WORK-LIFE BALANCE SATISFACTION FORMATION:
A QUANTITATIVE AND QUALITATIVE INVESTIGATION OF HOW WORKERS
CONTRIBUTE TO THEIR OWN WORK-LIFE BALANCE SATISFACTION FORMATION
WITHIN THE CONTEXT OF WORKGROUPS
By
Megan Huth

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Psychology - Doctor of Philosophy
2013

1

ABSTRACT
WORK-LIFE BALANCE SATISFACTION FORMATION:
A QUANTITATIVE AND QUALITATIVE INVESTIGATION OF HOW WORKERS
CONTRIBUTE TO THEIR OWN WORK-LIFE BALANCE SATISFACTION FORMATION
WITHIN THE CONTEXT OF WORKGROUPS
By
Megan Huth
Work-life balance (WLB) is defined as “satisfaction and good functioning at work and at
home, with a minimum of role conflict” (Clark, 2000, page 751) and is a topic that has garnered
increasing attention in the research and popular press over the past 40 years. This dissertation
seeks to more fully understand how individuals come to experience work-life balance and what
factors help contribute to feelings of satisfaction with balance. These factors include contextual
variables such as beliefs about team and manager support and team flexibility, individual
behavior variables such as behavioral detachment, flexibility use and work hours, individual
psychological variables such as psychological detachment, control or work, and autonomy, and
individual demographic variables such as family status and job level. Outcomes related to
feelings of work-life balance are also investigated, including performance, retention intentions
and emotional exhaustion.
A multi-level, multi-method approach was utilized to explore these relationships. All
three studies were conducted within the same organization that granted access for the purpose of
this dissertation. Study 1 utilized archival data that was gathered with intact teams so that team
level relationships could be investigated. Study 2 was a qualitative interview study that sought to
more fully capture nuances in individual experiences or work life balance and more deeply delve
into reasons why individuals were having difficulty maintaining a satisfactory balance. Study 3
2

was a quantitative survey study that used findings form Study 2 to increase the number of
constructs investigated from Study 1 and attempt to more accurately assess the complexity of
what contributes to work-life balance formation.
Among the many findings of these three studies, one of the strongest findings had to do
with psychological detachment, work life balance and emotional exhaustion. These findings
indicate that those who are better able to psychologically detach from work (even at higher levels
of workload) are more satisfied with work-life balance and experience lower levels of emotional
exhaustion as compared to their less detached peers. These results along with many others are
discussed along with future areas for research and practical implications.

3

Copyright by
MEGAN HUTH
2013

4

ACKNOWLEDGEMENTS

Pursuing a PhD takes the support of an incredibly large amount of people who are willing
to put up with challenging schedules and workloads, preoccupation with work and long days and
nights devoted to academic studies. I am indebted to so many wonderful people who have truly
gifted me with their time, patience, guidance and knowledge throughout this process.
First, I would like to thank my committee, Ann Marie Ryan, Steve Kozlowski, John
Schaubroek and Russ Johnson, who have given the gift of time and experience to help make this
dissertation what it is. Their expertise has increased the quality of this final project tenfold, and I
appreciate the time and energy they all invest in graduate students every day in developing us to
become the best psychologists possible. I would like to thank in particular my dissertation chair
and advisor, Ann Marie Ryan. Not only has Ann Marie been a tireless reviewer and thought
partner on the creation of this dissertation, she has also been my mentor in the broader sense of
the word. From my first day of graduate school, Ann Marie has helped guide and shaped my
thinking and my development as an organizational psychologist. All my current and future
success as an organizational psychologist owes a huge debt to her guidance and there are now
words of thanks large enough to express this.
Second, I’d like to thank my fellow graduate students who created a community of
inquiry that encouraged deep thought and spirited debate. In particular, I want to thank Jennifer
Wessel, James Grand, Katherine Corker and Charlotte Powers for making graduate school such a
wonderful and inviting place to be.
Third, I would also like to thank Charlotte Powers, Mary Keegin and Alexander Webb
for the many hours of interview coding they contributed toward this dissertation. The insights of
v

Charlotte Powers in particular were invaluable as they helped shape the direction of Study 2.
Thanks so much!
Fourth, I would also very much like to thank my partners at Genericorp who’s
partnership made this dissertation possible. The support and encouragement I received from so
many of the people I worked with there made me feel that doing field research was a breeze. I
thank them for the trust they put in me to research a potentially sensitive topic. In the interests of
confidentiality , I can’t name them by name, but they know who they are.
Finally I’d like to thank my family who bore the brunt of my preoccupation with my
dissertation as I juggled the writing and working part-time and then full time. For my mom and
dad, who suffered through terse phone calls when I needed to get back to editing. For my brother
and his family, with whom vacations sometimes had to be cut short. Your understanding means
the world to me.
And especially for my husband Aaron, who was an unending source of both emotional
and instrumental support. He kept our house running smoothly so I could immerse myself in
work when I needed to. Without him, I would not have fed myself (or the cats), and the dishes
and laundry would never have been done. In addition to practical help, he provided endless hours
of emotional support during the process and helped me think through more than one-- more than
ten!--thorny conceptual issues. Thank you for keeping me sane, Aaron. This would not have
been possible without you.

vi

TABLE OF CONTENTS

LIST OF TABLES

xi

LIST OF FIGURES

xix

CHAPTER 1: INTRODUCTION AND STUDY DESIGN
Introduction
A Note about Theory
Contributions
Work-life Research: An Overview
Conservation of Resources Theory
Role theory and COR Theory Hypotheses
Interrelationships of Antecedents
Relationships with Outcomes
Retention
Emotional Exhaustion
Refocusing on the System: Variables that Can Affect the Relationships of WLB
with Antecedents and Outcomes
Social Support
Border Theory
Flexibility
Idiosyncratic Deals
Norms for Work-Life Balance
Exploratory Research Questions
Methods Overview
Multi Method Approach
Research Site

1
1
3
4
7
14
15
16
22
23
27

CHAPTER 2: STUDY 1
Methods Study 1
Procedure
Location
Function
Manager Status and Direct Reports
Gender and Age
Sample Description
Method and Measures
Workload
Performance
Work-life Balance Satisfaction (Individual)
Work-life Balance Satisfaction (Manager)
Retention
Results Study 1

63
63
64
64
64
64
65
65
66
66
66
67
67
68
68

vii

32
33
40
42
49
52
57
59
59
60

Descriptive Statistics
Functional Contrasts
Correlations
Control Variables
Centering Variables
Hypothesis Testing
A Note on Statistical Models
Discussion Study 1
Summary of Results
Strengths and Limitations of Study 1

68
68
68
68
70
71
71
78
78
81

CHAPTER 3: STUDY 2
Study 2: A Qualitative Investigation of Work-Family Decision Making
Procedure
Interview Format and Focus
Participant Recruitment
Method
Results Study 2
Satisfaction
Detachment
Manager
Workload
Flexibility
Agency
Team/ Culture
Performance
Career
Anxiety/ Stress
Global/Commute Issues
Family
Top leaders
Managing
Rewards
Sustainability/Attrition
Discussion Study 2
Strengths and Weaknesses of Study 2

83
83
84
84
86
87
89
90
91
93
95
98
99
101
103
105
107
108
109
110
111
113
113
115
117

CHAPTER 4: STUDY 3
Study 3: Quantitative Survey of Individual Preferences and Perceptions of WLB
support
Hypothesis Development from Qualitative Results: Boundary Management
Beliefs About Boundary Flexibility
Work Control Factors
Psychological and Behavioral Detachment
Procedure

119
119

viii

119
120
121
122
123

North American sample
Non North American sample
Measures
Subjective Workload
Quantitative Workload
Work-life Interference
Negotiated Flexibility (I-deal flexibility)
Satisfaction with Flexibility
Used Flexibility
Preferences for Segmentation
Actual Integration
Emotional Exhaustion
Retention
Performance
Manager Emotional Support for WLB
Team Emotional Support for WLB
Perceived Manager Norm for WLB
Perceived Team Norms for WLB
Team Task Interdependence
Measures Added
Life Flexibility Ability
Life Flexibility Willingness
Work Flexibility Ability
Work Flexibility Willingness
Psychological Detachment
Control of Work
Autonomy
New Scales Developed Based on Study 2
Evening Work Style
Weekend Work Style
Vacation Work Style
Demographic Variables
Job Variables
Personal Variables
Household Demands
Child/eldercare Demands
Results Study 3
Descriptive Statistics
Functional contrasts
Regional Comparisons
Correlations
Control Variables
Hypothesis Testing
A Note on Statistical Models
Role Theory and Conservation of Resources Theory Hypotheses
ix

123
125
125
125
126
126
126
126
127
127
127
127
127
128
128
128
128
128
128
129
129
129
129
129
130
130
130
130
131
131
132
132
132
133
134
134
135
135
135
136
137
139
140
140
142

Coping Support Theory Hypotheses
Border and Idiosyncratic Deal theory hypotheses
A Note on Used Flexibility 148
Group Normative Theory Hypotheses
Discussion Study 3
Strengths and Limitations of Study 3

147
148
148
161
163
167

CHAPTER 5: DISCUSSION
General Discussion
Flexibility
Flexibility Satisfaction vs Used Flexibility
Perceptions of Flexibility Access
Border Theory and Boundary Management
Control and Autonomy
Social Support
Managers
Team
Family
Culture, Team Norms and Expectations
Emotional Exhaustion and Retention: Why Organizations Should Care
Emotional Exhaustion
Retention
Weaknesses
Future Research Directions
Practical Implications

171
171
171
172
173
176
180
183
184
185
187
188
191
192
193
194
195
198

APPENDICES
Appendix A: Selected Scales Used in Study 1
Appendix B: Structured Interview Script
Appendix C: Survey Items or Study 3
Appendix D: Tables
Appendix E: Figures

201
202
203
206
212
296

REFERENCES

330

x

LIST OF TABLES

Table 1 Summary of paper hypotheses, including whether they were supported
quantitatively by study 1 or 3

212

Table 2: Team Summary for Study 1

219

Table 3: Summary of Sample Characteristics
Table 4: Bonferroni Comparison for Work Life Balance Scores (n=810)

220

Table 5: Study 1 Correlation Matrix

220

Table 6: Means and Standard Deviations of Study 1 Variables

221

Table 7: Study 1 Unconditional Means Models for Work Life Balance and
Retention (n=809)

222

Table 8: Hypothesis 1: HLM models for workload predicting WLB (n=809)

223

Table 9: Mean comparison of performance scores for those who list workload as 224
a barrier to performance and those who do not (n=810)
Table 10: Hypothesis 3: HLM models for performance predicting WLB (n=740) 225
Table 11: Mean comparison of retention scores for those who list workload as a 226
barrier to performance and those who do not (n=809)
Table 12: Hypothesis 6: HLM models for WLB and performance predicting
227
retention Model 1= Null, Model 2 = WLB Fixed, Model 3 = Performance Fixed,
Model 4 = Interaction (n=809)
Table 13: Hypothesis 25: HLM models for workload and Manager WLB
228
predicting WLB, with Manager WLB moderating individual workload (n=688)
Table 14: Qualitative Coding Themes

229

Table 15: Satisfaction Categories

229

Table 16: Detachment Categories

229

Table 17: Manager Categories

230

Table 18: Workload Categories

230
xi

Table 19: Flexibility Categories

230

Table 20: Agency Categories

231

Table 21: Team/ Culture Categories

231

Table 22: Performance Categories

231

Table 23: Career Categories

232

Table 24: Anxiety/Stress Categories

232

Table 25: Commute/Global Categories

232

Table 26: Family Categories

232

Table 27: Top Leaders

232

Table 28: Managing Categories

233

Table 29: Rewards Categories

233

Table 30: Sustainability/Attrition Categories

233

Table 31: Study 3 response rates

234

Table 32: Study 3 Sample Characteristics

235

Table 33: Bonferroni Comparison for Retention Scores (n=1165)

235

Table 34: Correlation table

236

Table 35: Confirmatory Factor Analysis for Work/Life Flexibility Ability and
Willingness Scale (Using Direct Oblimin Rotation) (n=1140)

242

Table 36: Study 3 Unconditional Means Models for Work Life Interference,
Retention and Emotional Exhaustion (n=878)

244

Table 37: Hypothesis 1 and 17 Hierarchical linear regression predicting work 245
interference with life from used flexibility and the interaction of used flexibility
and work hours (n=991)
Table 38: Hypothesis 1 and 17 Hierarchical linear regression predicting work
246
interference with life from the interaction of negotiated flexibility and subjective
workload (n=991)
xii

Table 39: Hypothesis 2: Hierarchical linear regression of subjective workload
predicting performance (n=991)

246

Table 40: Hypothesis 2: Hierarchical linear regression of work hours predicting 247
performance (n=1000)
Table 41: Hypothesis 3 Hierarchical linear regression predicting work
interference with life from performance (n=993)

247

Table 42: Hypothesis 4: Hierarchical linear regression of subjective workload
predicting retention (n=998)

248

Table 43: Hypothesis 4: Hierarchical linear regression of work hours predicting 248
retention (n=995)
Table 44: Hypotheses 5, 6 and 7: Hierarchical linear regression predicting
retention from performance, work interference with life and the interaction of
work interference with life and performance (n=969)

249

Table 45: Hypothesis 8: Hierarchical linear regression of subjective workload
predicting emotional exhaustion (n=987)

249

Table 46: Hypothesis 8: Hierarchical linear regression of work hours predicting 250
emotional exhaustion (n=985)
Table 47: Hypothesis 9 and 10: Hierarchical linear regression of work life
interference predicting emotional exhaustion and partial mediation of work life
balance on the relationship of subjective workload predicting emotional
exhaustion (n=986)

250

Table 48: Hypothesis 10: Hypothesis 9 and 10: Hierarchical linear regression of 251
work life interference predicting emotional exhaustion and partial mediation of
work life balance on the relationship of work hours predicting emotional
exhaustion (n=984)
Table 49: Hypothesis 11: Hierarchical linear regression of the interaction of
252
Work interference with life and Manager support predicting retention (n=1033)
Table 50: Hypothesis 12: Hierarchical linear regression of the interaction of
Work interference with life and team emotional support predicting retention
(n=923)

253

Table 51: Hypothesis 13: Hierarchical linear regression of the interaction of
Work interference with life and team emotional support predicting emotional
exhaustion (n=1033)

254

xiii

Table 52: Hypothesis 14: Hierarchical linear regression of the interaction of
Work interference with life and team emotional support predicting emotional
exhaustion (n=924)

255

Table 53: Hypotheses 15 and 17 Hierarchical linear regression predicting work 256
interference with life from used flexibility and the interaction of used flexibility
and subjective workload (n=979)
Table 54: Hypotheses 15 and 17 Hierarchical linear regression predicting work 256
interference with life from negotiated flexibility and the interaction of negotiated
flexibility and work hours (n=991)
Table 55: Hypotheses 16 and 18 Hierarchical linear regression predicting work
interference with life from flexibility satisfaction and from the interaction of
flexibility satisfaction and subjective workload (n=991)

257

Table 56: Hypothesis 17 Hierarchical linear regression predicting work
interference with life from the interaction of used flexibility and subjective
workload (n=991)

258

Table 57: Hypothesis 17 Hierarchical linear regression predicting work
interference with life from the interaction of used flexibility and work hours
(n=991)

259

Table 58: Hypothesis 18 Hierarchical linear regression predicting work
interference with life from the interaction of flexibility satisfaction and work
hours (n=988)

260

Table 59: Hypothesis 19: Hierarchical linear regression of the predicting
retention from used flexibility (n=987)

260

Table 60: Hypothesis 19: Hierarchical linear regression of the predicting
retention from negotiated flexibility (n=975)

261

Table 61: Hypothesis 20

261

Table 62: Hypothesis 21: Hierarchical linear regression of the predicting
emotional exhaustion from used flexibility (n=988)

262

Table 63: Hypothesis 21: Hierarchical linear regression of the predicting
emotional exhaustion from negotiated flexibility (n=976)

262

Table 64: Hypothesis 22: Hierarchical linear regression of the predicting
emotional exhaustion from flexibility satisfaction (n=987)

263

xiv

Table 65: Hypothesis 23 Hierarchical linear regression predicting work
interference with life from the interaction of used flexibility and segmentation
preferences (n=988)

263

Table 66: Hypothesis 23 Hierarchical linear regression predicting work
interference with life from the interaction of negotiated flexibility and
segmentation preferences (n=976)

264

Table 67: Hypothesis 241 Hierarchical linear regression predicting work
interference with life from the interaction of subjective workload and actual
integration (n=987)

265

Table 68: Hypothesis 241 Hierarchical linear regression predicting work
interference with life from the interaction of work hours and actual integration
(n=985)

266

Table 69: Hypothesis 242 Hierarchical linear regression predicting work
interference with life from the interaction of actual integration and preferences
for segmentation(n=1004)

266

Table 70: Hypothesis 243 and 244 Hierarchical linear regression predicting
work interference with life from life flexibility ability and the interaction of
subjective workload and life flexibility ability (n=991)

267

Table 71: Hypothesis 243 and 244 Hierarchical linear regression predicting
268
work interference with life from life flexibility ability and the interaction of work
hours and life flexibility ability (n=989)
Table 72: Hypothesis 245 and 246 Hierarchical linear regression predicting
269
work interference with life from life flexibility willingness and the interaction of
subjective workload and life flexibility willingness (n=991)
Table 73: Hypothesis 245 and 246 Hierarchical linear regression predicting
270
work interference with life from life flexibility willingness and the interaction of
work hours and life flexibility willingness (n=989)
Table 74: Hypothesis 247 and 248 Hierarchical linear regression predicting
work interference with life from work flexibility ability and the interaction of
subjective workload and work flexibility ability (n=991)

271

Table 75: Hypothesis 246 and 247 Hierarchical linear regression predicting
work interference with life from work flexibility ability and the interaction of
work hours and work flexibility ability (n=989)

272

xv

Table 76: Hypothesis 249 and 2410 Hierarchical linear regression predicting
273
work interference with life from work flexibility willingness and the interaction
of subjective workload and work flexibility willingness (n=991)
Table 77: Hypothesis 249 and 2410 Hierarchical linear regression predicting
274
work interference with life from work flexibility willingness and the interaction
of work hours and work flexibility willingness (n=989)
Table 78: Hypothesis 2411 and 2412 Hierarchical linear regression predicting 275
work interference with life from control of work and the interaction of subjective
workload and control of work (n=990)
Table 79: Hypothesis 2411 and 2412 Hierarchical linear regression predicting
work interference with life from control of work and the interaction of work
hours and control of work (n=988)

276

Table 80: Hypothesis 2413 and 2414 Hierarchical linear regression predicting
work interference with life from autonomy and the interaction of subjective
workload and autonomy (n=991)

277

Table 81: Hypothesis 2413 and 2414 Hierarchical linear regression predicting 278
work interference with life from autonomy and the interaction of work hours and
autonomy (n=989)
Table 82: Hypothesis 2415 and 2416 Hierarchical linear regression predicting 279
work interference with life from psychological detachment and the interaction of
subjective workload and psych detachment (n=991)
Table 83: Hypothesis 2415 and 2416 Hierarchical linear regression predicting 280
work interference with life from psychological detachment and the interaction of
work hours and psych detachment (n=989)
Table 84: Hypothesis 2417 and 2418 Hierarchical linear regression predicting
work interference with life from evening work style and the interaction of
subjective workload and evening work style (n=991)

281

Table 85: Hypothesis 2417 and 2418 Hierarchical linear regression predicting 282
work interference with life from evening work style and the interaction of work
hours and evening work style (n=989)
Table 86: Hypothesis 2419 and 2420 Hierarchical linear regression predicting 283
work interference with life from weekend work style and the interaction of
subjective workload and weekend work style (n=991)
Table 87: Hypothesis 2419 and 2420 Hierarchical linear regression predicting
xvi

284

work interference with life from weekend work style and the interaction of work
hours and weekend work style (n=989)
Table 88: Hypothesis 2421 and 2422 Hierarchical linear regression predicting
work interference with life from vacation work style and the interaction of
subjective workload and vacation work style (n=990)

285

Table 89: Hypothesis 2421 and 2422 Hierarchical linear regression predicting 286
work interference with life from vacation work style and the interaction of work
hours and vacation work style (n=988)
Table 90: Hypothesis 25 Hierarchical linear regression predicting work
287
interference with life from the interaction of manager work life interference and
subjective workload (n=285)
Table 91: Hypothesis 25 : Hierarchical linear regression predicting work
288
interference with life from the interaction of manager work life interference and
work hours (n=284)
Table 92: Hypothesis 26 Hierarchical linear regression predicting work
interference with life from the interaction of manager instrumental support and
subjective workload (n=979)

289

Table 93: Hypothesis 26 Hierarchical linear regression predicting work
290
interference with life from the interaction of manager instrumental support and
work hours (n=980)
Table 94: Study 1 Unconditional Means Models to test for Aggregation for Team 291
Emotional Support, Team Instrumental Support and Work Flexibility Ability
(n=507)
Table 95: Hypothesis 28 Hierarchical linear regression predicting work
292
interference with life from the interaction of aggregated team emotional support
and subjective workload (n=507)
Table 96: Hypothesis 28 Hierarchical linear regression predicting work
293
interference with life from the interaction of aggregated team emotional support
and work hours (n=507)
Table 97: Hypothesis 28 Hierarchical linear regression predicting work
interference with life from the interaction of aggregated work flexibility ability
and subjective workload (n=507)

294

Table 98: Hypothesis 28 Hierarchical linear regression predicting work
interference with life from the interaction of aggregated work flexibility ability

295

xvii

and work hours (n=507)
Table 99 Supplemental Analyses Hypotheses 15-18: Hierarchical linear
regression of interaction of flexibly and workload predicting work life
interference (n=812)

323

Table 100 Supplemental Analyses: Hypotheses 241-2410: Hierarchical linear
324
regression of interaction of flexibility perceptions and workload predicting work
life interference (n=812)
Table 101 Supplemental Analyses Hypotheses 2411-14: Hierarchical linear
326
regression of interaction of control and autonomy and workload predicting work
life interference (n=812)
Table 102 Supplemental Analyses Hypotheses 102 2415-22: Hierarchical linear 327
regression of interaction of detachment and workload predicting work life
interference (n=812)
Table 103 Supplemental Analyses Hypotheses 25-28: Hierarchical linear
regression of interaction of team relevant variables and workload predicting
work life interference (n=256)

328

Table 104 Supplemental Analyses MANCOVA of Function and Region with
Bonferroni post-hoc tests in relation to Emotional Exhaustion, Work-Life
Interference and Retention

329

xviii

LIST OF FIGURES

Figure 1: Model of the Hypothesized Relationships and How They Fit Within Theoretical
Frameworks

298

Figure 2 Model showing COR Hypotheses

299

Figure 3 Model Showing Moderation Hypotheses of WLB on Outcomes

300

Figure 4 Model of Border Management Hypotheses and Moderations

301

Figure 5 Model and Manager and Team Hypotheses

302

Figure 6: Graph of Proposed Moderation for Hypothesis 7

303

Figure 7: Graph of Proposed Moderation for Hypothesis 11

304

Figure 8 Graph of Proposed moderation for Hypothesis

305

Figure 9: Graph of Proposed Moderation for Hypothesis 13

306

Figure 10: Graph of Proposed Moderation for Hypothesis 14

307

Figure 11: Graph of Proposed Moderation for Hypothesis 17

308

Figure 12: Graph of Proposed Moderation for Hypothesis 18

309

Figure 13: Graph of Proposed Moderation for Hypothesis 23

310

Figure 14: Graph of Proposed Moderation for Hypothesis 24

311

Figure 15: Graph of Proposed Moderation for Hypothesis 25

312

Figure 16: Graph of Proposed Moderation for Hypothesis 26

313

Figure 17: Graph of Proposed Moderation for Hypothesis 27

314

Figure 18: Graph of Proposed Moderation for Hypothesis 28

315

Figure 19: Graph of Frequency of Hours Spent on Housework

316

Figure 20: Graph of Frequency of Hours Spent on Child and/or Elder Care

317

Figure 21: Interaction of Actual Integration and Subjective Workload Predicting Work

318

xix

Interference with Life
Figure 22: Interaction of Control of Work and Subjective Workload Predicting Work
Interference with Life

319

Figure 23: Interaction of Autonomy and Subjective Workload Predicting Work
Interference with Life

320

Figure 24: Interaction of Psychological Detachment and Subjective Workload Predicting
Work Interference with Life

321

Figure 25: Interaction of Evening Work Style and Subjective Workload Predicting Work
Interference with Life

322

Figure 26: Interaction of Weekend Work Style and Subjective Workload Predicting Work
Interference with Life

323

Figure 27: Interaction of Vacation Work Style and Subjective Workload Predicting Work
Interference with Life

324

xx

CHAPTER 1: INTRODUCTION AND STUDY DESIGN

Introduction
In the past 30 years, research into how people balance their work roles with other life
roles has emerged as an area of industrial-organizational psychology research that has relevance
to academics and practitioners alike. Both groups see the topic as rife with potential to help ease
some of the strains that can affect people if life roles conflict. The benefits of well-balanced
employees seem utopian and far-reaching, with researchers showing links between work-life
balance and workplace commitment and job satisfaction (O’Neill, Harrison, Cleveland,
Almeida, Stawski, & Crouter, 2009) , improved mental and physical health (Frone, Russell &
Cooper, 1992; Frone, Russell & Barnes, 1996), improved marital and family functioning
(Ransford, Crouter, & McHale, 2008) and an overall improved sense of well-being (Lapierre,
Spector, Allen, Poelmans, Cooper, O’Driscoll, et al, 2008).
With all these benefits of balance, it would seem that employers, workers, researchers
and practitioners alike would be especially motivated to make work-life balance a priority and
that after 30 years of research all American employees should be well balanced. This, however,
is certainly not the case. Solving the “work-life balance problem” has not been simple. One way
organizations have attempted to address work-life balance issues for their employees has been
through the implementation of structural work-life policies at the organizational level. These
often include access to flexible work schedules, telecommuting options, nontraditional work
arrangements, onsite child care etc. The mere provision of these types of supports, however,
may not prove sufficient, as research shows that organizations that offer these programs may not
1

have significantly more balanced employees than those that do not offer these programs (Kossek
& Ozeki, 1998).
Kossek, Lewis and Hammer (2009) propose that structural supports are only half the
solution to the work-life balance problem. In addition to structural supports, the organization
must also provide cultural work-life support. By this, the authors mean that employees have a
sense from managers and coworkers that their obligations and responsibilities to both work and
non-work are valued and respected. Past research has shown that this type of cultural support is
critical in determining whether or not employees actually make use of the available structural
supports (Allen, 2001). The context of the employee, which includes the manager and the team,
can effect whether or not the employee feels the structural supports are truly available or
accessible to them personally (Hammer, Kossek, Yragui, Bodner & Hansen, 2009). This can
help drive the decisions the individual person makes about how to balance work and non work
responsibilities.
The goal of this paper is to better understand how and why people make work-life
balance decisions and how these decisions are impacted by team and manager support with the
ultimate goal of finding potential solutions to the work-life balance problem. By considering
both social support and structural support variables at the same time, it is possible to follow the
advice of Kossek, Lewis and Hammer and examine the dual role these two types of
organizational supports may play in affecting employee work-life balance. To best do this and to
test direct relationships of support to these variables, this paper also examines organizationally
relevant antecedents to work-life balance, namely productivity and workload, and outcomes of
work-life balance, retention intentions and emotional exhaustion. Through the three studies that
make up this paper, it will be possible to gain a better understanding of the forces that can
2

constrain individual decision making about work and life roles within the work context that can
lead to dissatisfaction with balance and the negative consequences thereof. This understanding
may lead to recommendations for more or less effective strategies individuals can use when
attempting to achieve a better balance in their own work and non-work lives.

A Note about Theory
One of the unique aspects of work-life research is that while much of organizational
psychology is defined specifically by role in which it takes place (the “work” role), work-life
research specifically focuses on the interaction of roles and is inclusive of roles outside of work.
When considering the conceptual space of examining how work and life interact, it is clear to see
that this encompasses a huge amount of human experience. This has led to some challenges in
coming to consensus on a single unifying theory, with major researchers arguing that with such a
large construct space to cover, it is unproductive and limiting to expect a single theory to
encompass the entire realm of work-life interaction and balance (Matthews, Allen, Barling, Eby,
Greenhaus, Kossek, & Poelmans, 2011). With this in mind, this paper seeks to explore work-life
balance through the lens of compatible but distinct theories in an effort to better understand how
individuals form and experience their own satisfaction with work-life balance.
Most work-life balance (WLB) research relies on role theory and demands and resource
theories to explain the particular focus on work and life roles, what role balance means, and why
negative outcomes can occur from imbalance (Kinnunen & Mauno, 2008). When attempting to
mitigate the negative effects of work-life imbalance, researchers and practitioners can choose to
either focus on factors that can affect outcomes after imbalance occurs or those that can mitigate
the effect of antecedents on WLB before imbalance occurs. Different theories are best suited to
3

understanding these different factors. The social support literature is best suited to understanding
how social supports can help individuals effectively deal with work-life imbalance in ways that
mitigate the effect of imbalance on outcomes. Border theory and decision making theory are
best suited for understanding the process that contributes to feelings of imbalance, such as how
individuals interact with their roles to manage a satisfactory balance within the context of work
and non-work demands.
This paper will attempt to take a broad view of the work-life research space and examine
factors that occur at different points in the formation of work-life balance perceptions and
experiences. First, the relationships between antecedents and outcomes of WLB will be
investigated using role theory (Kahn, Wolfe, Quinn, Snoek, & Rosenthal, 1964) and
Conservation of Resources Theory (COR; Hobfoll, 1989). Next, the role of manager and team
support will be examined to see how these support sources may help buffer negative outcomes
related to dissatisfaction with balance. Finally, this paper will take a closer look at how
experiences of WLB are formed by looking at the process individuals engage in when attempting
to balance competing work and non work roles. This section will make use of border theory,
flexibility, idiosyncratic deal theory, and group theory to help explain specific mechanisms that
may occur within this process. Please see Figure 1 for a heuristic of these relationships, as well
as the specific constructs that will be investigated.
Contributions
This paper contributes to the field of organizational psychology and work-life research in
a number of ways. First, this paper will add to the role theory/COR theory understanding of
WLB through a focus on the organizationally relevant variables of retention intentions and
performance that are often invoked in work-life research but are rarely measured (Kossek &
4

Ozeki, 1998). Although maintaining high performance is a clear concern for organizations, very
few researchers have directly studied the relationship of performance and balance, or tied
performance to retention through the lens of WLB (Kossek & Ozeki, 1998). This study will help
better clarify the relationship between performance and WLB that may eventually help
practitioners act in ways that can maximize productivity while minimizing harmful effects to
WLB to both the organization and the individual.
The second main contribution augments the support literature by looking at a potential
source of support that is rarely studied in WLB research, that of the team or workgroup.
Although manger support for WLB has been demonstrated to be an important factor in linking
WLB satisfaction and outcomes like retention and increased well-being (Lapierre, Spector,
Allen, Poelmans, Cooper, O’Driscoll, et al, 2008; Thomas & Ganster, 1995), the role of
coworkers or team members as support has largely been unexplored in the WLB literature
(Hunter, Perry, Carlson, & Smith, 2010).
Beyond emotional support in specific, this study also contributes to the literature by more
closely examining the ways in which teams and managers may set informal norms about work
and life balance. This is an area which WLB research has largely ignored, but may help explain
some of the pressures individuals experience when making decisions about balancing work and
family life.
One of the most important contributions this paper makes is through the investigation of
how individual workers participate in the formation of their own satisfaction with balance.
Using border theory to focus the investigation and idiosyncratic deal theory to help explain
individual decisions and actions, this paper seeks to find ways individuals can proactively
manage both work and non-work. Although the role of employee decision making has long been
5

a part of the construct of WLB (Greenhaus & Beutell, 1985), few researchers to this point have
focused on how these decisions are made as a process and what individual and contextual
differences might drive them. Powell and Greenhaus (2006) suggest that to truly understand
WLB, investigators need to begin focusing more on this type of process, and this study is one of
the first to explicitly do so.
The final contribution has to do with the antecedent and outcome based perspective this
paper takes on examining the work/life interface. Although this paper focuses exclusively on the
situations and experiences that occur at work that can affect the work-life interface (rather than
looking at family factors in depth), it does attempt to gain a broader understanding of these work
related factors at various stages of the work-life balance satisfaction formation process. Through
the use of both quantitative and qualitative data this paper will examine unexplored relationships
of organizationally and personally relevant variables. This paper attempt to get at the more
mutable elements of balance through first person accounts of the decisions and actions that
contribute to balancing work and life, and focuses on the elements that can happen before or
after conflict between roles occurs. This represents a step forward in considering WLB as
something that is constructed rather than solely relying on traditional survey research which
treats WLB as more of a stable, already constructed characteristic.
Before jumping in to specific hypotheses and relationships, it is necessary to set up the
research questions within the context of work-life research. This will be covered in the next
section. Following this review, the hypotheses will be introduced within the context of theories
that help support them. Role theory (Kahn et al, 1964) and Conservation of Resources theory
(Hobfoll, 1989) will be used to help explain the relationships between work antecedents and
work outcomes of imbalance. Next, the support literature will be used to help understand the
6

role of manager and team social support in buffering the negative effects of balance. Then,
border theory will be introduced to help explain how individuals might negotiate work and life
roles differently based on preferences about role interaction and permeability. Finally,
idiosyncratic deal theory and group normative theory will be used to help understand some of the
factors that might constrain decision making around these border negotiations and why
individuals may make decisions that do not always align with their preferences. Following these
sections, proposed methods, measures and analyses will be introduced.

Work-life Research: An Overview
The history of work-life research is one of testing evolving assumptions that mainly have
centered around the demography of those who are studied. When first conceptualized, work-life
research was focused specifically on work-family conflict and was seen as a problem mainly for
working mothers, a modern problem stemming from a change in traditional family structure and
the increasing participation in the workforce by women (Marecek & Ballou, 1981). Early
research under this paradigm often focused almost exclusively on the experience of working
women with families, which made it impossible to compare the experiences of men and women
directly and those with and without children (Barnett & Hyde, 2001). Role theory, with a focus
on gender roles in particular, was used to explain how work and life interacted.
A theory based entirely on gender roles did not explain real world experiences entirely,
however, as researchers and practitioners began to notice that work-family conflict was not a
strictly gendered phenomenon. As research accumulated it became apparent that although
gender differences may exist in some samples, these differences were often smaller than
anticipated by strict gender role theories and explained only a small part of the variance in
7

reports of work-family conflict (Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005; Ford,
Heinen, & Langkamer, 2007). This represented a shift away in the literature from a strictly
gendered lens of role theory, and allowed for the experiences of both genders to be addressed.
The next step in the evolution of work-life research dealt with the ways roles can conflict
within employees. Greenhaus and Beutell (1985) wrote a seminal article articulating the ways in
which the work role can interfere with the family role. The authors discuss how conflict can
occur in either the work-to-family direction or the family-to-work direction. They are careful to
point out, however, the specific direction the conflict occurs in is determined by the choices of
the individual. For example, an individual may experience the situation where they must decide
between attending a child’s soccer game and attending a work meeting. If the person decides to
go to the soccer game, then this becomes family-to-work conflict. If, however, they choose to
go to the work meeting, this is work-to-family conflict. Ultimately, how an individual
experiences work-life conflict is the product of various decisions he/she makes about how to
balance these roles. This theory gained broad support in the literature, along with other similar
conceptualizations that specifically draw on role theory (Allen, Herst, Bruck & Sutton, 2000;
Schaufeli, Bakker, Van der Heijden, & Prins, 2009; Um & Harrison, 1998).
A further shift in work-life research occurred when practitioners and researchers began to
be concerned that focusing on conflict did not explain the whole story about work and life
interaction. With a growing interest in positive psychology, the sharp focus on conflict did not
allow researchers or practitioners to investigate ways work and life could enrich each other
(Greenhaus & Powell, 2006). Thus, the sub-focus of work-family enrichment was born. This
expansion of work-life research was based in role theory as well, but with a focus on individual
identity development and the psychological needs met by both work and family roles (Wayne,
8

Randel & Stevens, 2006). Work-family enrichment has allowed a broader understanding of the
work-life interface and has been demonstrated to have complementary but distinct effects on
worker well-being and organizational variables as work-family conflict (Witt & Carlson, 2006).
A more recent shift in research focus has centered on the centrality and necessity of
having family be the specific role with which work can conflict. In many previous studies,
samples were specifically limited to those with easily defined families, e.g. spouses and/or
children (Casper, Eby, Bordeaux, Lockwood & Lambert, 2004). Some researchers felt that this
categorization was too narrow and that excluding a large subpopulation that had less traditional
family arrangements or who were single and childless missed valuable insight into the ways
work and life intersect (Beauregard, Ozbilgin & Bell, 2006). Due to this, some researchers in the
field have shifted away from the study of work-family role conflict in particular to a more
inclusive construct of work-life conflict. This shift still stays resolutely in role theory however,
with researchers simply expanding the specific roles they examine that may conflict with the
work role (Greenhaus & Singh, 2003).
In summary, over the years, work-life research has broadened from focusing exclusively
on the experiences of conflict of working mothers, to a more complex interaction of both positive
and negative experiences tied to individual roles and their importance for all workers. While this
expanded understanding of the ways work and life interact represents positive attempts to
increase generalizability and more effectively address real issues workers are facing, the broader
focus can create challenges of complexity, theory and focus as well as ambiguity about terms and
definitions.
When researchers begin designing studies to better understand how work and other life
roles intersect, they must make a series of decisions that are framed within these shifts of the
9

work-life literature. Some choose to focus on just conflict or enrichment, while others focus on
both (McMillan, Morris, & Atchley, 2011). Some choose to retain the family focus of the
research, while others alter measures and instructions in order to be more inclusive of a variety
of life domains (Casper et al, 2004).
As Gregory and Milner (2009) point out in an editorial on the state of work-life research,
the choice of terminology is not arbitrary as the labels we choose to attach to things can carry
weight in and of themselves. The terms conflict and enrichment both have clear focus on
valence, while the decision to focus on family roles vs. all non-work roles carries some
assumptions and implicit values about life stages and choices. Even the most common term
balance implies that when one role receives more attention than the other that the scales become
imbalanced and the less favored role is devalued. Gregory and Milner (2009) suggest that the
term “work-life harmony” may be more appropriate to capture the idea that the ideal
arrangement of work-life role is one in which conflict is minimized and enrichment maximized.
However this term change has been slow to catch on and the term “work-life balance” has won
the day in the current literature. Therefore the term work-life balance (WLB) will be used in the
methods and hypotheses to describe the work-life construct, which can be defined as
“satisfaction and good functioning at work and at home, with a minimum of role conflict”
(Clark, 2000, 751 page). It is important to note, however, that much of the past literature focuses
explicitly on conflict, so much of the supporting literature cited in this paper will use this term.
Although these shifts in the literature have affected the choices researchers make when
studying the ways work and the rest of life can interact, much of the current WLB literature is
focused on the antecedents and outcomes of WLB after instances of conflict or enrichment have
already occurred (Allen et al, 2000). Many of these studies are based in role theory and other
10

related theories, such as the Conservation of Resources Theory (Hobfoll, 1989), which help
explain why negative individual and organizational outcomes manifest when individuals are
dissatisfied with balance (Ford, Heinen & Langkamer, 2007). While these studies are valuable
in that they help identify especially important work and family characteristics that can contribute
to feeling of imbalance, and thus suggest potential intervention points, many of these studies
ignore the decision making aspect of WLB (Powell & Greenhaus, 2006). WLB is often treated
as a relatively static characteristic that is assessed in such a way that the decision making or
formative aspect of it is rarely addressed and is treated as something that remains relatively
stable. Despite the fact that individual decision making is an explicit part of Greenhaus and
Beutell’s (1985) conceptualization of work-life conflict, relatively little research explicitly
investigates the process individuals use to actually make these decisions or how workers may
attempt to arrange their work and non-work lives in order to minimize instances of imbalance
proactively.
Recent trends in research suggest that becoming mired down in this traditional focus may
be limiting the ways researchers are thinking about the construct of work-life balance. Some
recent researchers have looked at WLB over time via event sampling methodology and have
found that WLB does vary from day to day (Judge, Ilies & Scott, 2006; Livingston & Judge,
2008). This research points to the idea that satisfaction with WLB is a mutable construct that is
driven by a variety of organizational contexts (e.g. overwork; Butler, Gryzwacz, Bass & Linney,
2005) and individual preferences and actions (e.g. optimization behaviors, Baltes & HeydensGahir, 2003). Although these types of studies suggest that there is an individual decision making
process workers are involved in while managing work and non-work roles, few researchers have

11

looked directly at these processes and how people actually make the necessary decisions (Powell
& Greenhaus, 2006).
These decisions, however, do not occur in a vacuum. There are a variety of personal and
situational presses that can push work-life balance decisions in one direction or another, many
that occur within the employee’s most proximal environment, the team. This paper attempts to
understand WLB as a complex individual system that functions within this workgroup
environment. Through taking a broader view of the work-life balance system and incorporating
a focus on both individual preferences and decisions and perceptions about situational
constraints, I hope to be able to more fully understand how satisfaction with work-life balance
forms within individuals. With this understanding comes the potential of offering truly helpful
advice to organizations and individuals on proactive ways to decrease experiences of work-life
imbalance.
The foundational hypotheses of this paper will be built around the theories most common
in the work-life literature, namely role theory and COR theory. The following sections will
outline these theories and provide a literature base for the foundational hypotheses outlined in
Figures 1 through 5.
Role theory. Work-life balance research has its roots in role theory. Role theory posits
that individuals inhabit a variety of life roles that are determined by biology, culture, and social
differentiation (Kahn et al, 1964). These roles allow individuals to enact the appropriate types of
behaviors at the appropriate times. Kahn et al (1964) were among the first to specifically focus
on work roles and family roles as being of particular importance, since modern, industrialized
adults tend to spend the most time and place the most importance on these two roles (Greenhaus
& Beutell, 1985).
12

Negative individual and organizational outcomes can be expected when these roles
become overburdened (“role overload”) or when the pressures from one role interfere with the
pressures from another role (“role conflict”). When roles conflict, individuals feel stress at being
unable to meet felt obligations and this can affect satisfaction with both or either home and work
domains (Greenhaus and Beutell, 1985). Originally, role conflict was conceptualized as a
unidimensional concept, but further research has shown that the directions of work-interferencewith-life and life-interference-with-work can be considered distinct (Frone, Russell & Cooper,
1992). This paper will focus on satisfaction with balance as a general concept as well as
focusing on work-interference-with family but will not specifically examine family-interferencewith-work in great depth. The purpose of this is due to the focus on the work role, and the
constraints and supports the work role can offer to help improve balance despite role demands
originating from the work role. Family demands may conflict with work demands, but are less
amenable to workplace interventions and are thus not a part of the focus of this paper.
A great deal of research has been conducted based on this basic framework, with findings
showing all types of conflict being related to negative outcomes such as greater mental and
physical health complaints (Frone, Russell & Cooper, 1992; Frone, Russell & Barnes, 1996;
Thomas & Ganster, 1995), decreased job and life satisfaction (Ford et al, 2007; Greenhaus,
Parasuraman & Collins, 2001), and increased turnover intentions and absenteeism (Rode, Rehg,
Near & Underhill, 2007). These findings indicate that stressful incompatibility of the work and
family roles can cause detriments to individuals and organizations. A common means of
understanding these negative relationships between work-life conflict and many organizational
and personal outcomes is through the Conservation of Resources of Theory, which will be
discussed in the following section.
13

Conservation of Resources Theory. In the Conservation of Resources (COR) theory,
Hobfoll (1989) outlines how individuals possess a limited amount of resources that become
depleted or renewed throughout the day. If these resources are overtaxed without being renewed,
then an individual experiences strain and negative consequences that stem from that strain. The
types of resources that are most relevant to the study of work-life balance are personal energies:
“those objects, personal characteristics, conditions or energies that are valued by the individual
or that serve as a means for attainment of these objects, personal characteristics, conditions or
energies” (p. 516). The concept of energies encompasses the internal ability to invest emotional
and cognitive effort into tasks and interpersonal relationships.
COR is related to work-life balance in that it helps explain how the experiences in
one role can spill over and affect the other role. Within the context role conflict theory, COR
theory suggests that roles can come in conflict when the role demands of one role detract from
successful enactment of the other role. Research in work family spillover supports this
supposition, showing that the depletion of energies at work can lead to possessing fewer
resources at home to deal with family or personal life role demands (Small & Riley, 1990).
COR also helps explain the consistent and robust link researchers have found between
work demands, family demands and work-life conflict. Having a high workload, defined as
having high amounts of work, having to work fast or under pressure (van der Doef & Maes,
1999), is often related to decreased job satisfaction and general well being as well as increased
work-to-family conflict (Byron, 2005; Ford et al, 2007). Having high family demands, such as
responsibility for child or elder care or high household responsibilities, is also related to
decreased life role satisfaction and greater family to work conflict (Hammer, Kossek, Anger
Bodner, & Zimmerman, 2011). Because high workloads and high family demands require
14

greater amounts of focused and sustained energy to address, an individual’s finite resources are
drained in one domain, leaving insufficient energy left to effectively deal with the other domain.
Role Theory and COR Theory Hypotheses. The use of role theory and COR theory are
quite common in the work-life research literature, and a portion of this paper seeks to replicate
some of the findings already found in this literature. This paper will go beyond simple
replication, however, to extend the understanding of work-life balance within the context of role
theory and COR theory by examining variables that are rarely explicitly studied in work-life
research. This section will outline both the replication and extension hypotheses for role and
COR theory.
Within the framework of the COR theory, individuals have a set amount of resources
they can devote to work or non-work roles. If an individual has a workload they see as overload,
this can use up a greater share of the individual’s resources than may be desired, and thus leave
the non-work role impoverished, resulting in dissatisfaction with balance. If an individual feels
that workload is an inherent job characteristic within a particular organization or team, they may
feel that leaving the position (e.g. turnover) is the best solution to adjust work demands to a level
they see as more conducive to allowing for better work-non work balance (Boyar, Maertz,
Pearson & Keough, 2003). Additionally, problems maintaining a satisfactory WLB can affect an
individual’s well-being, and feelings of imbalance can lead to emotional exhaustion (Hall,
Dollard, Tuckey, Winefield, & Tompson, 2010).
The following section will present the ten hypotheses most centrally tied to the role and
COR theories, and will examine the how workload and performance act as antecedents to worklife balance and the outcomes of retention intentions and emotional exhaustion. For a summary
of a complete set of the entire paper hypotheses, please see Table 1. Please see Figures 6
15

through 18 for graphs of the interaction hypotheses and Figure 2 for a model representing the
hypotheses presented in this section.
Interrelationships of Antecedents. One of the more robust findings in the WLB literature
is that of the relationship between workload and balance (Geurts Kompier, Roxburgh &
Houtman, 2003; Peeters, Montgomery, Bakker & Schaufeli, 2005). Time after time researchers
have found that individuals with higher workloads also report more work-life conflict,
specifically in the work-to-life direction (Ford, Heinen & Langkamer, 2007). This finding
directly dovetails with the COR theory, and suggests that the extra energy needed at work to
invest in managing a high workload saps the energy necessary for an individual to enact the
home role to the best of her/his ability. Some studies have found correlations between workload
and work-life conflict to be as high as .60 (Ilies, Schwind, Wagner, Johnson, DeRue & Ilgen,
2007).

Hypothesis 1: Workload will be negatively related to WLB such that with higher
workloads will report lower satisfaction with balance.

One of the barriers that can stand in the way of organizations implementing work-life
balance friendly initiatives is the concern that performance will suffer if accommodations are
made (Fleetwood, 2007). Some studies suggest that an individual will choose to put in long
work hours at the cost of work-life balance in order to appear like a hard worker and top
performer and to mitigate negative evaluations of their performance (Gambles et al, 2006). It
would seem to make sense that individuals who have high workloads and successfully
accomplish them would be seen as better workers, and thus be rated as higher performers, than
16

their peers who accomplish less with smaller workloads. However, research suggests that when
workload becomes overly burdensome, workers begin to experience burnout and as a result
performance suffers (Shirom, Nirel & Vinokur, 2009).
One way to conceptualize the fine balance between having a sufficiently challenging
workload to maximize performance while avoiding role overload is through the activation theory
of job design (Gardner, Dunham & Cummings, 1988; Scott, 1966). The activation theory of job
design fits within the COR theory in that it views work and workload as a potential source of
both resource drain and resource renewal. The theory supposes that people have different levels
of work activation that are optimal to their individual productivity and motivation. Challenging
and interesting work can be a renewing agent that keeps a person motivated and focused on the
job at hand. If someone has too little work, then he/she is not fully activated, can find work less
motivating and interesting, and in turn productivity can fall off (Gardner, Durham & Cummings,
1988). If someone has too much work, however, they may not have the time or other resources
to effectively deal with such a high workload and then productivity may fall off here as well
(Gardner, 1986). The peak of the inverted U relationship between workload and productivity
falls at the point where the individual is experiencing their ideal level of activation: challenging
and interesting work without being overloaded.
Some researchers suggest that there is an optimal level of workload that maximizes
performance while not being overly burdensome in terms of role overload and job demands.
Each individual has an optimal level of job demands based on individual preferences and job
complexity. In a study that looked at the relationship between job performance and job
demands, Janssen (2001) found a curvilinear relationship between job demands and performance

17

showing that performance peaked at an optimal level of demands but then began to fall once
people began to be overloaded.
When workload is large and demanding enough that the individual feels that it is a barrier
to them performing their best, however, this can be understood as role overload. If there is too
much work for an individual to be able to complete up to his/her standards, then his/her
performance could decrease. Thus a general negative relationship between the two may be
observed when workload is considered to be too high.

Hypothesis 2: Workload and performance will be negatively related such that those with
higher workloads will be rated lower on performance.

Just as workload and performance have an uneasy relationship, so do performance and
WLB. Several researchers hypothesize that it is logical that sustained periods of low satisfaction
with balance and high work-life conflict negatively affect performance due to the increased
stress an individual experiences (Kossek & Ozeki, 1999). According to Hobfoll’s COR theory
(1989), stress occurs when the individual does not possess enough resources to deal with the
demands of one role or the other. In terms of performance, this lack of resources would suggest
that people are functioning at a level that does not meet the required obligations of the work role,
and would thus be evaluated as performing more poorly than their optimal potential or than their
more balanced peers (Grandey & Cropanzano, 1999).
Although some researchers attempt to link WLB research to performance in this manner,
few have actually studied it. In 1999, Kossek and Ozeki reviewed the available literature and
found just three studies that actually looked at performance and WLB together. Two found that
18

job performance and work-to- family conflict were completely unrelated (Greenhaus, Bedeian, &
Mossholder, 1987; Netemeyer, Boles & McMurrian, 1996) and the third found a very small
negative correlation with a global measure of conflict (Kossek & Nichol, 1992). It appears that
the family-to-work direction of conflict is more closely related to job performance as one study
found a positive correlation for this direction, but that direction of conflict is not the focus of this
paper.
In the intervening decade since Kossek and Ozeki’s review there has been little progress
made in linking WLB to performance variables. Although there have been a few studies that
have come out since then, they have mainly added some support for the idea that increased job
performance is related to higher levels of the family-interference-with-work direction of
imbalance (Karatepe & Kilic, 2007; van Steenbergen, Ellemers, & Mooijaart, 2007). These
studies have relied on self report ratings of performance and often do not examine the workinterference-with family direction of WLB at all, the primary direction of interest in this study.
This study’s proposed direct correlation between WLB and performance is not clearly supported
in the literature, but there are indications that higher performance may be associated with lower
satisfaction with balance.
In 1987, Greenhaus, Bedeian and Mossholder suggested that job performance may be
negatively related to satisfaction with balance due to the correlation of high performance with the
additional time investment needed in the workplace. This, in turn, detracts from the time needed
or desired for personal life responsibilities and can result in imbalance. This is a more time
based hypothesis and does not focus on strain as the primary consequence or driver of imbalance,
and thus does not fit precisely within the COR framework, but seems to describe a likely
work/life scenario nonetheless.
19

It is also possible that an individual may feel the effects of resource loss from work-life
imbalance but is able to manage it so that it does not obviously affect their functioning at work.
Gilboa, Shirom, Fried and Cooper (2008) looked at the relationship between work-family
conflict and self and manager rated work performance and found a significant relationship
between self-rated performance and conflict but no relationship for supervisor rated
performance. It may be that an individual is truly experiencing a net loss of resources from
overwork at work, but that the resultant strain is not enough to substantively push job
performance lower.
It is also possible that strain can be domain specific, which can help explain the findings
that family-interfering-with-work role conflict is more related to work performance than workinterfering-with-life role conflict (van Steenbergen et al, 2007). When work is interfering with
non-work-life, the resulting unmet obligations in the non-work role may cause concerns and
strain in that role specifically, while the increased time and effort spent in the work domain may
actually lead to increased performance at work.
A further possible explanation for this relationship may have something to do with
recovery time. Some of the research on recovery would suggest that individuals can recharge
and replenish their resources through restorative sleep and rest (Sonnentag & Bayer, 2005). Due
to the typical work schedule of full time employed Americans, individuals enter the work role
soon after waking up, and the strain they feel from a depletion of resources may be felt most
strongly near the end of the day, time that is typically spent in the non-work role (Sanz-Vergel,
Dermerouti, Mereno-Jimenez, & Mayo, 2010). Therefore, strain originating from workinterfering-with-life would not have an effect on role functioning until the individual may have
already exited or be exiting the work role.
20

A final potential explanation for the findings thus far may have to do with the centrality
or importance work holds for individuals. In a study focused on professionals, Halbesledben,
Harvey and Bolino (2009) found that those who were highly involved with their work and found
work to be personally rewarding experienced higher levels of work interference with family.
This suggests that people who enjoy their jobs and who have a personal identity stake in their
profession may be more likely to experience work interference with family. These individuals
are also more likely to be high performers (Diefendorff, Brown, Kamin, & Lord, 2002). This
sort of research suggests that performance may be an antecedent of WLB rather than the other
way around. High performers are more likely to invest time and energy in their work, and this
may cause interference with the non-work role, resulting in lower satisfaction with balance.
To sum up, the role performance plays in WLB remains unclear from the present
research. Although some researchers theorize that poorer performance may be an outcome of
poor balance, findings are sparse and longitudinal research supporting this claim is nonexistent.
On the other hand, there are several reasons to suppose that higher performance may actually be
an antecedent to poorer WLB, as high performance may demand greater time and energy
investment in the work domain at the cost of the home domain (Greenhaus, Bedeian &
Mossholder, 1987) or the nature of cross domain influences and recovery may mean imbalance
may not significantly impair work performance (Ford et al, 2007). Therefore, this paper treats
performance as an antecedent to WLB rather than as an outcome.

21

Hypothesis 3: Performance will be negatively related to WLB such that those who are
rated as higher performers will indicate less satisfaction with balance.

1

Relationships with Outcomes. In addition to considering variables that may contribute to
WLB, it is also important to look further at outcomes that may be impacted by WLB. In
organizational research, outcomes that affect the well-being of both the individual and the
company of are particular interest. One organizational outcome that has been shown in past

1

The relationships between performance, workload and WLB are likely to be quite

complicated. As argued above, workload is likely to have a negative relationship to both WLB
and performance (Hypotheses 1 and 2, respectively), and these are likely to be negatively related
to each other (Hypothesis 3). Past research has in particular not examined performance well or
consistently in the WLB or workload literature (Janssen, 2001; Kossek & Oseki, 1997) so it is
unclear how these relationships may play out in the final data. It is possible that the
hypothesized relationships may hold when considered in isolation, but that the stronger of the
correlations will swamp the competing relationships when considered in tandem. For example,
the work-life literature shows a quite strong link between workload and WLB. This may mean
that in performance is considered as a moderator of this relationship, its negative relationship
with WLB and workload may be subsumed by the stronger relationship of WLB and workload.
There is not enough of a literature base to presuppose a particular moderating relationship
between workload, performance and WLB, but these relationships will be investigated in the
data analysis phase.

22

research to be of particular interest in performance, workload and WLB research is that of
retention intentions, that is intent to stay with the company. At the individual level, emotional
exhaustion is also often studied in conjunction with these antecedents as well. The sections
below will further elaborate on the proposed relationships.
Retention. Reducing voluntary turnover in employees has long been a focus of
organizations and organizational psychologists have been investigating the phenomenon for
more than 50 years (Brayfield & Crockett, 1955; Campion, 1991). Losing qualified employees
represents costs to the organization not just in terms of loss of those employees’ performance but
also in terms of replacement costs of searching for, hiring and training new employees. For
knowledge-based companies that rely on the expertise of highly trained employees that are in
high demand and short supply, turnover can be especially dysfunctional for the organization
(Trevor, Gerhart, & Boudreau, 1997). If organizations have already gone through the process of
selecting high performers, then ideally they would like to retain these employees if at all possible
to avoid time and monetary loss associated with replacing them.
Although actual retention (observing the actual behaviors of employees staying with the
company) may be the most desired metric for studying retention, due to time and data collection
constraints many researchers must rely on retention intentions rather than observed turnover.
Retention intentions are correlated with actual turnover (Hom, Katerberg & Hulin, 1979), and
can also be used as an “early warning system” to identify employees that may be at the
beginning stages of withdrawal behaviors before they have fully disengaged from the
organization (Podsakoff, LePine, & LePine, 2007). Retention intentions also have the practical
value of being able to be assessed through a survey instrument rather than longitudinal
organizational records.
23

Workload is one job characteristic that can lead to an increased desire to leave the
company. Those who feel overloaded at work may see leaving the job as the only way to
decrease this overload. Research in this area supports this contention (e.g. Rosin & Korabik,
1995; Bolino, Turnley, Gilstrap, & Sauzo, 2010). In a study by Bolino and colleagues (2010), the
researchers found that perceptions of work role overload were positively related to intentions to
quit. These findings have even been replicated across cultures, as well. In a study of both Anglo
and Asian cultures, Spector, Allen, Poelmans, LaPierre et al (2007) found a positive relationship
between workload and intention to quit across both cultures.

Hypothesis 4: Workload and retention intentions will be negatively related such that
those with higher workloads will be less likely to indicate an intention to stay with the
company.

Performance has a slightly more complicated relationship with retention intentions
because there are presses and pulls that can both encourage high performers to stay while
providing greater opportunities for employment elsewhere. Past research has suggested that high
performers have greater job mobility and job options and that this knowledge of mobility may be
related to a greater potential to turnover (Nyberg, 2010). However, high performers are also
more likely to feel more positively about the company and be more committed to it (Becker &
Cropanzano, 2011). These positive feelings about the organization may help explain why high
performers generally indicate a lower intent to quit despite greater perceived mobility. Metaanalytic results show that all types of ratings of performance (self, manager and objective) are
negatively related to intentions to quit (Zimmerman & Darnold, 2009).
24

There are also organizational level factors that can contribute to the turnover intentions of
a high performer. Hausknecht, Rodda, and Howards (2009) conducted a study in which they
found that high performers were more likely to stay at the company if they perceive job
advancement opportunities at the company, if they are rewarded for high performance and if they
perceive the organization as prestigious. The organization that will be used in this study uses a
performance management system that directly links bonuses to individual performance and
makes a concerted effort to maximize career development opportunities available for high
performers. In addition, this organization holds a position of relatively high prestige in its
industry. All of these factors combined would suggest that high performers would be more
likely to indicate a desire to stay with the company.

Hypothesis 5: Performance and retention intentions will be positively related such that
those with higher performance ratings will be more likely to indicate an intention to stay
with the company than those with lower performance ratings.

Turnover intentions have been repeatedly been invoked as one of the reasons
organizations should care about and attempt to alleviate problems with WLB (Eby et al, 2005).
Despite this, retention intentions have infrequently been directly studied in work-life research
(Allen, Herst, Bruck & Sutton, 2000). What findings there are indicate that those who feel more
conflicted are more likely to show a greater intent to turnover (Rode, Rehg, Near & Underhill,
2007; Hammer et al, 2011; Blomme, van Rheede, & Tromp, 2010). As mentioned above, for
knowledge-based companies, which are often characterized by an intensive (and expensive)

25

selection process, retaining high talent employees is especially important (Trevor, Gerhart, &
Boudreau, 1997), and improving WLB may be a way to positively affect retention intentions.

Hypothesis 6: Work-life balance will be positively related to retention intentions such
that those who report higher balance will also report higher intentions to stay with the
company.

Satisfaction with WLB could also serve as a moderating variable between job
performance and retention. Due to the inconclusive research addressing job performance and
work-life balance, there is no research available directly examining this potential relationship.
However, work-life balance variables have been found to moderate a variety of relationships
between job characteristics (including workload and autonomy) and outcomes such as retention.
Hypotheses 5 and 3 state that those who are higher performers are more likely to show
intent to stay with the company and are also more likely to experience dissatisfaction with WLB.
Dissatisfaction with balance, in turn, is hypothesized to relate to lower intent to stay with the
company (Hypothesis 6). WLB may moderate the relationship between performance and
retention intentions such that satisfaction with WLB will affect the strength of the relationship
between performance and retention. As mentioned earlier, high performers are more likely to
have greater perceived mobility options, although co-occurring higher job engagement and
organizational commitment may often result in greater retention (Becker & Cropanzano, 2011).
WLB dissatisfaction, however, could result in high performers feeling dissatisfied with
the company as a whole, thus reducing the job commitment characteristics that buffer against
highly mobile employees seeking positions elsewhere. Research has shown a link between WLB
26

and employee perceptions of organizational support or organizational values (Chalofsky, 2008).
Those who are dissatisfied with WLB can see their organization as unsupportive and out of sync
with personal values (Thompson, Beauvais, & Lyness, 1999). This may lead to feelings of
“misfit” between the individual and company. Misfit, in turn, is related to turnover (actual and
intentions) as employees seek a better fit elsewhere (Dawley, Houghton, & Bucklew, 2010,
Verquer, Beehr, & Wagner, 2003). With the greater mobility options available to high
performers, dissatisfaction with WLB may particularly affect their intentions to remain with the
company.

Hypothesis 7: WLB will moderate the relationship of performance with retention
intentions such that those with higher WLB and higher performance will be more likely
to indicate intentions to stay with the company than those with lower performance. See
Figure 6 for a graph of this moderation.

Emotional Exhaustion. In addition to the more organizationally focused outcome of
retention intentions, other outcomes that affect employee well-being and happiness have been
examined in relation to WLB. Emotional exhaustion is one such variable. Wittmer and Martin
(2010) define emotional exhaustion as a “general, yet severe, lack of energy that is accompanied
by feelings that one’s resources are spent” (p. 608). Emotional exhaustion is a component of
burnout, a concept that was initially defined by Maslach (Maslach, 1984; Maslach & Jackson,
1984) as a unique type of stress syndrome that relates to work and affects individual mental and
physical health (Shirom & Melamed, 2005). Maslach’s (1984) conceptualization of burnout
typically breaks the phenomenon into three components, namely emotional exhaustion,
27

depersonalization (also called cynicism) and a decreased feeling of personal accomplishment.
Initially, burnout was theorized to apply mainly to workers in the service industry who have to
deal with customers or clients on a regular basis (Maslach, 1984) and much of the research
focuses on these service-type occupations (e.g. Abdallah, 2009; Mikkelsen & Burke, 2004).
Over time, however, it has become apparent that employees in non service work also can
experience burnout and research in this area has expanded to include jobs that are decidedly not
service oriented (Hetland, Sandal & Johnsen, 2007; Lewin and Sager, 2007). For non-service
employees, it appears that the facet of emotional exhaustion is the most central in understanding
experiences of burnout (Baba, Jamal & Tourigny 1998) and within the work-life research stream
emotional exhaustion generally shows the strongest relationships to work-life variables (e.g.
Montgomery, Panagopolou, & Benos, 2006; Thompson, Brough, & Schmidt, 2006).
Although many researchers choose to examine all three factors of burnout in empirical
research, Maslach herself argued that burnout should not be considered a single concept but
rather that each sub-component has unique relationships to various outcomes and antecedents of
interest (Maslach, 1984). This would support the use of a single component rather than
considering all three in a research study if only one was of particular interest. Shirom (1989)
argues that emotional exhaustion is the true central element of burnout since depersonalization
and feelings of diminished personal accomplishment can be understood as coping mechanisms
for or reactions to emotional exhaustion. Cynicism can be seen as a coping mechanism as it
serves as a way for individuals to distance themselves from work they find to be stressful. A
diminished feeling of personal accomplishment may also result from feelings of emotional
exhaustion as individuals experience this type of stress as indicative of their performance at

28

work. For these reasons, the emotional exhaustion facet of burnout will be used as the outcome
of interest independent of the other facets of burnout.
In addition to the qualities described above, emotional exhaustion is theorized to be
more prevalent in occupations where there is a high amount of personal job involvement or
where there is a mismatch between demands and resources (e.g. Bakker, Demerouti, &
Schaufeli, 2002, Maslach, Schaufeli & Leiter, 2001). These findings would suggest that
employees who work in knowledge based professions, characterized by strong ties to the work
role, may be at a greater risk of emotional exhaustion than other types of organizations.
Emotional exhaustion has been theoretically linked to the COR theory and role overload such
that individuals who have demands that exceed their resources will experience feelings of stress
at these points of mismatch (Hobfoll, 1988), and prolonged experience of this stress can lead to
emotional exhaustion (Boles, Johnston, & Haire, 1997).
Emotional exhaustion is an outcome that is of considerable interest to both organizations
and employees. From an individual standpoint, emotional exhaustion decreases individual
happiness and well-being (Schaufeli, Bakker, van der Heijden, & Prins, 2009), decreases life and
job satisfaction (Um & Harrison, 1998) and decreases mental and physical health (Iacovides, et
al 2003; Shirom & Melamed, 2005). From a monetary perspective, stress related costs from
emotionally exhausted employees can be quite high—the American Psychological Association
and the American Institute for Stress estimate that absenteeism, productivity loss, turnover and
health care costs stemming from job stress and emotional exhaustion account for $300 billion per
year (Rupprecht & Grawitch, 2010). Because of the high costs of emotional exhaustion,
researchers and practitioners alike are very interested in indentifying antecedents of emotional
exhaustion with the ultimate goal of mitigating these problem areas before they result in
29

widespread exhaustion. Two antecedents that have frequently been examined in the literature are
workload and work-life balance.
Workload has the strongest base in the literature and is a very robust predictor of
emotional exhaustion. In a study by Xanthopoulou, Bakker, Dollard, Demerouti, Schaufeli et al
(2007), the researchers found that workload was more strongly related to emotional exhaustion
than other types of demands (e.g. emotional demands), and that the workload/emotional
exhaustion relationship proved particularly resistant to moderators. These types of findings have
been replicated numerous times (e.g. Ilies, Dimotakis & DePater, 2011; van der Doef & Maes,
1999). Workload is a relatively pure example of resource loss at work, and represents resource
depletion of the COR theory. Through this depletion, workload may be a primary driver in the
experience of feelings of emotional exhaustion by individual workers.

Hypothesis 8: Workload will be positively related to emotional exhaustion such that
those who report higher workloads will also report more emotional exhaustion.

Work-life balance has also been frequently tied to emotional exhaustion in the
organizational psychology literature. If there is a mismatch between work demands and the
available resources such that the non-work role is routinely left impoverished, this net loss of
resources can be directly related to experiences of emotional exhaustion. This contention has
been supported by a number of studies that have found a positive relationship between work-life
conflict and various aspects of burnout (e.g. Aryee, 1993; Kinnunen & Mauno, 1998; Netemeyer
et al, 1996). Allen and colleagues (2000) reported the weighted mean correlation across studies
to be .42 between experience of work-life conflict and emotional exhaustion.
30

Research into this area supports the contention that WLB issues are antecedent to
burnout. Innstrand, Langballe, Espnes, Falkum and Aasland (2008) looked at longitudinal data
to see if the relationship of work-family interference led to burnout, if burnout led to workfamily interference or if the relationship was reciprocal. They found support for the model
indicating work-family interference predicting lagged burnout, thus supporting the conceptual
placement of burnout as an outcome.

Hypothesis 9: WLB will be negatively related to emotional exhaustion such that those
who report higher balance will report less emotional exhaustion.

The role WLB plays in the relationship between workload and emotional exhaustion will
be more direct than the role WLB plays in the relationship of performance and retention. While
Hypothesis 7 states that WLB will moderate the relationship between performance and retention,
essentially constraining the positive relationship between performance and retention, the role of
WLB in the workload/emotional exhaustion relationship is one of mediation. Within the
framework of COR theory, when work and life conflict, this creates stress points, which in turn
can lead to feelings of emotional exhaustion (as stated in Hypothesis 9).
Workload, on the other hand can follow two paths to emotional exhaustion. For the
direct path, workload can drain personal resources without resulting in work-life imbalance per
se. Thus the direct relationship proposed in Hypothesis 8. In addition to the direct path,
however, workload can also drain personal resources through the extra stress that is experienced
when high workloads result in interference with the non-work role. In this case, the effect of
workload on emotional exhaustion occurs at least partially through the extra resource drain
31

experienced when WLB is thwarted by high workloads. This means that WLB can partially, but
not fully, mediate the effect of workload on emotional exhaustion. The relationship is
hypothesized to be partial rather than full based on aforementioned logic as well as previous
findings (discussed below), as well as on the contention from Baron and Kenny (1986) that in the
social sciences, partial mediation is more likely due to the multi-determinability of many
psychological phenomena.
The research literature supports this idea of partial mediation. In a study by
Montgomery, Panagopolou and Benos (2006), the researchers tested work-family interference as
a mediator between job demands and job burnout. They found that job demands directly
predicted emotional exhaustion, as did work-family interference. In addition, work-family
interference partially mediated the relationship between job demands and emotional exhaustion,
showing that when work-family interference was considered, the strong direct effect of workload
on emotional exhaustion was considerably reduced but still significant. Other researchers have
come to similar conclusions, finding WLB variables to partially mediate the relationship between
workload and various dimensions of burnout (Geurts et al, 2003; Peeters, deJonge, Janssen &
van der Linden, 2004; Peeters, Montgomery, Bakker & Schaufeli, 2005).

Hypothesis 10: WLB will partially mediate the relationship of workload and emotional
exhaustion such that when included in the relationship the effect of workload on
emotional exhaustion will be diminished.

Refocusing on the System: Variables that Can Affect the Relationships of WLB with
Antecedents and Outcomes
32

The paper up to this point has focused on the main path of the conceptual model proposed
in Figure 1 and outlined in the hypothesis model shown in Figure 2. This has focused on direct,
moderated and mediated paths from the conceptual antecedents of workload and performance to
WLB and the outcomes of retention and emotional exhaustion. The next two sections of the
paper will focus on factors that moderate these direct relationships. The first section will focus
on factors that can help individuals deal with WLB once it has already formed, which is
represented in box 2 of Figure 1, namely through support provided by managers and coworkers.
The hypothesized relationships of manager and team support on WLB and retention and
emotional exhaustion are presented in Figure 2. The second section will focus on factors that go
into the formation of WLB (box 3 of Figure 1). The proposed hypotheses that deal with
flexibility and border management are presented in Figure 4 and the hypotheses that deal with
team and manager norms for WLB are presented in Figure 5. This section will include a
discussion of boundary theory, idiosyncratic deal theory, workplace flexibility and the role of
group norms. To aid in clarity of understanding the relationships of the proposed moderated
relationships, graphs of all study moderations have been included in Figures 6 though 18.

Social Support. Within the COR framework, social support can be seen as a way
individuals can gain access to additional resources to compensate from the drains experienced
from work and family demands. Social support has been an explicit aspect of work-life research
from early forays into the field. Greenhaus and Beutell (1985) theorized that “support from
significant others” had the potential to moderate the relationship of work-life imbalance and
outcomes such as psychological well being.

33

Although the social support literature recognizes both instrumental types of support, such
as practical help (like with a work project or child care), as well as emotional support, such as
having a sympathetic listenber to “vent” to (e.g. Michel, Mitchelson, Pichler & Cullen, 2010;
Shumaker & Brownell, 1984), this study will focus on emotional social support. The reason for
this is that emotional support is explicitly seen as a way of dealing with stressors once they occur
(such as feelings of imbalance) while instrumental support can be seen as a way of preventing
stressors from work-life imbalance from occurring (Michel et al, 2010). This portion of the
paper focuses on ways workplace supports can help buffer the effects of low WLB on retention
and well-being outcomes, so a focus on emotional support is more appropriate.
The work-life research literature has found some correlation between organizational and
supervisor support for work-life balance and increased WLB over a variety of studies (e.g.
Carlson & Perrewe, 1999, Michel et al, 2010). However, several researchers have indicated that
support for balance can occur at various levels within the organization—at the top management
level (the level often studied as “organizational”), the direct manager level, and the co-worker
level (Bowen & Orthner, 1991; Eby et al, 2005). In a meta-analysis on support and WLB,
Mesmer-Magnus and Viswesvaran (2005) concluded that although preliminary results indicate
that supportive workgroups and supervisors may help alleviate work-life conflict, more research
is needed to fully examine the impact of these types of support.
Organizational and manager support for balance are often studied in conjunction with
individual characteristics in an attempt to explain more about what affects levels of balance and
how this balance is related to outcomes of interest. In general, the findings are as expected, that
individuals who feel more supported by the organization or manager in balancing their work and
life responsibilities report fewer negative organizational outcomes, such as turnover and
34

absenteeism (Ford et al, 2007). Some researchers propose that the way manager support helps
alleviate the negative effects of higher work-life imbalance is through the simple provision of
social support, which can be a coping mechanism for stressful situations (Behson, 2002). With
reference to Figure 1, this social support can help mitigate some of the negative effects of
experiencing conflict.
Having a good relationship with one’s manager has been shown to be very important to
individual job satisfaction and commitment (Pisarski, Lawrence, Boble & Brook, 2008). In the
work-life balance literature, managers are seen as being especially important in the ways they
explicitly support direct reports in efforts to balance these roles (O’Neill, Harrison, Cleveland,
Almeida, Stawski & Crouter, 2009). One of the ways managers can impact the relationships
between WLB and retention is how they impart the values of the corporation. Managers are often
seen as the frontline representative of the organization’s values and how committed the
organization is to helping employees maintain a happy and healthy lifestyle (Leiter, Gascon &
Martinez-Jarreta, 2010).
Research has shown that manager support for work-life balance is related to greater
employee job satisfaction, retention intentions and improved balance (Lapierre et al, 2008;
Thomas & Ganster, 1995). Manager support can therefore be conceptualized as a resource
individuals may possess that can be used to deal with lower satisfaction with balance. This
resource can buffer the effects of low satisfaction with WLB on outcomes such as retention
intentions, allowing individuals access to greater reserves to deal with imbalance without
negative effects to intentions to remain with the company.
Manager support for balance may also be seen by direct reports as a commitment to help
them deal with WLB issues. This may in turn provide social support and a belief that even if an
35

individual is dealing with imbalance currently; the supports are in place that there is potential for
solving the problems leading to imbalance (Brough & O’Driscoll, 2010). This can increase
commitment to the organization and may make turnover a less likely mechanism the individual
will choose to help deal with dissatisfaction with balance.

Hypothesis 11: Manager support for balance will moderate the relationship between
WLB and retention such that those who report more supportive managers will report
higher intention to remain with the company as compared to those with similar levels of
WLB with less supportive managers. See Figure 7 for a graph of this moderation.

In addition to managers, teams also represent an important means through which an
employee can receive social support at work. As organizations restructure to more distributed
decision making models, with flatter and more responsive hierarchies to meet the needs of a fast
paced, changing marketplace, teams and work groups have become a more preferred way to
organize work and responsibilities (Mathieu, Maynard, Rapp & Gilson, 2008). This has resulted
in an increased importance of the need to consider the ways teams create a context for individual
behavior. Teams have been infrequently studied in work-life balance research and there are calls
for more research in understanding how individual work-life balance is affected by the group
(Mesmer-Magnus, Murase, DeChurch & Jimenez, 2010). As such, the attention this paper turns
on the role of teams in individual WLB is an important contribution to the research.
Like managers, teams can provide resources for dealing with work-life balance
dissatisfaction. As organizations are increasingly relying on teams as the primary work
environment (Tekleab, Quigley & Tesluk, 2009), the frequent interaction of teammates means
36

that they are ideally situated to become an informal support network (Hunter, Perry, Carlson &
Smith, 2010) for individual employees and can offer emotional support in much the same ways
managers can. Only one study has explicitly looked at work-life balance and team support, and
the researchers did not distinguish between team instrumental or emotional supports or look at
outcomes such as retention. However, the researchers did find support for the contention that
individuals in teams with more resources did experience greater work-life balance (Hunter et al,
2010).
Despite the fact that there is little literature directly linking team support, WLB and
retention, the literature that does look at team social support suggests that it would act as a buffer
between stressful events and negative outcomes, such as instances of work-life conflict and
intentions to leave the company. In a study by Heaney, Price and Rafferty (1995), the
researchers found that those who felt more socially supported by their teammates felt better able
to cope with stressors, which in turn buffered their experiences of negative outcomes. It is this
type of experience we would expect in a WLB scenario.

Hypothesis 12: Team support for balance will moderate the relationship between WLB
and retention such that those who report more supportive teams will report greater
intention to remain with the company as compared to those with similar levels of WLB
with less supportive teams. Figure 8 for a graph of this moderation.

The research on manager and team support also indicates that manager support for WLB
likely has an impact on individual experiences of emotional exhaustion as well as retention (as
addressed in Hypothesis 11). Manager and team support for balance will moderate the
37

relationship between WLB and emotional exhaustion in much the same theoretical manner as it
would for retention. The support would act as a buffer for the resource loss experience by a poor
satisfaction of balance, essentially refilling or replenishing the resources lost through the
experience of this type of stress. Van Daalen et al (2006) suggest that one of the mechanisms
this replenishment occurs through is that it increases an employee’s energy levels, one of the
resources central to the COR theory (Hobfoll, 1989). In the work-life context, a supportive
supervisor or team can help an employee cope with dissatisfaction with WLB through
discussions of family-related problems, reinforcing the self-image of the employee and by
reducing stress by showing understanding of the employee’s efforts to balance work and family
life (Halbesleben, 2006; Lapierre & Allen, 2006; ten Bremmelhuis & van der Lippe, 2010). This
type of support is not aimed at fixing any conflict problems a priori; that is, the supervisor is not
reducing the workload or adjusting the hours of the employee, rather he or she is providing social
support for coping with efforts to balance work and family life. This coping in essence occurs
after WLB satisfaction has already formed and would buffer the effect of WLB on the outcome
of emotional exhaustion.
There is some research that supports the idea that supervisor support would buffer the
WLB-emotional exhaustion relationship. In a study that looked at both manager and coworker
support (Hammer, Saksvik, Nytro, Torvatn, & Bayazit, 2004), the researchers found that leader
and coworker support interacted with work-family conflict such that those with higher support
experienced less emotional exhaustion than those with less support.

Hypothesis 13: Manager support for WLB will moderate the relationship between WLB
and emotional exhaustion such that those who receive more support will be buffered from
38

the negative effects of low WLB on emotional exhaustion. See Figure 9 for a graph of
this moderation.

Hypothesis 14: Team support for WLB will moderate the relationship between WLB and
emotional exhaustion such that those who receive more support will be buffered from the
negative effects of low WLB on emotional exhaustion. See Figure 10 for a graph of this
moderation.

This study replicated some important findings in the literature as well as extended the
COR theory to investigate some areas that are less well studied in the work-life literature, such as
links to performance, retention, and the role of teams in the experience of work-life balance.
This represents a substantial contribution to the literature and can help begin to explain the
context within which work-life balance forms.
However, through maintaining a narrow focus on role and COR theory, it is not always
possible to address the finer points of the process of WLB formation and how people achieve it.
Role theory can focus researchers too narrowly on the idea of role conflict as something that
happens to a passive person rather than something actively experienced and constructed by an
individual (Clark, 2000). While role theory does a good job outlining the problem space where
conflict can occur, it is less well adapted to understand the process of how roles interact and how
these interactions are negotiated by the individual. For this, border theory offers better insight.
The purpose of focusing on the process of WLB formation is to try to get a better
understanding of work-life balance. There has been a lot of research in the area, and poor worklife balance has been linked to a host of negative outcomes (Allen et al, 2000), but there is still
39

very little information about what individuals and organizations can do to effectively change or
deal with imbalance (Hammer et al, 2011). Once the process of how individuals actually make
work-life balance decisions is better understood, hopefully it will be possible to target
interventions at this process that can help maximize performance in both roles and equip people
to make decisions that best reflect their personal values and preferences. With that in mind, this
paper will begin the exploration of this decision making process by considering how border
theory may help explain those decisions, and the role idiosyncratic deals and workplace
flexibility can play in helping employees achieve their ideal work-life balance.
Border Theory. Border theory builds upon role theory but focuses on what occurs at the
borders of the roles of work and non-work to better understand how individuals develop their
own sense of work-life balance and satisfaction. Border theory is based on the idea that borders
represent real demarcations between two concepts and harkens back to Lewin’s concept of “life
space”. Lewin posited (as reported in Rychlak, 1981) that people organize their different roles
within this overarching life space and different roles can be closer or more distant from each
other dependent on the individual’s preferences and beliefs about roles. The borders of these
roles can be more or less flexible depending on the degree to which each role interacts with
another. For example, a person may have very strong beliefs about treating family members
with respect but at work continually belittles and devalues coworkers. This person would have a
strong border between family and work behaviors as their behavior is so different in one role
from the other.
Border theory defines borders as occurring in several potential locales (Clark, 2000).
Borders can be physical, temporal or psychological. Physical borders are those that occur by
virtue of physical space. For example, non-work can be defined by the physicality of an
40

individual’s home or community and can be defined by the literal walls of these places.
Contemporary work in industrialized nations, on the other hand, often occurs at a place
specifically separate from non-work locations. For this type of border, roles are enacted within
their particular spaces, such as non-work happening within the home while work happens at the
office. Temporal borders are the second type of border. These have to do with times specific
roles are usually enacted. For individuals working a standard full time schedule, weekdays
between the hours of 8:00 am and 5:00 pm are the typical hours where work roles are enacted,
while earlier mornings, evenings and weekends are hours where non-work roles are enacted.
Border theory suggests that the greater degree of flexibility a person has to create their
own more or less permeable or flexible boundaries, the better he or she is able to construct a
work-life arrangement that aligns with his or her own preferences. In a truly flexible work
environment, someone who prefers less rigid boundaries would be free to arrange their schedule
to allow for frequent home/work boundary spanning events. Someone who prefers a more
segmented experience of work and home life would be free to construct a more rigid work and
life schedule that minimized the amount of boundary spanning events.
One of the main ways organizations have attempted to accommodate employee border
management preferences is through the provision of flexible work arrangements. Many
organizations have explicit policies that state specific flexible arrangements that are supposed to
be accessible to employees, such as flextime or telework options, although using these policies is
often left up to the discretion of individual managers and workgroups (Richman, Civian,
Shannon, Hill & Brennan, 2008). The next section will discuss some of the literature that
supports why flexibility may a good tool to help employees better balance their work and non-

41

work lives and the role idiosyncratic deals with managers might play in an individual’s ability to
take advantage of these flexible arrangements.

Flexibility. One tool organizations can provide their employees to help them balance
work and non-work is in the flexibility in time and place of work (Kopelman, Prottas,
Thompson, & Jahn, 2006). Most research on flexibility of work indicates that providing this
type of accommodation is generally beneficial to both employee and employer (Grawitch &
Barber, 2010). Telecommuting, the ability to work from home or another location through the
support of communication technology, has frequently been cited as a major way in which
organizations can provide flexibility in work schedules for workers (e.g. Morganson, Major,
Obron, Verive, & Heelan, 2010; Standen, Daniels, & Lamond, 1999). In a white paper
summarizing research findings and making recommendations for best practices, the Center for
Work and Family recommend that organizations make more use of flexible work arrangements
to increase the ability for workers to effectively manage work and life (Hamilton, 2010).
Although some managers and organizations are reluctant to institute flexible work arrangements
due to fears of productivity loss and/or alienation of employees, research has indicated that
productivity is not affected by use of flexible work arrangements (Dunham, Pierce & Castaneda,
1987; Lingard, Brown, Bradley, Bailey & Townsend, 2007) and employee commitment is not
negatively affected by telecommuting for half or less time per work week (Golden, Veiga &
Dino, 2008).
Flexible work arrangements have been linked to the work-life literature. When
organizations seek to help alleviate work-life conflict strains on employees, they frequently look
to two main types of provisions—those that provide tangible supports to ease conflict (like
42

childcare or concierge services) and those that encourage flexible work accommodations
(Giardini & Kabst, 2008). Some of the research on the effectiveness of these programs as a
whole has been equivocal, making organizations rethink the utility of providing these services to
employees (Kossek & Ozeki, 1998). However, Grawitch & Barber (2010) took a closer look at
the work-life balance program efficacy. They found that support for resource supports was
ambiguous, but the benefits of flexibility work arrangements to work-life balance were well
supported. Kopelman, Prottas, Thompson and Jahn (2006) point out that in comparison to other
types of workplace work-family supports (such as onsite childcare) flexibility may actually help
prevent work-life conflict before it occurs rather than provide backend support once it already
happens. This supports the proposition that use of and satisfaction with flexible work
arrangements would have a direct effect on WLB satisfaction.

Hypothesis 15: Frequency of use of flexible arrangements will be positively related to
work-life balance satisfaction such that those who report greater use of flexible
arrangements will report greater satisfaction with balance.

There has been very little research directly comparing employee satisfaction with
flexibility to actual use of flexible arrangements. Most studies look at either one or the other,
and both types of studies have found similar findings (Lee, Magnini, & Kim, 2011, Stavrou &
Kilaniotis, 2010). For the purposes of this paper, there are no reasons to presuppose that actual
use of or satisfaction with flexibility should function differently with regard to these
relationships.

43

Hypothesis 16: Satisfaction with flexible arrangements will be positively related to worklife balance satisfaction such that those who report greater satisfaction with flexible
arrangements will report greater satisfaction with balance.

In addition, it is possible that the use of flexible work arrangements affects WLB through
the ways it can moderate the effects of high workload. Hill, Erickson, Holmes and Ferris (2010)
did an in depth study where they examined place and time flexibility in a large, multi-national
tech firm. Place flexibility was defined as the ability to work remotely on occasion (telework)
and time flexibility was defined as having some discretion regarding start and stop time for work
days. They found that both types of flexibility were related to decreased work-life conflict. In
addition, they found that employees who reported more flexibility were able to work longer
hours before reporting work-family conflict, often 8 to 16 more hours of work per week than
other employees with less flexibility.
It would appear, therefore, that flexible work arrangements may make it possible for
employees to arrange their work and personal lives so that the two roles do not conflict, thus
avoiding instances of work-family conflict. This would suggest that for employees with equal
workloads, those with flexible arrangements would be better able to manage that workload
without detrimental effects to satisfaction with balance.

Hypothesis 17: Frequency of use of flexible arrangements will moderate the relationship
between workload and WLB such that those who report more use of flexible
arrangements will also report greater work-life balance despite high workloads. See
Figure 11 for a graph of this moderation.
44

Hypothesis 18: Satisfaction with flexible arrangements will moderate the relationship
between workload and WLB such that those who report more use of flexible
arrangements will also report greater work-life balance despite high workloads. See
Figure 12 for a graph of this moderation.

The benefits of workplace flexibility extend beyond just the effects it may have on
improving work-life balance experiences. Research indicates that flexible work arrangements
are also directly related to the outcomes of retention intentions and emotional exhaustion.
Theoretically, this contention fits well within the boundary literature as well as the COR theory
literature. Having flexibility gives individuals more control over their own schedules and allows
them resources to respond to demands as they arise and within the boundary conditions they find
most optimal (Voydanoff, 2005). The increased resources individuals may accrue from
flexibility may enhance their feelings of fit with the organization and in turn increase their
intention to stay with the company (Edwards, 1996).
For retention, the available literature suggests that satisfaction with flexibility is related to
a increased desire to stay with the company. Lee, Magnini and Kim (2011) found that
satisfaction with schedule flexibility predicted lower turnover intentions. Similarly, a study by
Stavrou and Kilaniotis (2010) found that as schedule flexibility increased (work from home
options and flextime), turnover decreased. As these two studies demonstrate, both perceptions of
workplace flexibility and actual use of flexible arrangements are related to intentions to stay with
the company (Richman, Civian, Shannon, Hill & Brennan, 2008).

45

Hypothesis 19: Frequency of use of flexible arrangements will be positively related to
retention intentions such that those who report more use of flexible arrangements will
also report greater intentions to stay with the company.

Hypothesis 20: Satisfaction with flexible arrangements will be positively related to
retention intentions such that those who report more satisfaction with flexibility will also
report greater intentions to stay with the company.

For emotional exhaustion, the theoretical basis for the direct relationship of flexibility to
emotional exhaustion is similar to that of retention. Flexibility can provide the individual with
greater resources with which to deal with demands, which in turn means less net resource loss
and therefore less emotional exhaustion. In the burnout literature, flexible work arrangements
have been found to be related to better health, including reduced emotional exhaustion and better
physical health (Butler, Grzywacz, Ettner, & Liu, 2009). Studies that look at flextime and those
that look at telecommuting both find that individuals who use those flexible supports experience
lower emotional exhaustion than those that do not use the supports (Costa, Sartori, & Åkerstedt,
2006; Grzywacz, Carlson, & Shulkin, 2008).

Hypothesis 21: Frequency of use of flexible arrangements will be negatively related to
emotional exhaustion such that those who report more use of flexible arrangements will
also report lower levels of emotional exhaustion.

46

Hypothesis 22: Satisfaction with flexible arrangements will be negatively related to
emotional exhaustion such that those who report more satisfaction with flexibility will
also report lower levels of emotional exhaustion.

Preferences for types and frequency of border crossings are a relevant aspect of border
theory and apply to the concept of flexibility as well. As mentioned above, individuals can vary
in the degree of permeability they prefer between the borders of work and non-work. Some may
prefer to the switch back and forth between work and non-work several times a day while others
may prefer to have much more rigid boundaries. The research literature has divided these two
types of preferences into the concept of segmentation and integration.
Nippert-Eng (1996) discusses how individuals prefer to cross the border between work
and home life. There are those that prefer to keep work and life entirely separate (called
segmentors) such that borders are only crossed a minimum amount during the day. For such an
individual, when she is at work, she does not attend to home issues at all and when she is at
home, work does not intrude. Others prefer to enmesh work and life roles more (called
integrators) such that borders between work and non-work can be crossed multiple times during
the day and at many places. An example of an integrator may be someone who feels comfortable
switching gears at work to pay personal bills online and also feels comfortable taking a work call
at home at the end of the day.
It is important to note that neither being an integrator or a segmentor is necessarily better
or worse. The degree to which an individual can arrange their work and life so that the
boundary-crossings occur most in line with their values can help determine their satisfaction with
both roles (Olson-Buchanan & Boswell, 2006). The empirical literature is very sparse, however,
47

and the relative benefits or detriments to segmenting or integrating work and life domains are
largely unknown (Bulger, Matthews, & Hoffman, 2007).
Although flexible work arrangements have found to be largely beneficial to the WLB
satisfaction of employees, it is possible that certain individual difference preferences could
impact the degree to which flexible supports are beneficial to employees. In border theory,
people are assumed to have preferences for either segmentation or integration of work and nonwork-life (Bulger, Matthews & Hoffman, 2007). Individuals who prefer a more integrated
approach may find flexible work arrangements to be ideally suited to maintaining these more
flexible boundaries. For those who prefer a more segmented relationship between work and nonwork-life, however, greater flexibility in work time and place may pose an unwelcome intrusion
into non-work time. Therefore, integrators should benefit from flexible work arrangements to a
greater degree than segmentors.

Hypothesis 23: A preference for segmentation will moderate the relationship between
flexibility and WLB such that those who prefer segmentation will not report higher WLB
satisfaction with increases in flexibility. See Figure 13 for a graph of this moderation.

Border theory makes it possible to better understand how individuals can work to create
their own balance. Some of the danger of this theory, however, has to do with the degree of
responsibility this places on an individual who may be having trouble balancing their own work
and life demands. Taken at the extreme, the argument could be made that every individual is in
charge of his or her own boundaries between work and life, and any problems he or she
encounters in managing his or her work-life balance can be attributable to a non-optimal
48

management on the part of the individual. This view, of course, ignores the fact that people do
not make decisions in a vacuum and there are presses from both work and home life that can
disrupt preferred patterns of work-life management. Indeed, research shows that although
organizations often offer flexible work options, frequently employees do not take advantage of
these programs even if they experience a need for them or a desire to enroll due to felt pressures
in their work environment (Grawitch & Barber, 2010).
Idiosyncratic deal theory is one way to try to understand how it is that individuals are
able to enact their preferences within the context of their workgroup and particular job.
Idiosyncratic deals are a particularly formalized way people can take advantage of flexible work
opportunities and the discussion does not preclude the possibility that some people obtain
flexible working arrangements in a much more informal manner. However, idiosyncratic deal
theory will be used as a frame for thinking about a particular way people can negotiate their
border preferences. The next section will explore how idiosyncratic deal theory can be used to
help better understand the contextual presses from work that could interfere with the effective
utilization of individually optimized boundary management strategies.

Idiosyncratic Deals. Idiosyncratic deals (i-deals) are a construct developed by Rousseau
(2001) to help explain how people are able to construct work arrangements that fit more closely
with individual preferences and abilities than the strict job description allows. She defines ideals as an “individualized arrangement between valued workers and their employers [that] is the
product of negotiation” (p. 260, Rousseau, 2001). Rousseau argues that i-deals are likely on the
rise due to three employment trends. First, a hypercompetitive market means that there is an
employment demand for knowledge workers that have specific and distinct competencies, and
49

this demand means that these employees have greater power to negotiate within current
positions. Second, as the traditional employment model of “one company for life” is weakened,
the decreased job security based model allows for greater customization of employment
conditions. Third, as choices within the employment market expand, employees may have
greater expectations for non-standardized jobs. The contemporary use of i-deals is mainly
focused on retaining valuable workers.
Rousseau (2001) compares i-deals to psychological contracts in that they are related to
the exchange relationship of employee and employer but have an important distinction. A
psychological contract “refers to individual beliefs regarding the exchange relationship between
themselves and the organization” (p. 261, Rousseau, 2001). The belief aspect of this definition is
central to the concept of psychological contract and is the key distinguishing factor between ideals and psychological contracts. While psychological contracts rely on the subjective beliefs
of employees about their treatment by the organization, i-deals are based on actual differences in
treatment. Rousseau (2001) goes on to explicitly state that i-deals are not politics or favoritism
as the purpose of i-deals is a positive outcome for both firm and employee such that they fulfill
actual, legitimate needs of both while reflecting fair differences in treatment rather than personal
bias.
Another aspect of i-deals is the idea that they are explicitly negotiated between an
employee and the employer (Rousseau, 2001). Different organizations and/or groups can have
difference zones of negotiability that dictate what may be possible or not possible to negotiate
and these zones can be shaped by government or union regulations, corporate culture and
industry norms and individual differences. For more bureaucratic organizations, for instance,
there is greater standardization of job descriptions and less room for flexibility in negotiating
50

changes to these descriptions. More knowledge-based and innovative companies, on the other
hand, have a larger zone of negotiability as they seek to retain in-demand workers with large
amounts of business and market knowledge in a competitive environment. The ability to make ideals allows managers greater latitude in motivating employees and increasing engagement.
Idiosyncratic deals are related to WLB mainly through the ways they help explain how
individual employees may be more or less able to take advantage of flexible work arrangements.
Although the research base linking i-deals to WLB is not extensive, preliminary findings show
that greater flexibility i-deals are linked to lower work-family conflict. The main aspect of
idiosyncratic deals that is seen to relate to work-life balance is that of flexibility. In a study by
Hornung, Rousseau and Glaser (2008), the researchers found that those who had negotiated ideals related to flexibility (specifically part-time work or telework options) had decreased workfamily conflict related to their peers who had not negotiated these types of i-deals.
Idiosyncratic deals are unique in that they require specific negotiation with supervisors
and are one way workers can take a proactive approach in creating a more flexible work
schedule. It is also possible that there are other ways and other strategies individuals use to
create a working schedule and environment that support the work-life balance they seek. Part of
the purpose of the qualitative investigation portion of this study is to obtain a better
understanding of the specific things individual employees do to help balance their work and nonwork-life that may include informal actions that go along with these more formal i-deal
negotiations.
Study 2 was conducted to investigate specific strategies for balance that individuals use
to help alleviate the negative effects of workload on balance. The specific hypotheses that
stemmed from the findings from Study 2 will be further explicated following the presentation of
51

the results from that section. Hypothesis 24 is presented here to hold the place of these findings.
Ultimately, an additional 24 hypotheses were developed based on Study 2 and incorporated in
Study 3, mainly having to do with boundary management behavior, detachment and control. For
a full list of these hypotheses, please see Table 1.

Hypothesis 24: WLB management techniques will moderate the relationship between
workload and WLB such that those who use more techniques will be buffered from the
negative effects of high workload on WLB. See Figure 14 for a graph of this moderation.

The working environment is important to work-life balance satisfaction formation not
just in the ways it impacts the specific techniques and negotiations an individual uses and makes
in balancing work and non-work roles, but also in the ways it can constrain these types of
behaviors. Work-life researchers hypothesize that one of the main reasons that workers do not
take advantage of work-life supports offered by organizations to a larger degree is that they
perceive more informal barriers to use from their direct manager or workgroup (Eaton, 2003;
Lewis, 1997). The following section will explore how perceptions of normative experiences of
work-life balance of both managers and teams as well as actual experience of work-life balance
of managers and teams can create an environment that may constrain individual WLB
satisfaction.

Norms for Work-life Balance
A recent paper by Mescher, Benschop and Doorewaard (2010) discusses the use of WLB
supports within organizations within the framework of implicit hegemonic power. The main
52

thesis of this argument is that within organizations broadly and workgroups specifically, a great
deal of power is wielded by groups of people not by coercion but rather by reliance on normative
explanations of the “way things go” in organizations or groups (Doorewaard & Benschop, 2003;
Dooreward & Van Bijesterveld, 2001). The authors argue that these hegemonic normative power
routines regulate behavior within the workgroup without being openly questioned, and that the
norms that are supported by this power are gradually built through solidifying the assumptions,
norms and values of the dominant group.
The process of norm formation compels consent and compliance from members even
though some members might not wish to follow these practices. A work-life example may be
that a work group has created a norm of “face-time” in which members of the team are expected
to be at their desks from 8am until 6pm every day regardless of workload and to stay late if
needed, creating a norm of a greater than 50 hour work week. This would result in reduced
flexibility options and possibly long hours that might be in opposition to a stated organizational
goal (flexible options for employees) but is supported by the manager and other members of the
workgroup as being legitimate and based on “the way things are done around here”. In an
organization where greater power and decision latitude is given to individual work teams, it is
even more likely that workgroups may form their own hegemonic routines based on implicitly
agreed upon behaviors.
One way norms can be formed is through the role modeling of proximal leaders. In the
work-life literature, managers are important to the formation of these norms due to the way they
can act as role models and interpreters of organizational culture. A manager may explicitly tell
her direct reports that work-life balance is encouraged, but if these workers observe her putting
in long hours and feeling stress from not meeting personal life obligations, these encouragements
53

can lose their value. In this way, manager support for balance can be interpreted by the direct
reports as merely lip-service rather an actually enacted support (Gregory & Milner, 2009). It is
possible that manager role model WLB behavior could act as a resource to help people manage
larger workloads without experiencing detriments to WLB.
There is some research to suggest that manager level of work-life balance is a valuable
variable to study apart from perceptions of norms for WLB. In a study by O’Neill et al (2009)
the authors found that those who had managers with higher work-to-family conflict reported
higher conflict themselves as well as a greater intent to leave the company. This may suggest
manager WLB can influence subordinate WLB through mechanisms of role modeling in such a
way that effective means for dealing with challenging workloads may be more apparent for those
with well balanced managers than for those without.

Hypothesis 25: Manager work-life balance will moderate the relationship of subordinate
workload and work-life balance such that those with more balanced managers will also
be more balanced, regardless of level of workload. See Figure 15 for a graph of this
moderation.

It may also be possible that it is the perception of having a manager who values WLB
that can affect subordinate WLB rather than manager WLB per se. If an individual feels that
their manager supports prioritizing both work and life responsibilities, he/she may feel more
comfortable making decisions that make WLB a priority as well (Thompson, Brough & Schmidt,
2006). In essence, the individual perceives a norm of WLB from their manager, whether or not
the manager is actually enacting this norm or not. This could lead the employee to make
54

decisions in balance work and non-work-life in such a way that more optimally reflects their
preferences without feeling that he/she will suffer undue professional consequences for arranging
their work and life responsibilities in such a way.

Hypothesis 26: Perceptions of manager expectations about work-life balance will
moderate the relationship of subordinate workload and work-life balance such that those
with mangers what expect balance will also be more balanced, regardless of level of
workload. See Figure 16 for a graph of this moderation.

In addition to perceived manager and organizational support for balance, a third variable
has been proposed to understand how individuals experience WLB, namely that of the team. The
WLB of a team is important to consider not necessarily as a team level construct that requires
agreement to truly represent the group as a whole, but rather as an environmental type of
variable. It is entirely possible that individuals within a team will vary on their levels of WLB
due to reasons that do not pertain to the workgroup specifically. For example, one group may
consist of some members with small children who place many demands on their times and some
members fresh out of college with few non-work demands and these people may experience
WLB differently for reasons that do not have to do with the workgroup. The mean level of team
WLB may be relevant to individual WLB, however, in the way that it creates an environment
that the individual functions within, and can help influence norms and behaviors at the team
level.
Using the distinction set forth by Kozlowski and Klein (2000), team WLB will be
considered a compilation variable rather than a composition variable. Therefore “team WLB”
55

does not mean the equivalent of individual WLB; that is, we do not expect teams to necessarily
have uniform values for WLB satisfaction. The team level of WLB is more of a compilation
variable because it takes on different properties at the team level, namely those of a more
environmental nature.
Work-life researchers, while acknowledging the social effects on individual experiences
of balance have rarely looked at how team experiences of balance can affect individual
experiences, both through social support and through more instrumental support of allowing
employees to make decisions about time and resource allocation that may occasionally prioritize
life roles over work. With reference to the norms literature mentioned above, it is possible that
the WLB of other members of the team may help shape the group norms surrounding WLB of
the team which may make it more or less difficult for individuals to make decisions that more
closely reflect their personal values about work-life balance. As of yet, very little research has
been done that has looked at the effect of team WLB or team norms on individual experiences of
WLB.

Hypothesis 27: Team WLB will moderate the relationship between individual workload
and individual satisfaction with balance such that those in teams with higher balance will
be buffered from the negative effects of workload on balance. See Figure 17 for a graph
of this moderation.

Perceptions of team WLB norms would also be important for individual levels of WLB
satisfaction. If an individual makes a decision to attend to life responsibilities at the cost to work
responsibilities, teammates should be the most proximally affected by this decision. If team
56

members are willing and able to make these adjustments in response to individual attempts to
balance work and life, then it stands to reason that individual balance could be affected by these
team variables. In addition, work norms are often created at the team level that make explicit
what sorts of hours and workloads are appropriate as well as what sorts of family interferences or
accommodations are seen as acceptable or not within the group context. It is possible that some
groups are more flexible about meeting the changing WLB demands of their members while
others are less so. Those who are part of teams that are more accommodating would be expected
to be better equipped to deal with higher workloads without suffering negative consequences to
WLB.
It is also important to acknowledge the contextual power of coworkers and teammates.
Because i-deals are predicated on unequal treatment of individuals based on preferences, job
requirements and negotiation, coworkers can be seen as interested third parties. Lai, Rousseau
and Chang (2009) found that coworker acceptance of i-deals can affect their effectiveness. If
coworkers perceive i-deals as legitimate and the result of a beneficial social exchange between
workers and employers rather than an economic exchange they are more likely to accept and
support their use within the work group.

Hypothesis 28: Perceptions of team WLB norms will moderate the relationship between
individual workload and individual satisfaction with balance such that those in teams
with norms for higher balance will be buffered from the negative effects of workload on
balance. See Figure 18 for a graph of this moderation.

Exploratory Research Questions.
57

In addition to these specific hypotheses, this paper includes some exploratory research
questions that will be included in the qualitative portion of the project. Because there has been
little research done specifically examining how people make decisions about work and non-work
balance (Powell & Greenhaus, 2006) and even less qualitative research, there are many potential
fruitful areas the qualitative interviews investigated. The structured interview questions aimed to
clarify and dig deeper on how workers make these decisions, what they are happy and unhappy
about in their own work and non work balance, and what they see as effect and ineffective
management strategies. Some of the questions explored are as follows:

How do workers describe their work-life balance and what strategies do they indicate
they use to help them manage before conflict occurs as well as after? How do people use flexible
arrangements—is it through formal negotiation or a more casual assumption of use? What role
does the manager and/or team play in a person’s decision to use flexible arrangements, or other
methods of helping balance work and life?

58

Methods Overview
Multi Method Approach.
This project represents a multi method approach toward understanding WLB, it
antecedents, outcomes, moderators and formation. Experts in the fields of social science
encourage the use of multiple methods in order to fully assess the construct space of a variable of
interest (Singleton & Straits, 1999). Morgan (1998) recommended a thoughtful application of
qualitative and quantitative methods in order to achieve the goals of a particular research project
and this paper aims to do so with a mixed qualitative and quantitative design. This paper offers
many contributions to the literature of work-life balance, including furthering the quantitative
understanding of how role theory applies organizationally relevant variables of performance and
retention and more closely investigating variables that can affect both the formation of WLB
satisfaction and buffer the affects of WLB on outcomes. Working together, these three studies
will be used to both help develop new understanding of WLB formation and to test novel
findings in a quantitative manner.
The first study replicated and extended some findings from the work-life balance
literature linking work-life balance with retention, while more closely looking at the role
performance, workload and teams play in this relationship. Study one served as a first step in
thinking about how the WLB of those around an individual, including managers and team
members, can affect the degree to which an individual remains committed to staying the
company. This study mainly focused on the explanatory power of role theory and COR theory
for these relationships and empirically tested several of the hypotheses presented in the
introduction (see Table 1 for a summary of which studies examine which hypotheses). The

59

purpose of the quantitative Study 1 was to test these hypotheses as well as guide purposive
sampling for Study 2.
Studies 2 and 3 took a closer look at what can affect WLB satisfaction formation or
buffer the effects of WLB satisfaction on the outcomes of retention and emotional exhaustion. In
order to better understand how individuals experience work-life balance, Study 2 consisted of
focused qualitative interviews that specifically asked participants questions about their work-life
balance, how they managed work and life roles, how supportive they perceived managers and
coworkers, what aspects of their work environment they saw as challenging to balance work and
life roles, and the degree of flexibility they perceive within their roles to work in non-traditional
locations or times. Singleton and Straits (1999) suggest that qualitative methods are particularly
useful when investigators want to more closely assess nuance and meaning, which makes its use
in understanding the process of WLB formation particularly relevant.
Study 3 built on the findings of Study 2 by quantitatively examining how individual
preferences for boundary management, coworker and manager support, and formal and informal
flexible work arrangements can affect the relationships between workload, WLB and the
outcomes of retention and work related emotional exhaustion. Study 3 also was able to sample
participants at the group level; further investigating relationships at the group level of analysis
and further exploring ways group dynamics may play into work-life balance formation and
experiences. This use of quantitative methods replicated and extended some of the analyses that
were tested in Study 1 and it was possible to delve more in depth into a greater array of
constructs related to WLB.
Research Site. Conducting research within a single organization has both potential
benefits and drawbacks. On one hand, the single organization context may limit generalizability
60

to other firms or industries. On the other, having broad access to a single organization can give
the researcher a greater degree of latitude and control in study design. Ideally, if a researcher is
confined to one organization, this organization should encompass enough diversity of job and
work groups that individuals in different roles throughout the company would have sufficiently
different experiences to drive testable variance.
The type of organization and its structure becomes important when considering this type
of field study. When considering work-life balance formation and experiences, much of the
current research has been done with professional samples that are assumed to have a greater
degree of autonomy and flexibility than less professional workers (Konrad & Mangel, 2000).
What this approach loses in generalizability it gains in maximizing variance as the greater degree
of autonomy in professional jobs creates a weaker situation. In addition, selecting an
organization in a knowledge-based industry (such as technology) that has a less hierarchical
structure than more traditional industries should also contribute to weaker situations and thus
greater individual and group variance. These weaker situations, as opposed to the more
confined, rule based strong situations associated by rigid rules and hierarchies should allow for
more variance among teams and individuals in order to test the hypotheses.
My particular research site, Genericorp (pseudonym), is a large, multi-national
technology based firm. From a work-life balance standpoint, this firm has multiple qualities that
recommend it for research. First, Genericorp is a well know and prestigious company that is
frequently cited as a Best Company to Work For (Fortune, 2011, 2010, 2009, 2008). Prior
research in organizational support for WLB suggests that even companies with espoused support
for work-life balance may have variation in ways family supportive initiatives are supported or
enacted throughout the organization (McCarthy, Darcy & Grady, 2010). Indeed, Genericorp is
61

also known for its relatively flat structure, where a lot of decision latitude about work design is
placed in the control of managers and work teams. In these more dispersed structures,
organization espoused values have a greater chance of differential interpretation (Pocock, 2005).
Finally, Genericorp is characterized by many different job types that range from more
entry level positions to those that require extensive schooling and experience. These job
descriptions are often vague and non-specific, and allow for considerable customization at the
individual level. This type of situation should allow for individuals the greatest possible latitude
in enacted personal preferences for work-life balance while also fostering a diverse array of team
contexts that may be more or less supportive of individual customization.

62

CHAPTER 2: STUDY 1

Methods Study 1
Study 1 examined several of the proposed hypotheses using archival data available from
the annual employee attitudes survey. Studying these relevant variables within the context of an
annual employee survey is important from an organizational standpoint. Although annual
surveys can be psychometrically hindered by survey and item length constraints, studying the
ways workers as a whole experience work-life balance, performance and retention on an annual
basis can provide valuable information to organizations who seek to more fully understand the
work lives of their employees. This can help the organization focus on the areas that show
problems or that are of most importance when making action plans based on survey results.
This study focused on using archival data to examine whether or not work-life balance
satisfaction can be explained by performance ratings, self reported workload overload and how
factors such as manager WLB can affect the relationship of individual balance with retention
intentions. Although the measures are limited with regard to construct validity and the ability to
make statistical conclusions based on the measures (generally due to the small number of items
for each scale), the utility of this analysis lies in its potential to pinpoint problems areas in a way
that can be scaled to an annual survey and analysis. An additional benefit of using a sample
from the archived employee survey is that it allows for a powerful team analysis. Unlike field
research where gaining adequate response rates from teams and managers can have adverse
effects on statistical power (Raudenbush & Liu, 2000), through using archival data it is possible
to select a sample of teams with sufficient power to adequately analyze the results.

63

Procedure
Data was gathered from an archival database from Genericorp. This data was collected as part of
the annual employee survey distributed to all employees in the fall of 2010. A subset of an
archival database was randomly selected based on the following strata of interest.
Location. Although Genericorp is a large, global company with offices in a variety of
countries, due to potential geographical effects in work-life research (e.g. Stavrou & Kilaniotis,
2010), the location was held constant for the selected subsample. Only employees where the
majority of the team is located in North America were selected for inclusion in Study 1. Teams
were not eliminated if one or two members were located in a region outside of North America,
but the majority of the team members had to be situated in the North American region for the
team to be selected.
Function. The structure of Genericorp is divided into three, roughly equally sized
functions. These functions are engineering, sales and general and administrative. Research has
shown that both work-life balance issues as well as job flexibility are influenced by the type of
job an individual holds (Hornun, Rousseau & Glaser, 2009). In order to get a good
representation across function to use as a control variable, the manager headed teams will be
equally sampled from across the functions. Teams were selected on a roughly equal basis
randomly from each of the functions. In sum, there were 41 teams in the Engineering function,
44 teams in the Sales function, and 37 teams in the General and Administrative function.
Manager Status and Direct Reports. In research on teams, groups of three or more are
usually considered sufficient for investigating team level phenomena (e.g. Hollenbeck, Ilgen,
Tuttle, & Sego, 1995). In a review of team field research, Rassmussen and Jeppesen (2006)
found that 20% of the studies had teams of 3-8 members (as opposed to 9-14 (73%) or 15 or
64

more (9%)) but that there were no consistent differences in the variety of psychological factors
the authors examined based on team size. For Study 1, 122 teams of 4 to 10 members were
selected from the archival data set, and only those with 60% or more of team members reporting
were included in analysis. Teams were defined as groups of 4 or more employees who reported
to the same manager. In addition, all teams had a single manager who provided survey
responses as well. Please see Table 2 for a break down of team size and average response rate to
the survey per team.
Gender and Age. The main goal of Study 1 was to maximize the number of intact teams
with high proportions of team members reporting to ensure adequate power to test the multi-level
hypotheses, but to meet this goal and respect the privacy concerns of the cooperating
organization, the demographic variables of gender and age had to be abandoned in Study 1.
Because the employees had not consented to allow this data to be made public since it
was archival, my contacts at the company did not feel comfortable disclosing this information
although they did feel comfortable allowing me to gather this information in the supplementary
quantitative study to be described later due to the fact that employees had a chance to explicitly
grant consent. For Study 1, I was unable to collect gender or age data. These variables were
collected in Study 3, however, so their effects in similar contexts can be discussed.
Sample Description. The archival sample consisted of 122 teams. Each team had
precisely one manager who provided survey responses. The final sample was made up of 122
managers and 688 non managers, for a total sample size of 810.

Demographic variables

assessed include function, tenure, manager status and job level. Please see Table 3 for a
summary of individual sample characteristics.

65

Other demographic variables that are routinely studied in work-life research were
unavailable, such as gender, age, race, marital and parental status. Several of these variables will
be available in Study 3 and will be discussed then.
Method and Measures
Because space is at a premium for annual surveys, scale length was shortened to
maximize the types of questions that can be assessed each year. Because of this, some constructs
are represented by single items. Many of the relationships investigated in Study 1 will be
replicated in Study 3 using more robust scales to attempt to get around the some of the
drawbacks of using impoverished scales. However, although single item measures are not ideal
in research due to issues that may arise due to validity and reliability as well as to issues that
result from restricted variation between people, some researchers suggest that single item
measures may be appropriate to use with the construct of interest represents a single,
unidimensional construct, is clear to the participants and is a narrow construct (Sackett & Larson,
1990; Wanous & Hudy, 2001). Please see Appendix A for the scales used in Study 1.
Workload. Workload was measured in a binary manner. Respondents were asked to
respond to the question “Which of the following are the biggest barriers to your productivity?
Please select up to 3 from the list below”. Respondents could select from a dropdown menus
that listed 14 possible responses including one having to do with workload. Participants were
coded as 1 if they listed workload as one of the barriers to getting things done and were coded as
0 if they did not list workload as a barrier. The average score on this item was .40, meaning 40%
of respondents indicate that workload is a barrier to getting things done.
Performance. Performance data were obtained from Genericorp’s human resources
department. At Genericorp, performance is directly tied to incentive based compensation, and
66

due to this the company conducts performance reviews on a quarterly basis. Individuals are
rated on a scale with 1 being exceptionally below average and 5 being exceptionally above
average. Scores are given in increments of 0.1 and are determined based on performance on
goals set by the employee along with input from his/her manager at the beginning of each
quarter. The performance rating process includes a horizontal component as well that includes
peer input. Employees and their managers together determine 2-5 peers most appropriate to
provide feedback for the target employee. Peer feedback is provided in addition to manager
feedback on performance. Ratings from the mangers and peers are then calibrated by Human
Resources personnel, taking into account all of the evaluation recommendations made by
managers in the business unit, and each individual is assigned a rating. To obtain the most
appropriate performance rating, the performance ratings for the three quarters of the year that fell
before the survey was administered (Quarter 1, 2, and 3) were averaged for each respondent.
Average performance rating was 3.45, with a standard deviation of .25.
Work-life Balance Satisfaction (Individual). Work-life balance satisfaction was
assessed through three items that were rated on a 1 to 5 agreement scale, with 1 being Strongly
disagree and 5 being Strongly agree. The items were “I am satisfied with the balance between
my work-life and my personal life”, “I am able to detach from work during non-work time”, and
“My manager supports my efforts to balance my work and personal life.” Cronbach’s alpha for
this scale was .78.
Work-life Balance Satisfaction (Manager). The work-life balance scale score of the
manager of each team member was inputted into a column for each team member to be tested in
later hierarchical linear models.

67

Retention. Retention was assessed through four items that were rated on the same 1 to 5
agreement scale as mentioned above. A sample item is “At the present time, I am not seriously
considering leaving Genericorp”. Cronbach’s alpha for this scale was .83.

Results Study 1
Descriptive Statistics
Functional contrasts. This data set was made up of participants from three major
occupational functional groups: engineering, sales and general and administrative. Two oneway, between subjects ANOVAs were conducted to compare the effects of function on the
variables of work-life balance and retention, with Bonferroni post hoc test of differences. There
were no significant differences between the functions in retention scores [F(2,822)= .34, n.s.],
with the mean for Engineering being 3.34, Sales being 3.70 and G&A being 3.78.
For scores on WLB, the function General and Administrative scored significantly lower
than the function Engineering [F(2, 822)=3.83, p<.05]. The variable of function was
dichotomized in any subsequent analyses , with 1 representing the General and Admin function,
and 0 representing the Sales and Engineering functions. This was done as the difference between
Engineering and Sales was negligible relative to the difference between General and Admin and
the other functions. Please see Table 4 for the results of the ANOVA testing the effects of
function on work-life balance.
Correlations. The correlations between the nine variables examined in this study are
presented in Table 5 and the means and standard deviations are presented in Table 6. There do
not appear to be any unduly worrying trends in the correlation matrix. The outcomes of
68

interest—retention intentions, individual work-life balance and manager work-life balance, all
appear to have adequate standard deviations. The means are slightly high (all between 3.5 and
3.75 on a 5 point scale) but not so high that a ceiling effect would restrict the range to a
debilitating degree. Only one variable, performance rating, might present some psychometric
difficulties. The standard deviation for this item is quite limited which suggests that
performance ratings may be range restricted and may limit the findings that will be possible with
regard to this variable.
Control Variables. When testing the hypothesized relationships, it is important to take
into account variables that may affect the variables of interest in addition to those being
investigated, and thus control for their potential effects. The correlation matrix (see Table 5)
reveals that team size and function have a relationship to the work-life balance variables, both for
the individual and for the manager work-life balance scores. These scores indicate that people
on larger teams experience greater balance (as do their managers) and people in the functions of
Sales and Engineering experiences better balance than those in General and Administrative.
Interestingly, tenure and job level are not strongly related to work-life balance. Originally, job
level was to be used as a proxy for age in the control variables. Age is a variable often used in
work-life balance research (Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005) that I was
unable to collect for this data set, but given that job level is unrelated to the work-life balance
variables, it is unnecessary to use it as control, and it cannot serve as a proxy. For hypotheses
relating to work-life balance as an outcome, the control variables of team size and function will
be utilized.
Retention is related to different demographic variables. As evident from the ANOVA
performed above, retention intentions have no relationship to function. The correlation matrix
69

also reveals that they have no relationship to team size. Instead, retention intentions are related
to job tenure and, less strongly, to job level, with people showing decreasing intentions to stay
with the company the longer they have been there and the higher positions they have obtained.
For hypotheses that address retention, tenure will be used as a control. Only tenure will be
controlled for considering tenure and job level are correlated with each other and controlling for
both could result in problems with multicollinearity. Additionally, tenure may be logically
related to retention in that new employees are less likely to be for different career opportunities
and therefore may show artificially inflated retention intentions. Some research indicates that for
highly technical and/or specialized employees (such as those employed by Genericorp),
employees may be more likely to jump between companies for promotion and/or career
development opportunities (Barley & Kunda, 2004) This suggests that longer tenure at a single
company may be related to increased external job searching, thus linking tenure to retention. As
this link is irrelevant to the current study, the relationship of tenure and retention will be
controlled.
Although team size is also related to retention, these variables may share real variance
with the predictors of interest (e.g. small teams may have larger workloads per member). Spector
and Brannick (2011) suggest that controlling for variables that share real variance with variables
of interest artificially limits statistical tests and is inappropriate. Therefore, team size will not be
controlled for in further analyses.
Centering Variables. For both moderated regression and hierarchical linear modeling
(HLM) it is necessary to center, or standardize certain variables if the scales do not have a
meaningful 0 point. To grand mean center a variable, the grand mean of that variable is
subtracted from each individual’s score so that overall mean of the standardized variable is now
70

0. For both types of analysis, centering reduces the multicollinearity between the variables of
interest. In moderated regression, it reduces the multicollinearity between the interaction term
and the main effect of those variables (Baron & Kenny, 1986). For HLM, variables need to be
standardized in order to avoid remove multicollinearity between the random intercepts and
slopes as well as between first and second level variables (Krefts and deLeeuw, 1998). In HLM,
the researcher can choose to either grand mean center or group mean center. Grand mean
centering centers the variables around the overall mean or group mean center. Group mean
centering, on the other hand, calculates the mean for each group based on level two grouping
variables and subtracts the appropriate group mean from the members’ scores that fall within
each particular group (Krefts and deLeeuw, 1998). Because Hypothesis 25 states that manager
work-life balance (a level 2 variable) will moderate individual workload (a level 1 variable) to
predict individual retention, grand mean centering is most appropriate for the analyses in this
paper and in all HLM models the predictors have been grand mean centered.

Hypothesis Testing
A Note on Statistical Models. Because parts of Study 1 take place at two levels (the individual
and the work group) some analyses were conducted through hierarchical linear modeling (HLM)
while others were conducted through linear regression. Hierarchical linear modeling is a
statistical method that accounts for the fact that belonging to a group may make individuals more
alike to each other than random individuals that may not belong to the same group (Davison et
al, 2002). HLM is the most appropriate statistical method to use when studying data that is
nested, as it attenuates the effect group clustering can have on results and decrease the likelihood
of Type 1 errors (Hoffman, 1997).
71

When dealing with data that is nested within in groups, the first step is to test whether the
dependencies created by the fact that people’s work experiences are similar due to the fact that
they interact with each other can affect the error rate of standard regression (Hoffman & Griffin,
2000). To test this, it is necessary to look at whether the degree of variability between groups is
sufficiently greater than the variability within groups such that a regression model is not biased
to find a result when similarities in responses may be due not to real similarities in the variables
of interest, but rather to similarities due to group membership (Hoffman & Griffin, 2005). In
order to assess the “groupiness” of the variables, an unconditional means model in a special form
of regression, hierarchical linear modeling (HLM), is conducted. This is an empty model with
only the variable of interest estimated with group membership included as a nesting factor to
assess the relative impact of between and within variability. From the unconditional means
model, the interclass correlation (ICC) can be computed. Although there are not hard and fast
agreed upon standards of appropriate ICC(1) values to test for “groupiness” (LeBreton & Senter,
2008), general rules of thumb suggest ICC(1) of .08 to .2 are sufficient to indicate that the group
is a factor and needs to be addressed. For this study, retention and work-life balance were the
two variables that needed to be assessed for group agreement. See Table 7 for the unconditional
means models. Another use of HLM is to justify aggregation of individual level responses to
create a group level variable—this process will be discussed in more detail in Study 3.
Work-life balance shows a sufficient ICC(1) value (.16) to indicate that this can be
considered a group variable, while retention does not. In addition, the Wald statistic show
significance for work-life balance but not for retention, further supporting treating work-life
balance as a clustered variable but not retention. That means that members of the same group
tend to have similar work-life balance satisfaction but different retention intentions. Because
72

work-life balance satisfaction is affected by the nested effect of “groupiness”, all further analyses
with work-life balance need to be done with HLM. Because retention does not show group level
effects, it can be analyzed using hierarchical linear regression which does not take into account
group nesting.
Due to the limited scale availability in the archival data set, only a subset of the proposed
hypotheses were testable in Study 1. See Table 1 for a complete summary of all hypotheses
tested across studies including whether they were supported or not. The following section will
review the hypotheses for support in a sequential manner. For a summary of supported
hypotheses across all three studies, please see Table 1.
Hypothesis 1 stated that workload would be negatively related to WLB such that those
with high workload would report lower satisfaction with balance. In this study, workload was
measured using a dichotomous variable where participants could choose workload as something
that stood in their way of getting things done. To test this relationship, this dichotomous variable
was entered into the basic HLM model containing work-life balance first as a fixed effect and
then as a fixed and as a random effect. See Table 8 for four models. Model 1 illustrates the
empty unconditional means model that shows that WLB meets the standards of ICC(1) values for
group agreement (e.g. greater than .08, Lebreton and Senter, 2008). Model 2 and Model 3 show
the standard model without workload as a predictor with just the control variables included with
WLB considered as both a fixed (Model 2) and random (Model 3) effect. Chi-square difference
tests between Models 2 and 1 and 3 and 1 are non significant (Model 1 and 2: Chi2(1, N = 809) =
0.76, n.s; Model 1 and 3: Chi 2(2, N = 809) = 0.76, n.s). There are no differences in these models
which indicates either fixed or random models can be used. The final model (Model 4) shows
workload included in the model. Including workload decreased the values of Deviance, AIC and
73

BIC, which indicates a better model fit, and a chi squared test of the change in model fit, as
based on the change in Deviance scores, (Chi 2(3, N = 809) = 0.76, p <.01) shows that fit did
indeed increase from the unconditional means model. Thus, Hypothesis 1 was supported.

Hypothesis 2 stated that workload and performance would be negatively related such that
those with high workloads would have lower performance. To test this hypothesis, a one-way
ANOVA was conducted to compare whether or not the performance scores of people who
indicated workload was a barrier to getting things done differed from those who did not indicate
workload as barrier[F(1, 738) = 9.24, p<.01]. The finding was significant, however, it was in the
opposite direction than was hypothesized. People who feel workload is a barrier receive
significantly higher performance ratings than those who do not cite workload as a barrier to
getting things done. See Table 9 for a summary of the mean differences. Although the mean
differences are not large in performance scores (.06), it is important to note that small differences
in scores can have a large impact on employees, as performance scores are tied to bonuses,
promotions and transfer decisions within Genericorp. Hypothesis 2 was not supported.
Hypothesis 3 stated that performance and work-life balance would negatively related
such that who with higher performance scores would indicated lower satisfaction with balance.
To test this hypothesis, the HLM models were conducted so that model fit could be compared.
See Table 10 for model comparison of the unconditional means model for WLB, the model with
performance as just a fixed effect, and the model with performance as both a fixed and random
2

effect. A Chi Squared test (Chi (1, N = 740) = 185.78, p<.05) on the difference between the two
models (between Model 1 and Model 2) shows that Model 2 is a better fit. There is no difference
74

2

between Model 2 and Model 3 (Chi (1, N = 740) = 1.3, n.s ). However, internal statistics to the
HLM model, such as the Wald test of significance of random effects and t-tests of significance of
fixed effects, show that performance does not meet the threshold of .05 significance.
Researchers often advise interpreting model fit statistics over these internal statistics due to their
inconsistency (LeBreton & Senter, 2008), however these findings, coupled with the relatively
small (albeit significant) change in model fit significance, suggests that the relationship between
performance and work-life balance might be slight enough to be of little practical significance.
Hypothesis 3 is supported, with the caveat that it be interpreted with effect size in mind.
Hypothesis 4 stated that workload and retention intentions would be negatively related
such that those with higher workload would be less likely to indicate an intention to stay with the
company. To test this, a one-way ANOVA was conducted to compare the differences in mean
scores between the groups that either indicated that workload was a barrier to getting things done
versus those that did not indicate that workload was a barrier to getting things done [F(1, 807) =
5.44, p<.05]. See Table 11 for a summary of the mean differences. Because the test is significant,
this may indicate that those who cite workload as a barrier to getting things done show lower
intentions to remain with the company that those who do not list workload as barrier. Thus,
Hypothesis 4 was supported, although the effect size is small. However, when considering
retention behaviors, even small effect sizes can have a large impact on organizations due to
increased costs that accompany employee work product loss and employee replacement costs.
Hypothesis 5 stated that performance and retention intentions will be positive related
such that those with higher performance ratings will show greater intention to stay with the
company. See Table 12 for the regression analysis (Step 2 indicates the regression test for
75

Hypothesis 5). After controlling for tenure, performance did not result in significant additional
prediction of retention. Therefore, Hypothesis 5 was not supported.
Hypothesis 6 stated that work-life balance would be positively related to retention
intentions such that those with high work-life balance would report greater intentions to stay with
the company. Four HLM models were tested to investigate these relationships and subsequent
model fit was investigated. Please see Table 12 for a summary of this analysis. Model 2 is the
2

best fit of the three non-null models (Model 3 and Model 2: Chi (1, N = 809) = 150, p<.05)
which indicates that work-life balance scores are related to retention intentions such that those
with lower satisfaction with balance show higher intentions of leaving the company. Thus
Hypothesis 6 was supported.
Hypothesis 7 stated that work-life balance would moderate the relationship of
performance and retention such that those with higher work-life balance and performance would
be more likely to indicate intentions to stay with the company. See Table 12 for a summary of
this analysis. The interaction term did not explain additional variance in retention as the term was
not significant and Model 4 did not fit the data significantly better than Model 2 (Model 2 had
2

higher model fit statistics than Model 4, Chi (1, N = 809) = 180, p<.05). Therefore, Hypothesis
7 was not supported.
Hypotheses 8 through 24 incorporated variables that were not assessed in the archival
data and will be addressed later in Study 3.
Hypothesis 25 stated that manager work-life balance would moderate the relationship
between subordinate workload and subordinate work-life balance such that those with more
balanced managers will be more balanced, regardless of workload. Manager work-life balance is
76

a level 2 variable as individuals are nested within managers such that to test the hypothesized
relationship each individual team member’s work-life balance score was linked to his/her
manager’s score. Therefore, each member of the same team was linked to the same manager
work-life balance score. To statistically test this, 4 models were compared. Model 1 was the
unconditional means model for individual work-life balance. Model 2 incorporated individual
workload as a predictor of individual WLB as fixed effect (note: workload was not included as a
random effect due to the small effect this inclusion had in the previous models presented in Table
8 above). Model 3 incorporated the level 2 variable of Manager WLB as a direct effect predictor
in addition to workload. And Model 4 added the interaction term of manager WLB and
individual level workload. See Table 13 for a summary of these models. A Chi square difference
test reveals that the only significant difference between the models occurs between Models 1 and
2

2 (Model 1 and Model 2: Chi (1, N = 688) = 426.95, p<.05, Models 2, 3 and 4 are not
significantly different from one another). suggesting the Manager WLB and the interaction of
Manager WLB and individual workload do not increase model fit, and therefore do not add
explanatory power to individual level work-life balance. Therefore, Hypothesis 25 was not
supported.
Hypotheses 26 and 28 included variables not assessed in Study 1. Hypothesis 27 stated
that team WLB would moderate the relationship between individual workload and WLB.
However, since team WLB is an emergent variable stemming in part from individual WLB
scores, it is impossible to test this hypothesis using this data set.

77

Discussion Study 1
Summary of Results
Study 1 offered relatively robust support for Hypotheses 1 and 6, which link workload to
work-life balance and work-life balance to retention intentions. These findings indicate that
those who view workload as a barrier to getting things done also tend to have lower work-life
balance and those with lower work-life balance also, in turn, tend to have lower intentions to stay
with the company. These findings line up both with prior findings in the work-life literature (e.g.
Binneweis, Sonnentag & Mojza, 2009) as well as with logical reasoning about time constraints
and work satisfaction. Individuals with high workloads are more likely to experience lower
satisfaction with balance due, at least in part, to time constraints and conflicts that arise from
working more hours to deal with large workloads (Ilies et al, 2007). Those with lower work-life
balance may also show a decreased desire to stay with the company in an effort to help rebalance
their work and nonwork lives through a different position or through general job dissatisfaction
stemming from imbalance (Deery, 2008).
Other significant relationships also emerged from the results, although the effect sizes are
small and should be interpreted with these in mind. Workload as a barrier to getting things done
has a small negative relationship to retention intentions (Hypothesis 4), suggesting that
employees who are less satisfied with their workloads show an increased tendency to show
intentions to leave the company. Even though the small effect size is small, small effect sizes are
important to organizations if it means that their higher performers are more likely to leave the
company. As workload is often related to higher performance, this could be a key relationship in
help explaining why higher performers are choosing to leave the company. It is also important to

78

point out that the measure used to assess workload in this instance is impoverished, as it is a
single item and not ideally worded. This question will be more fully investigated in Study 3.
The hypotheses relating to performance also showed only small relationships that, while
statistically significant, may have limited practical implications. Hypothesis 2, which predicted
workload would be negatively related to performance, was not supported and performance was
actually found to be related positively to this variable. This could be due to the possibility that
people with high workloads could be accomplishing more and thus be earning higher
performance ratings than those who do not report high workloads. Although the literature
suggests that an inflection point would be reached in which burdensome workloads would
overwhelm the employee at which point performance would suffer (Janssen, 2001), it is possible
that the inflection point has not occurred with this sample and that those with heavy workloads
are simply working more and that more work is translating into higher performance scores.
Another possibility is that the performance scores at Genericorp are subjective interpretations of
performance as filtered through supervisor discussions and it is possible that these discussions
use work hours and workload as a proxy to determine performance. Although the relationship is
small, the performance rating variable is so range restricted and constrained that even small
effects can have real world implications. This is particularly true when considering that real
world decisions such as promotion, bonuses and other compensation based decisions are based
on performance scores where small increments in score make real differences in decision
making.
The second supported hypothesis that has to do performance is Hypothesis 3 and
indicates that performance is negatively related to work-life balance. This relationship is
negative such that those with higher performance ratings report lower satisfaction with work-life
79

balance. Although stress and strain research suggest the performance will ultimately suffer if
individuals spend a long time in a state of work-life balance dissatisfaction (Hobson, Delunas
and Kecic, 2001), this relationship has not been reliably demonstrated in the literature. Past
research in this area has been equivocal, with those with low satisfaction with balance linking
inconsistently with performance (Beauregard & Henry, 2008), perhaps due to individual
differences in preferences for role involvement. Again, this effect size is quite small, so, although
statistically significant, may not represent much practical distinction and should be interpreted
with caution.
Hypothesis 7, the hypothesis testing the relationship of performance to retention was not
supported. This is not particularly surprising as performance likely has a complicated
relationship to retention intentions. High performers can be both more and less likely to leave a
company. For a company that has reward and advancement systems that are tied to performance
metrics (as does Genericorp), high performers are often very satisfied with both their work and
career opportunities and this can manifest itself in increased organizational commitment and
intent to stay with the company (Guest, 1997). However, high performers are also more likely to
have increased job mobility options as their skills are in high demand from competing
organizations (Garger, 1999; McEvoy and Cascio, 1987). Thus, high performers can feel
pressure to both leave their current positions and stay at them, which makes finding systemic
relationships between high performers and retention intentions difficult.
The two hypotheses that tested interactions, Hypothesis 7 and 25 were not supported.
Hypothesis 7 suggested that performance would interact with work-life balance to predict
retention intentions. This was not the case. An explanation for this null finding in addition to the
complicated retention/performance relationship discussed above is that the performance scores
80

for this sample are relatively range restricted, with the majority of the organization falling within
a narrow band. This leaves relatively little variance with which to find interrelationships.
The second interaction hypothesis predicted that manager WLB would interact with
individual workload to predict individual workload. This hypothesis was not supported. One of
the reasons for this may be that as level increases, work-life balance scores decrease as indicated
by the negative correlation between job level and work-life balance. This is likely due to
increased responsibilities and work demands that come with higher levels. The level/balance link
may be strong enough to overcome any team norm linkages that may arise from manager WLB
scores affecting their subordinate teams.
Strengths and Limitations of Study 1
Study 1 had some unique strengths and limitations that stem from the source of the data
itself. Because it was based on archival company survey data, one of the strengths was that the
response rate was high and it was possible to sample mostly intact work groups. This allowed for
an investigation of work-life balance and retention at the group level in a way that is often very
difficult to do in a field setting. At Genericorp, there is unusually high company buy-in for the
annual survey, which ensures a response rate that allows for team level analysis.
However, because archival data was used, there was diminished control over both what
types of questions could be asked and what kind of background and demographic information
could be gathered. Due to the agreement with Genericorp, only a limited set of demographic
information could be connected to this data, and it excluded gender and age variables.
Additionally, this study was constrained to the items asked in the annual survey, and had to use
post hoc measures of constructs such as workload and balance that do not completely fit or fully
capture the entire construct space.
81

Study 1 attempted to maximize team representation at the cost of robust measures and
measurement control. Studies 2 and 3 compensate for the measurement deficiencies of Study 1 to
get a more comprehensive understanding of work-life balance and boundary management at
Genericorp.

82

CHAPTER 3: STUDY 2
Study 2: A Qualitative Investigation of Work-Family Decision Making
Study 1 furthered the literature through examination of infrequently examined constructs
(namely retention and performance) but it did so using a very frequently used research paradigm,
the survey. While the use of the cross sectional survey is common and useful from a practical
standpoint, survey research can often treat phenomena such as work-life balance as stable within
an individual and may not be the most appropriate method for capturing more nuanced
characteristics of the phenomena (Hinkin, 1998). The current measures of balance satisfaction
and work-life correlates often focus on this type of general understanding rather than seeking to
figure out how these experiences of aggregate satisfaction of balance come to occur (Matthews,
Barnes-Farrell & Bulger, 2010). Study 1 is able to reveal some of the relationships between
variables, but Study 2 will be able to add depth of meaning to these relationships and help shed
light as to why these relationships exist and how they might form and evolve. Although the
dynamic relationship of resource provisions and depletions is explicitly outlined in work-life
balance research, few researchers have directly assessed what contextual and personal factors
individuals see as especially important in contributing to satisfaction with WLB.
The purpose of Study 2 is to more closely examine the roles managers, coworkers and
individuals play in creating and managing satisfaction with balance. These interviews will be
focused to help encourage participants to reflect on their perceptions of agency in making worklife balance decisions and how this agency if affected by the context each person finds
themselves in. In addition to the structured questions, a subgoal of the interviews is to allow the
participants to contribute insights into their own experiences that may not be captured by the
work-life literature. For this purpose, interviewees will be encouraged to add whatever they feel
83

is relevant to their own experience of work and life balance. The goal of Study 2 is to help
develop a better understanding of how employees feel more or less empowered to take control of
their own work and non-work lives.
Procedure.
Interview Format and Focus. Individual interviews were conducted between the
interviewer and a single employee at a time. One on one interviews are the best mode of
research as larger focus groups might inhibit individuals in discussing personal challenges with
balance or the specific details of negotiations and i-deals that may have been made.
Individual interviews were conducted by the primary researcher on-site with employees
from the main office of Genericorp. The interviews consisted of a series of semi-structured open
ended questions that specifically targeted individual WLB, perceptions of manager and group
support, perceptions of barriers to achieved satisfactory work-life balance and the utilization of
flexible arrangements. The interview focus questions were based on border and idiosyncratic
deal theory, as well as research done on manager and organizational support. The goal of the
questions was to more deeply probe individual experiences in creating and maintaining his/her
own satisfaction with balance and what s/he see as potentially helpful and hurtful to this balance
within the context of their work group. During the interview, interviewees often directed the
discussion to areas that were of particular interest to them, so the interviewer occasionally
diverged from the semi-structured interview script to follow the topics the interviewee wished to
talk about. Please see Appendix B for a copy of the semi-structured interview questions
The semi-structured interview is formatted in such a way as to probe at the areas of
decision making that are most theoretically relevant to work-life decision making within the
context of COR theory and border theory frameworks. The questions are focused on both
84

tangible barriers an individual may experience (such as workload) and intangible barriers (such
as norms for working long hours). Manager and team support are also specifically addressed.
These questions are designed to try to cover a range of experiences workers may have so that
each person’s experience can be thoroughly explored during the interview.
The structure of the interview was arranged in such a way as to encourage participants to
feel comfortable and to open up to the interviewer. It consisted of four distinct parts that I, as the
interviewer followed for each interview as a best practice (Krueger, 1994; Krueger & Casey,
2000), namely the Introductory phase, the Biography phase, the Structured Interview phase and
the Closing phase. In the Introductory phase, I briefly introduced myself, explained the purpose
of the interview and asked the participants their first names. At this point, I had the interviewee
provide a brief description of his/her work and home schedule of a typical and/or a busy work
day as an opening question. This was meant to get them thinking about a typical instance of
how they balance work and life before we get into discussing more abstract concepts and is
meant to create a factual base for the more opinion based dialogue to follow (Krueger, 1994).
Following the Introductory phase, I transitioned to the Biography phase where I asked
participants to tell me in their own words about their work and home lives and whether they felt
satisfied with their balance. This phase was meant to get interviewees firmly present in the
interview and thinking about the work-life issues of interest in the study.
After the Biography phase, I introduced the Structured Interview phase which represented
the key questions I am interested in. These key questions were focused on trying to really
understand what drives individual decisions about work-life balance and what role her/his team
and manager may play in helping or hindering their efforts to create their ideal balance.

85

The interviews concluded with the Closing phase in which I asked the participants if
there was anything they wished to add to the interview or if they had any further questions for
me.
Prior to conducting the one on one interviews, the interview questions were discussed and
revised with a group of 4 Genericorp volunteers. Feedback from this group was used to fine tune
and enhance the questions that were asked during interviews.
Participant Recruitment. 41 employees were identified to participate. In order to get
good construct coverage as well as representation across the company, participants were
randomly selected from a variety of demographic and construct relevant strata. Using data from
the previous year’s employee survey, roughly half the approached sample scored low (in the
lowest quartile) and the other half scored high (in the highest quartile) on the work-life balance
satisfaction item from the survey. The sample was also stratified on manager status, with about
40% of the approached sample being managers. The sample was also distributed by function
with roughly 1/3 falling into each of the three main functions of Engineering, Sales and General
and Administrative. The interviewer was blind as to what level of WLB the interview
participant had scored on the previous year’s employee survey so as not to bias the interview
process.
Participants were approached via email from an internal executive in the HR department
of Genericorp, who asked each of participants to participate in 30 minute interview on work-life
balance. Interviews were set up via an internal electronic calendar system and interviewees
merely accepted or declined the calendar invitation to indicate their desire or disinclination to be
interviewed. An incentive was provided, and each interview participant received a $15 gift
certificate to Amazon.com.
86

A final sample of 30 participants agreed and followed through to participate in the
interviews (a 73% response rate). The sample consisted of 20 men (66.7%) and 10 women
(33.3%), 12 managers (40%) and 18 non managers (60%), and had the following representations
from the occupational functions: Engineering: 9 (30%), Sales: 11 (36.7%), and General and
Administrative: 10 (33/3%). Interviews were audio recorded (with the consent of participants),
transcribed and then the audio recordings were destroyed.
Method
For the analysis of the qualitative data, I used both inductive and deductive methods of
content analysis. For the deductive plan, I used border and i-deal theory as guidance for themes
to look for. For border theory, I looked for instances where participants mention ways they
manage transitions from work to home and back again, the use of technology in these transitions,
and references to physical locations and times work and home issues are dealt with. For i-deal
theory, I coded for the ways that individuals explain their flexible (or non flexible) arrangements
to determine whether they are the result of implicit or explicit negotiation, what specific type of
flexible arrangement they may use, and what role manager/coworker support may play in the use
of flexible arrangements.
In addition to deductive coding, I enlisted the help of a fellow graduate student to aid in
inductive coding processes as recommended by Hinkin (1998). As this project is among the first
to explicitly examine the nuanced ways individuals experience their own work-life balance, it is
likely that topics will arise during the interviews that do not fit closely with the above deductive
categories. Inductive coding is based on grounded theory techniques outlined by Glaser and
Strauss (1967) that discuss ways to develop theory from qualitative data. However, unlike other
usages of grounded theory that rely on no theoretical framework, the inductive techniques used
87

in this paper still exist within an “orienting theoretical perspective” of COR and boundary theory
which can “guide researchers in what they should pay attention to but does not focus research so
narrowly as to exclude data whose importance may not be recognized at the outset of a project”
(p. 20; Locke, 2002). This type of blended, theoretically guided inductive research has been
used with success before (Kreiner, Hollensbe & Sheep, 2009; Ashforth, Kreiner, Clark & Fugate,
2007).
The inductive coding took place in a series of cyclical steps, in which the codes were
derived, then agreed upon, then more codes were created, then those codes agreed upon, etc. In
the first step, two coders, the principal investigator and graduate student unfamiliar with the
WLB literature read through each interview transcript independently looking for themes or
concepts. To ease the process of inductive coding, at this point the interviews were broken into
thematic chunks based on deductive reasoning outlined above, as well as the specific questions
that were asked during the interviews. During this process, we created new specific codes as
necessary to accommodate the data, subcategorized within the specific themes. We repeated this
iterative process for all the themes.
Once a code book of coding categories had been agreed upon by the principal
investigator and the graduate student, coding began in earnest. Two undergraduate research
assistants aided in assigning codes to the interview text. Two coders independently coded all the
interviews and then met to discuss code assignments. Discrepancies in coding were generally
minor and mostly due to oversight on the part of a coder or minor disagreement on code
interpretation. At coding agreement meetings, the coders discussed differences in coding and
reached mutually agreed upon conclusions. Coding agreement during these meetings converged

88

to 100%, although prior to meetings hovered between 70% and 80% depending on theme or
coding pair.

Results Study 2
At the onset of coding, the goal was to create more specific categories than may be
strictly necessary to catch potentially important nuances in the narratives. Once all the
interviews were coded, the coding categories were consolidated, so that thematically similar
categories were collapsed across individuals so as to create a simpler means of interpretation.
This consolidation was discussed with another expert in Genericorp and the decisions were
verified as logical, an approach that is recommended in qualitative research process (Seidman,
2012). Originally, 255 categories were defined and coded. After consolidation, 65 distinct
categories remained. All told, the 30 interviews yielded 1,267 unique coding points across 65
categories within 17 themes. See Table 14 for a summary of these codes and the number of
interviewees who mentioned them during their sessions. Also noted in Table 14 is whether this
theme was “Targeted” or “Emergent”. Targeted themes are those that were specifically probed in
the interview script (see Appendix A), while Emergent themes are those that arose naturally
through the course of the interviews and were not investigated with a priori questions.
About a third of the themes that emerged showed up in all or almost all of the interview
sessions (Satisfaction, Detachment, Manager, Workload, Flexibility and Team). The prevalence
of these specific themes is due to the fact that the interview protocol specifically called out these
areas for discussion. More interesting, perhaps, are the themes that arose organically during
discussion that were not part of the original interview protocol. Those that were mentioned

89

frequently include the Culture, Agency, Performance, and Career. Below, I will discuss the
findings for each of the 17 themes.

Satisfaction
The interview protocol specifically asked about each individual’s self perception of
satisfaction with work-life balance. Table 15 shows the categories that emerged within this
satisfaction category. The sampling methodology of participants for the interviews tried to get a
balance of people who were satisfied and dissatisfied with balance so that a full spectrum of
experiences would be represented in the interviews. Ultimately, the interviews ended up slightly
skewed toward those who were unsatisfied with balance, with 18 of 30 interviewees indicating
that they were dissatisfied or very dissatisfied with their current levels of work-life balance. The
remainder of interviewees (12) indicated that they felt good about their work-life balance.
When discussing satisfaction with balance, an interesting trend came up for about half of
those who indicated that they were not satisfied with their balance. Ten interviewees felt that
their balance was unsustainable in the long term, and several said that this meant they were
coming to a point where they needed to make some real changes to their work lives in order to
better accommodate their personal lives. While solutions for this varied (see attrition section),
there was a feeling that the current pace of work could not be sustained indefinitely for this
group.
The flip side of this is the eight individuals who reported dissatisfaction with balance but
felt that was sustainable to a certain extent. This is illustrated best in a quote from one
interviewee who was very dissatisfied with her balance but felt that she was not near “burn out”

90

because the work itself was so rewarding for her. This indicates that work can serve both a
replenishing as well as an exhausting role in the interviewees’ lives.

“I think from where I work and what we work on that we’re so into what do and who we
work with, like, the client, that no… [I’ll]be okay. It will because we really care about
what we do and enjoy what we do, and the people that we work with, and that’s huge. I
don’t think it’s burn out. And like I said, I think when it gets to a certain point, we have
to speak up. A lot of it is up to us as well.” –Participant 28

Detachment
The inability to detach from the office came up relatively frequently in the interviews.
See Table 16 for a summary of the categories within the Detachment theme. People tended to
talk about detachment—or, more accurately, the inability to detach-- in a couple of ways, mainly
having to do with physically detaching from the workplace vs. mentally detaching from the
workplace. On the physical side, most of the interviewees mentioned doing at least some work
during off work hours (26 interviewees, 87% of those asked) and most of the work required
using some sort of technology, such as a laptop or a mobile device, to log on to physically tether
them back to the workplace. A third of interviewees also mentioned that they have difficultly
mentally disengaging from work, such that when they are at home their minds are still churning
with work related topics. As one interviewee put it:

“I only get out an hour or two a weekend, I’ll often find myself thinking about work.
Political stuff tends to nag me a lot, so I’ll often find myself waking up in the middle of
91

the night worrying about political stuff. So, 50 hours of hard work, easily another 5 to
10, depending on the week, actively thinking about work.” – Interviewee 12

Another interesting aspect of physical detachment is that although the majority (26)
mention that they try to set clear boundaries between work and non-work, about half also say
that they like having the flexibility to switch between work and non-work tasks at their
discretion. Clearly there is some ambivalence within individuals about which schedule
arrangement they most prefer or enact. This may speak to a larger trend seen throughout the
interviews that will be discussed further in the “Agency” section. This was the tendency of the
interviewees to say they were actively seeking work-life arrangements that best suit their
individual preferences and work styles. This search is not always as successful as the individual
hopes, so this may explain why some of the interviewees like both separate and integrated worklife boundaries.
A micro trend showed up in the interviews and had to do with parents. Only one third of
interviewees (10) identify as being parents, but all of them mention a specific weekday schedule
arrangement that helps them balance work and home obligations. They all talked about
reserving time specifically for family in the evenings where they consciously do no work. After
the kids are asleep or dinner is eaten, these parents check back into work, checking email again
before bed or logging back onto the office for longer stretches of an hour or two to work in the
evenings.

92

“I leave 7 to 9 p.m. just strictly for anything home related. Now a days I’ve gotten into
the habit where I just close the computer at that time. And then, check up on the emails a
little bit in the evenings, do anything that needs to be done” – Interviewee 30

“Head back home and then just spend time with the kids. Then dinner, then dish wash,
this type of stuff, and then around 9:00 put the kids to bed. Then, log back in and stay
online for another two or three hours.” – Interviewee 7

Manager
Manager support was a theme that was specifically addressed in the interview script, so
all interviewees addressed it in some way. See Table 17 for a breakdown of the categories within
this theme. The majority of interviewees (20) responded that their managers were supportive of
maintaining a good work-life balance. People often spoke how their managers were generally
supportive of them across a variety of domains (e.g., career development, work interests, etc.)
and how the manager typically had the best interests of the employees at heart and did their best
to ensure that they were able to do their best work in a way they were most happy with. A
substantial minority of interviewees (10), however, did indicate that their managers were not
supportive of work-life balance efforts. A few interviewees (5) fell into both of these categories
as they compared and contrasted experiences with both supportive and unsupportive managers
they had had at Genericorp. Five interviewees did not categorize their managers into either
supportive or unsupportive categories because they had never discussed work-life balance issues
with them (“they never come up”) so felt they could not comment on them.

93

An interesting theme that ran through the interviews when people talked about their
managers is that more people viewed their managers as poor role models than good role models
for effective work-life balance (16 vs 10, respectively). Although some interviewees felt that
their managers were supportive of their subordinate’s work-life balance, several felt that the
managers themselves did not have the type of work-life balance that they would like to emulate.

“I know very little of my manager’s private life, but I do see him online at various times
of the day, so if … Yes, I think, people here become managers by putting in a lot of work,
and when they become managers they continue to put in a lot of work and they continue
to put in many hours, and yes I think they are a model for putting in a lot of work rather
than a model for separating work and life.” – Participant 18

“My manager is unfortunately not, maybe, the ideal case to study. He works more than
anybody I know, more than anybody on our team, more than anybody actually should.” –
Participant 27

When it comes to manager support, interviewees talk about two types of support. One
type revolves around communication and discussion and fits more closely with emotional type of
support. The other type of support is more agentic and involves the manager actively doing
something to adjust workload balances within the team to help ease the burden on individual
employees. Half of the interviewees (15) feel that their managers do not communicate well about
work-life balance, either that it doesn’t come up during regular conversations or that it would be
uncomfortable to talk about. The other half, however, do say that they show strong verbal
94

support for work-life balance and encourage active dialogue about balance issues within the
team.
When it comes to more proactive behaviors, however, such as reconfiguring workloads or
reprioritizing to reduce workload expectations, 9 interviewees feel that their managers had little
ability to actually make substantive changes or improvements in their workloads or balance. So
while managers may be doing a good job keeping communication lines open around work-life
balance issues, they may still face challenges when it comes to taking action to make changes
that could help solve the problem their subordinates face.

“Craziness, heroism, you know, setting impossible things and then achieving them
works. But they are no less impossible because of that. They are… and also, we never
meet the deadlines, we always slip. But we always… the deadline gets extended on the
last day before the deadline. Everybody’s racing to meet the date as if it’s possible and
then on the very last date you extend it. Yes, I think my manager supports work life
balance, I think… In practical situations, if I ask, can I work from home? Sure. Can I,
you know, can I… I was never denied vacation. I was never denied any time off I would
ask, any flexibility I would ask, it’s always been granted. There are definitely things to
like… There’s certainly a lot of flexibility, but there’s also a very, very high bar to meet,
very high expectations and those are impossible to meet without putting in a lot of hours,
I guess”. --Participant 18

Workload

95

Another topic that was covered in all interviews was workload. See Table 18 for a
summary of the categories covered in the Workload theme. Workload has a variety of aspects,
including volume of work, type of work and how work is shared among team members.
One of the most interesting findings looking across interviews is that people had very
different subjective experiences to objectively similar workloads. Where one interviewee
(Participant 30) mentioned working 55 hours a week and that workload being too much, another
mentioned working the same number of hours and that amount being not enough (Participant
11). However, this variation around subjective feeling about workload only really occurred in
the 45 to 55 hours per week workload range. For the interviewees who regularly worked greater
than 60 hours per week (n=3), they all indicated this workload was too much. This finding
would suggest that subjective feelings about workload are particularly important for work hours
that slightly exceed the “typical” 40 hour work week, but that at high amounts of hours worked
per week, subjective feelings about workload are likely less variable.
An observation about work structure that came up in the interviews is that most
interviewees (28) mentioned that their workload teds to ebb and flow, with some times of year
being busier than others. Depending on the type of work the interviewee did, these busy times
could be tied to specific calendar dates or project deadlines, but many mentioned that they
routinely dealt with discrete periods of greater workload where they expected their work-life
balance to suffer. When probed about whether the less busy times were used to consciously cut
back and “make up” for the over work of the busy time, two of the interviewees indicated that
although they tended to take things easier for those weeks, they were still working typically long
(40+) hour weeks during the off cycles and not officially flexing extra time working during busy
times into these periods.
96

When it comes to type of work, about two thirds of the interviewees (21) mentioned that
the volume of meetings and emails made it difficult to balance their workload. Others (15)
mentioned having to deal with last minute, reactive tasks as being very disruptive to their
workday as something else that pushed planned work into evenings and weekends. Specifically,
many felt they had to work in the evenings and/or weekends to catch up on the other work that
they didn’t get to during the work day. Working during evenings and weekends is typical
behaviors across interviewees, with three quarters saying they routinely at least check email and
often do more substantive work during these off hours. However, many interviewees mention
that checking in is not a bothersome activity to them and helps them feel on top of their
workload.

“This particular role that I’m doing now, it’s very much peaks and valleys of work. So
sometimes I have a lot to do, sometimes I do not have enough to do, so on average it’s
fine.” -- Participant 11

“If your day is so full of meetings then you’re like ridiculously backed up on email that
you can’t get to things. And then you either have to spend two hours at night, or at 5 o
clock, my meetings are done but I don’t want to go home with a hundred emails in my
inbox. So I’m then sitting there until 7 just making sure I’ve checked everything off for
the day. And those two hours, I mean, if I didn’t have so many meetings I should have
been able… Ideally, I shouldn’t have to sit there for two hours after 5, right?” –
Participant 20

97

Flexibility
Interviewees were all specifically asked about flexible work arrangements and whether
they had access to them or used them in addition to whether they were satisfied with their current
levels of flexibility. See Table 19 for a summary of the categories in the Flexibility Theme.
The most common form of flexible work arrangements mentioned was location
flexibility, and the majority of interviewees indicated that they had at least some degree of
location flexibility (23). This location flexibility had individual flavors, however. One
interviewee spoke about how he had a standing “work from home” day that he took every single
week. He highly valued this day as it increased his ability to spend time with this wife and young
child and reduced his commuting time. Another interviewee mentioned that he liked to spend
time visiting family in another state and his manager allowed him to work remotely during these
times. A more common trend was that people didn’t work from home on a regular basis, but that
they felt they had the option if they needed or wanted to, such as if they needed to meet a repair
person at their home or if they needed to be close to home for a doctor’s appointment.
An interesting trend was that many people really felt that their level of flexibility was
indicative of the “rule” for Genericorp as a whole, even though these levels differed rather
drastically from person to person. For example, here are two quotes that speak to this belief, even
though these two individuals have very different perceptions of what the rule is:

“Interviewer: Do you ever work from home?
Participant 4: No/ [Genericorp] has a pretty stringent policy on working from our office
locations.” –Participant 4

98

“So the schedule, you see, what I like about it here is the schedule is very flexible, nobody
really… you know from my team, if you ask, when does this person really come to the
office. I don’t know. I don’t care. As long as that person’s doing the work. Generally
that’s pretty common at [Genericorp]. So far virtually every company I’ve worked for
has offered flexibility like that [working from home]. Flexibility is not really unique at
[Genericorp], at a technology company that is pretty common.” – Participant 7

One interesting thing to come up in about half the interviews (16) is that some people felt
that flexibility is a double edged sword. On one hand they liked that they have the option to
work from wherever to meet personal life demands, but they felt like the portability of work
means that they are essentially “on call” to work all the time.
Others (11) spoke of the possibility that flexibility may not be good for productivity. A
few mentioned that it can make collaboration more difficult if half of the team is sporadically out
of the office working from home. Additionally, two interviewees mentioned that, as managers, it
can make it more difficult to make sure that all employees are staying on top of their work if they
are doing more of their work outside of the office.
Agency
One of the most interesting themes that arose organically from the interviews was the
idea that work-life balance was something that individuals had personal responsibility for or
control over. See Table 20 for a breakdown of topics in the Agency Category. Interviewees
discussed personal agency with regards to achieving balance in a variety of ways, but although
the interview script did not probe the issue of control specifically, it was a topic that arose
naturally in the majority (28) of interviews.
99

The most common way interviewees talked about their own control over balance was in
the ways they exerted intentional effort over trying to manage their work-life balance. Over two
thirds (22) mentioned being actively engaged in trying to effectively manage their work and
personal lives. More concerning perhaps, is the large proportion (60%) of interviewees who
identified balance as a personal goal, but felt that they, personally were “bad” at achieving it.

“I’m trying to make a concrete push to care less about work, as strange and as cold as
that sounds, but I think. I’ve been trying to do that, but I don’t think I’ve been doing that
very successfully. I’m sort of still in this sort of experimental phase of this process, I
don’t know how successful I’m being with it at the moment. But if I’m not being
successful, then I don’t know how much longer I can last. It seems like every day, it’s
very cyclical and very, almost, bipolar. Some days it’s like, okay, it’s not so bad, and I
actually felt like I had a decent day and got to do some fun stuff and other days it’s like,
wow, where’d the week go and this sucks.” –Participant 36

“But then you as an employee really need to just pick and choose and try to strike that
balance yourself, I think. And so, I’ve taken, I went and took a time management class
that they offered, that this company offered and I think it was okay, I mean it was good. I
mean, I get it, it was one of these things where I feel I need to improve. It wasn’t like…
and so, there is pressure in some ways. You know, I don’t know if I have the answer,
that’s the thing. It’s a very difficult… like it’s just not an easy thing. Because I think
that’s why a lot of [Genericorp employees] struggle with because it’s just not obvious
what the right solution is.” –Particpant 16
100

The theme of control continued in the degree of autonomy people felt over being able to
manage their work and personal life balance. A third of interviewees felt that their work was out
of their control, that they did not have the agency to reduce or redesign their workload in order to
help balance their lives in a way they would prefer. This clashed with some of their efforts to
take personal responsibility for balancing work and life since their feelings of control over at
least one part of the equation were very reduced. The group of interviewees that felt out of
control of their workload also tended to be the ones who wanted a better work-life balance but
were unable to achieve it and did not know where to go from there.
A small but important group of interviewees (5) mentioned an alternative view to worklife balance that emphasized agency and choice but de-emphasized balance per se. These
interviewees identified that they led imbalanced lives (in that they were tipped more fully
towards work than life) but that this was an intentional choice they were making. This choice
was generally driven by career goals, life stage (e.g. most interviewees in this group were young)
and the sense that such intense work focus was something they planned to do for a relatively
temporary amount of time to jump start their careers before settling down into a more balanced
pattern.

Team/ Culture
The interview script specifically probed how interviewees felt about their teams and the
culture of Genericorp. See Table 21 for the categories in this theme. When it comes to the local
team culture, two interesting macro topics emerged. First, most interviewees (25) were quick to
identify their teammates as a source of emotional and instrumental support, for both work-life
101

balance issues as well as for other issues they may be encountering in their work. Second,
although interviewees rarely said that they were in direct competition with their teammates, they
often mentioned (21) a fear or concern of “keeping up with” or “distinguishing themselves from”
the others on their team. These concerns often led the interviewees to feel like great performance
wasn’t good enough and they would have to work harder and longer hours just to stay in the
same place relative to their team. For some, this seemed to stem from a concern that they wanted
to avoid appearing incompetent (5 participants), while for others, this stemmed from an implicit
desire to prove their competence (6 interviewees). This is interesting thinking through the lens of
approach/avoidance orientation and how that may be related to concerns over performance and
how these concerns may drive behaviors and balance satisfaction (Porath & Bateman, 2006). So,
although it doesn’t appear that overt competition or negative cultures arise at the team level, the
strong performance pressure individuals seem to place on themselves when surrounded by other
high achieving individuals may be driving some of the tendency to overwork.
People were more likely to speak in less positive terms about the global culture of
Genericorp with regard to work-life balance. When asked about whether they thought
Genericorp supported WLB, a minority reported feeling very supported by the company (5),
while close to two thirds (17) mentioned that at Genericorp performance is the highest priority,
so work-life balance gets second shrift. Several interviewees were quick to point out that
performance should rightly be Genericorp’s number one priority, and this is the way
organizations are run and they did not judge the company or the leadership harshly for it.

“I think, they understand the concept and understand why it’s important. I think in
practice, it’s a difficult thing to do at [Genericorp]. And again, it comes back to the types
102

of personalities we hire and the notion that to have visibility and to be able to contribute
and to have the impact is hard to do within a 40 hour work week for most people. I
applaud the effort. I would be very impressed if they could figure out how to make it all
work. Because I think it just two opposing forces that are very difficult to wrangle in such
a way that works at [Genericorp]. Mostly, just because of the types of the hiring profile
of the people we hire, expectations, the… and by expectations I mean, we hire all these
superstars and we expect everyone to do well.” –Participant 39

“I don’t think it’s possible to slow down because our competitors will not slow down,
start ups will not slow down, if you slow down, you’ll die. So, it’s definitely… I don’t
think the pressure is ever going away, and there will always be this very… this industry
will never be for people that like to work 9 to 5.”—Participant 18

Performance
The performance theme is related to some of the themes that arose in the team and culture
theme above, mainly the degree to which pressure to perform is embedded in a social context in
which ones peers are also high achieving. See Table 22 for the categories in this theme. The
majority (24) of the interviewees specifically mentioned performance demands during the
interviews.
One topic that came up frequently (20 interviewees) was that in a field of high achieving
peers, distinguishing oneself through performance became even more difficult and that
expectations for performance were high. Many interviewees described this as a factor of the
“type” of people Genericorp hires, generally describing an ambitious, driven, high achieving
103

archetype used to being the best in the class or in a work group. A few mentioned that a
company full of these “Type A” personalities would inevitably make performance pressure an
ongoing and difficult-to-overcome challenge.
Related, many interviewees (two thirds) explicitly felt that the way they performance
management system is structured at Genericorp rewards employees for living lives that are more
tipped toward work than life. Additionally, as a portion of each employee’s pay is determined by
their individual performance, how one is able to demonstrate excellence on the job is tied to real
financial outcomes for the employee. Most of the interviewees stated that their peers were very
competent so in a field of very competent peers, the way to distinguish oneself, or even keep up
with basic expectations, within the performance management system is to simply do more work.

“The models of how much work one should take on are kind of insane. And so in order to
do that well, it’s not like it’s about face time or putting in a certain amount of time, it’s
just that the time required to get stuff done which is just the sort of expected is more than
is reasonable. It’s not a 40 hour week.” –Participant 34

“I don’t see examples of people encouraging work-life balance. I see examples of people
pressing us to grow faster and run harder and all that kind of thing.” –Participant 29

It is important to note that a small but interesting minority of interviewees (2) stated that
they believe that their work quality suffers when they are leading imbalanced lives. This was
matched by two interviewees that believed that no matter how imbalanced they were, they would
produce more work (and it would remain of adequate quality) than if they sought balance.
104

Although these two groups represent a small portion of the interviewees, it is important to note
that individuals can have different beliefs and preferences about their own performance and
abilities and that one solution may not fit well for all individuals.

Career
Closely tied to the Performance theme is the Career theme. See Table 23 for the
categories in this theme. Most of the employees at Genericorp want to perform well in service to
their careers so that they can grow in their roles and advance within the company (e.g. be
promoted). Almost two thirds (19) of the interviewees expressed concern that work-life balance
was not compatible with the career progression they envisioned for themselves, which often
included regular promotion and increases in responsibility.
Although not all the interviewees seemed satisfied with where they were at with regard to
their career and work-life balance, for the interviewees who were satisfied with their career
progress they generally fell into two camps. First, there is a type who is imbalanced, on a career
fast track and fully satisfied with that situation (3 interviewees fell into this camp). This type
generally felt she had agency over her choices, chose to work long hours and invest a lot of her
energy in her career. Second, there is a type who had consciously removed himself from striving
for the next promotion, was focused on creating sustainable work-life balance, and found areas
within his role to grow and excel within well-defined time and energy boundaries (three
interviewees fell into this camp). The characteristic that these two camps shared was that they
both emphasized choice and agency over their behaviors.

105

“I feel that if there ever comes a time when I physically cannot dedicate as much time,
then I will be able dedicate less time and continue to be successful … like I won’t have
any performance issues… At this point, I value being in the vortex, value being on the
team that is changing the industry, so I’m not as concerned. And if things get rough, then
I’m sure I would be able to find something else.” – Participant 29

“This is where my personal situation is sort of more interesting because I am at a level
where to get to the next level I think would require a ton of work. I am sort of resigned to
working… So my main interest is making sure my team is moving forward, so that’s
where I focus and concentrate my efforts …I’ve made in my own head a conscious
decision to manage my workload in such a way that I maintain that work-life balance.” –
Participant 39

Additionally here were three people who appeared to be in a third camp, a “deciding”
phase. What distinguished them is that they seemed to show less agency and control over their
situations than those above but they were moving in that direction. These interviewees were
actively evaluating their personal life and career goals and determining if they were compatible.
The rest of the interviewees either were satisfied with their work-life balance or were struggling
with trying to achieve both balance and traditional success.

“So why am I so fixated on this other, this other carrot. And if it’s to the detriment of
these other things… So I’m, I think I might be getting close to the level, and maybe I am I
just need to kind of have my aha moment of taking a step back and pulling away a little
106

bit, whatever those small increments, percentages, like I said, like “Okay, maybe I
respond to that one email tonight, but I don’t need to stay online 4 hours tonight pushing
forward all this different stuff”… or “this weekend, I’m not working at all”. Just,
Monday will come and it will be there waiting for me. I’m trying to find where that line
is but it doesn’t go so far, I don’t pull back so far that it impacts job performance and so
on. There’s also a little bit on an internal fear in myself of if you, I don’t know if this is
the right analogy, but if you step off the treadmill, or turn it too low, you won’t be able to,
kind of, ramp it back up again if you need to.”- Participant 13

Anxiety/ Stress
The interview script did not specifically call out feelings of stress or anxiety, but for a
substantial minority of interviewees, this theme arose during the course of the interview session.
See Table 24 for categories in this theme. More than one third (12) of the interviewees
specifically mentioned feeling anxious, frustrated or stressed by their work or work-life balance.
Interestingly, when asked for further elaboration, all cited similar reasons for this stress.
Related to themes mentioned above, all twelve mentioned that the performance
expectations and ambiguous performance goals are a source of stress as there is always a feeling
that one can be working more and more and that the work never ends. This can make it difficult
for individuals to draw boundaries around their work and personal lives. Those who felt stress
about their work lives mentioned that this boundless work was an area that caused stress. They
were also likely to say this quality was not unique to Genericorp and was likely a function of the
fact that Genericorp is part of a fast paced industry where speed of innovation is expected to take
place as a very accelerated pace. With such external pressures on the company, many felt that
107

similar pressures are communicated internally as well to keep up the pace to help the company
succeed.
“With the work world, especially at Genericorp, there’s always way more stuff out there
to do so you never can be really done for that day or that week or whatever… That’s
really hard to figure where you go, okay, I’m done working right now.” – Participant 13
“I don’t think it’s possible to slow down because our competitors will not slow down,
start ups will not slow down, if you slow down, you’ll die. So, it’s definitely… I don’t
think the pressure is ever going away, and there will always be this very… this industry
will never be for people that like to work 9 to 5.” –Participant 18

Fewer people (9) mentioned that workload or issues at work caused them to lose sleep of
feel tired. This is still a substantial minority, however, and given the importance of sleep and rest
in mental and physical health (Strine & Chapman, 2005) and productivity (Barens & Hollenbeck,
2009), it bears noting.

Global/Commute Issues
Another theme that rose organically from the interviews had to do with where work was
located. See Table 25 for categories related to this theme. Slightly more than half (16) of the
interviewees had long commutes to and from work, often averaging over an hour each way, and
many mentioned that these commutes made it difficult for them to successful balance work and
personal life as the time they had at their homes in the evenings was greatly diminished.
108

“One thing that puts pressure on it is the fact that I have a commute. I will say that one
thing that I have thought about is getting a job where I don’t have a commute.” –
Participant 34

Another location challenge that emerged for many interviewees had to do with some of
the challenges that come with being part of a company that has a presence in several locations
and across several time zones. Although all the interviews took place at the headquarters in a
single time zone and office location, Genericorp has many office locations across the world and
collaboration with team members in remote locations is common for employees. Eight
interviewees mentioned coordinating meetings across time zones as something that caused some
problems with work-life balances, as sometimes meetings ran into personal life time to
accommodate disparate time zones of teammates.

Family
See Table 26 for a summary of the categories in this theme. During the interviews, less
than half of the interviewees mentioned family (such as a spouse, children or other close
relatives) as an important presence in their personal lives (13). Of these, ten mentioned having
children (for a discussion on detachment and children, see Detachment theme section).
In general, those who mentioned having a family felt that their work-life satisfaction was
impacted by the feeling that they were unable to spend as much time with their families as they
wanted. For a few (3), they specifically talked about how the interviewee’s decision to work the
amount they were working (whether it was a 45 hour work week or a 60 hour work week) was a
109

decision they made as a family and one that their spouses supported. For the two interviewees
that had this arrangement and worked very long hours, they also had a spouse that stayed at
home and took care of home related responsibilities, a more traditional work-family model that
could sustain long working hours for one adult member.
As a counterpoint, there was one single, working mother in the sample and she was
conflicted about the time she spent at work and at home. She wanted to spend more time with her
children, especially as they got older (they were teenagers), but was worried that she would be
sacrificing too much in how others would perceive her performance and that her job would be in
jeopardy. In this case, the interviewee was both the head of the household and the breadwinner,
and she seemed to be feeling the strain of balancing both roles.

“I would, for example, I would love to take some time off next summer to spend some
time with my older daughter, doing whatever she needs to attend to prepare for
college. But I really worry, I get stressed out about my job security here. And I, it’s a
really, really bad feeling.”—Participant 30

Top leaders
Several interviewees were asked about how they felt Genericorp top leadership (at the C-Suite
level) supported work-life balance. The categories in this theme are summarized in Table 27.
Similar to the Team/Culture theme above, the results from this theme show a bit of a disconnect
between espoused values and enacted ones. While eight of the interviewees say that they receive
a clear message from top leadership that they support work-life balance, nine feel that the top
leaders themselves are not good role models for balance. The lack of good role models then
110

tends to lead the interviewees to believe that the support the leadership says it has for balance
may just be lip service, and the way to succeed at Genericorp is to lead a life heavily skewed
toward work.

“Arguably, perhaps, there’s lip service toward work-life balance within the company, but
I don’t really, it’s very, very rare that that plays out in a positive way. I think that that’s
the exception and not the norm, when it does.” --Participant 30

“No, I mean I look at [my leader’s] life and I’m like “Good Lord, I don’t want
that”. [laughs]. She works all the time, she’s online all the time … The only people who
I can think of who I can look at and I think “Oh yeah, that looks better, I can kind of
envision that, or whatever” are people who are low, at my level or lower… don’t see
anyone above who seems to… or at least creates the image of a more sane [balance].” –
Participant 34

Managing
The categories for this theme are summarized in Table 28. About half (14) of the interviewees
were themselves managers. As part of the interview script, they were asked to elaborate on how
they would help their direct reports deal with work-life balance issues if they arose. Most (11)
mentioned offering emotional support in the form of encouraging open and honest
communication, encouraging the subordinate to disengage and take time off, and talking through
the issue. An equal number mentioned trying to do something proactive about the problem, such
as shuffling around the workload to other teammates, reprioritizing projects to lessen workload,
111

requesting additional resources or assisting on pushing back on requests that come from outside
of the team to help reduce the workload.

“I’ve made sure, made… that whatever their family responsibilities are that they are not
getting out if it. If I see someone stressed out, I do make it a point to talk to them about it,
is there anything we can do. Um, if it starts to… sometimes there are health related
issues as well. Um, if they start to take up too much of their time, then there’s another
conversation to be had. Um, if they have a newborn or something I make it a point to
talk about it once in a while. I found that every time is kind of not as productive, but
every now and then talk about it.” – Participant 30

That said, half of those who manage people (7) mentioned they struggled with how to
manage the well-being and work-life balance of their subordinates. A couple mentioned that
they were unsure when it was an appropriate time to bring something up to a subordinate, or if
they should just wait until they were approached. Three mentioned the uncomfortable position a
manager is often of delivering a quality outcome from the team in a certain amount of time and
also trying to look out for the best interests of their team. To achieve ambitious goals (which
seem to be the norm for Genericorp), this often means that managers have to put more workload
and expectation pressures on their teams at the expense of work-life balance. A few managers
mentioned that they felt bad about doing this, but did not know of a workable solution to fix it
given the amount of work that needed to get done and the number of people they had to do it.

112

“And that’s a difficult thing to try to manage because everyone has too much work to do
so you can’t really just at the end of the day say “so, did you get everything done?”
because no one’s going to get everything done. So it puts a lot, you know, so it puts a lot
of burden on someone who has to lead all these people because they each have a different
workload. And I have to, I think, more carefully make sure that everyone has a full
plate.” – Participant 29

Rewards
This section is somewhat related to the performance section above, but is distinct in that
it related more closely to the Conservation of Resources Theory (Hobfoll, 1989) discussed in
earlier parts of this paper. The categories for this theme are summarized in Table 29.
Interviewees here discuss how work itself can be something that is rewarding or draining,
and this nature of work has an effect on work-life balance. Eleven interviewees mentioned that
if work is going poorly or if they feel that they are not being adequately rewarded for the effort
they put into their jobs, that their work-life balance suffers even more. Conversely, for those (8)
who feel that they work is very intrinsically rewarding, they appeared to be somewhat buffered
from work-life dissatisfaction as the work itself provided a resource that prevented them from
becoming too depleted.

Sustainability/Attrition
The final theme that emerged from interviews is that of sustainability and attrition and is
summarized in Table 30. It is important to note that not all interviewees who were dissatisfied
113

with their balance felt that their current level of work focus was unsustainable in the long term or
that it was not possible for them to make changes in their current role to achieve a more
satisfactory balance. Ten of the interviewees, however, did say that they felt they would need to
leave their roles in one form or another in order to achieve a more satisfactory balance. Some (6)
felt that they could do this through an internal transfer to another role, while others (7) felt that
they only way to really re-balance their lives would be to leave the company entirely.

“I don’t think it’s sustainable long term. Um, it’s been an ongoing concern for me since
I joined [Genericorp] 5 years, almost 5 years ago. Um, but I’ve come to the acceptance
that it’s just the nature of the job in the field that I’m in and so, I sort of rationalize it
away in that sense. Like, it’s not a long term solution, but for now I’m in a position
where I can afford to invest the time, and I feel like I’m compensated for my time, so that
makes it justifiable if not totally enjoyable.
Interviewer: If in the future you’re not willing to make that tradeoff, what would you do?
Participant 26: I’d leave [Genericorp].” –Participant 26

Although only one of the interviewees mentioned actively seeking out other positions at
different organizations during the interview, the fact that work-life balance seems an
unachievable goal at Genericorp for fully a third of the interviewees is concerning from an
employee retention perspective. As Study 1 explored (and Study 3 will investigate further), even
small effects to retention can have large cost implications to an organization that employs mainly
highly qualified, professional employees that are more difficult and more expensive to source
and hire.
114

Discussion Study 2
The qualitative interviews provide a rich context for how employees across different
levels, functions and tenures at Genericorp are experiencing work-life balance satisfaction. In
summary, 16 meta-themes arose from the interviews, eight which were specifically targeted in
the interviews and eight of which arose organically during the discussions.
The eight targeted themes revealed important similarities and differences in how
interviewees experience detachment, work-life balance satisfaction, workload, manager support,
flexibility, team and culture support, top leadership support, and managing subordinates. Among
the most interesting differences is how strong the local culture can be in influencing the
individual’s conceptualization of the global culture of Genericorp. Only two interviewees had
insight into the idea that their own experiences may differ from those of Genericorp employees
in different teams or functions. Most felt that their experiences with flexibility, pressure, and
expectations were indicative of a larger culture, even if these impressions were vastly different
between interviewees. This type of local to global culture extrapolation has been observed by
others in the academic literature (Caplan, 1987; Kristof, 1996), but is interesting in that it may
indicate a potential roadblock to attempts to change culture from the top-down.
Perhaps more interesting are the themes that emerged organically through the course of
the interviews. These eight themes are important in that they highlight the benefits of a
qualitative design in allowing for opportunities to explore ideas that were outside the original
scope of the research question. The themes that emerged were agency, control, performance,
anxiety/stress, commute/global issues, family, rewards, and sustainability/attrition. The issue of
control in particular appears to be an extremely important one for work-life balance. This comes
through very clear especially for the interviewees who appear to be among the highest
115

performing but still satisfied group. Although this is a very small minority in the interviews
(only two interviewees fall in this category), what sets these two apart is the rigid control they
exert over their schedules and work. Although originally not the focus of this paper, the link
between autonomy, control and work-life balance has been established in the literature, although
more rigorous research in the area is needed (Batt & Valcour, 2003; Thompson & Prottas, 2006).
One surprising finding from the interviews had to do with team dynamics and how this
related to performance pressure and expectations. Going into the interviews and based on past
research on team norms and dynamics, the expectation was that there would be more or less
explicit norms for long work hours, “facetime” and imbalance on teams where individuals felt
less satisfied with their work-life balance. This, however, was not to be the case. Rather, the
norms on the team were generally explicitly supportive of balance, with teammates trying to
encourage each other to take breaks, disconnect and adopt balanced work and life habits. Few
interviewees also felt that there was direct competition within their team, although two
interviewees did feel this competition. Generally, the team norms arose less from explicit,
unsupportive talk or behavior and more from implicit beliefs about performance and
expectations. Most interviewees felt that in order to distinguish themselves (or even keep up)
with highly competent and performance-oriented peers, they needed to push themselves to take
on more and more work, past their own personal preference levels. If all teammates in a group
hold the same implicit belief about their peers, it is easy to see how this could create a layer of
ever escalating performance pressure that would exist without ever being made explicit.
When thinking holistically about the interviewees as being either satisfied or dissatisfied
with work-life balance, some general qualities emerged from the satisfied group that tied them
together and distinguished them from those who were dissatisfied. Those who were satisfied
116

with their balance were more likely to mention having manageable workloads and work hours
(e.g., less than 55 hours per week), manager and team support, control over their work, adequate
rewards, and general engagement with their work. Those who were dissatisfied often lacked at
least one piece of this puzzle (e.g., lack of control, too many hours, or lack of engagement) and
this seemed to tip the scales into imbalance. It often took only one of these elements to be out of
whack for the individual to experience the negative effects to their balance, although those with
multiple problems showed larger signs of distress with their work-life balance and were more
likely to say they were contemplating looking elsewhere for work if things did not improve.

Strengths and Weaknesses of Study 2
The main strengths of Study 2 lie in the fact that it is a qualitative study and as such is
able to add depth to the understanding of work-life balance formation. Because of the open
ended nature of the qualitative interview, themes that were unplanned emerged, and from these
themes could come a deeper understanding of the nuances of team cultures, agency and
individual’s feelings about their own balance.
The main weaknesses of Study 2 are those that plague many qualitative studies. In the
pursuit of more detailed understanding, generalizability is sacrificed. A small sample size of 30
is not large enough to make generalized claims that the experiences of this group of employees
are applicable to the larger whole. Additionally, although an attempt was made to select
interviewees from a cross section of the Genericorp population and across satisfaction levels, it
appears that there may have been some self-selection bias in those that chose to be interviewed.
As attempt was made to create equal groups of those satisfied and unsatisfied with work-life
balance, but in the end the interviews skewed toward those who were unsatisfied. This is likely
117

not an accurate representation of the company as whole, as from Study 1, satisfaction with worklife balance averaged higher than such numbers would indicate.

118

CHAPTER 4: STUDY 3
Study 3: Quantitative Survey of Individual Preferences and Perceptions of WLB support
The purpose of the third study was to more fully explore the hypotheses presented in this
paper that were not addressed by Studies 1 and 2. Study 3 replicated some of the investigations
of Study 1 using different measures in an effort to triangulate the relationships as well as look at
additional constructs such as emotional exhaustion, flexibility and segmentation that were
unavailable in the archival data. In addition, Study 3 used scales to assess some of the more
nuanced aspects of Study 2 to explore whether some of these constructs could be examined at the
cross sectional level in survey data. It is important to note that in Study 3 the measure used to
assess the work-life variable is work-life interference (e.g the degree to which work interferes
with life) rather than work-life balance. This means that the valence of this variable changes in
all related work-life hypotheses.
Hypothesis Development from Qualitative Results: Boundary Management
Before diving into the results from Study 3, it is necessary to take a moment and describe
the supplemental hypotheses developed from the input from Study 2. Hypothesis 24 stated that
work-life boundary management techniques will moderate the relationship between workload
and work-life interference such that those that use certain techniques will experience less worklife interference than those who do not. Hypothesis 24 was left deliberately vague to allow for
additional variables to be added upon reflection on the interview data from Study 2. See Table 1
for a summary of all Hypothesis 24 sub hypotheses.
One boundary management technique, actual integration of work and home life, was
proposed a priori to the interviews and will be used as one of the potential moderators tested
under this hypothesis (H24.1). An additional hypothesis was added to investigate whether
119

individual preference for integration techniques had an interactive effect with actual behaviors
based on the number of interviewees who seemed to want a more separated work-life schedule
but were unable to enact it (H24.2).
To explore other moderators used in this study, the interviews pointed toward three main
areas where individuals may experience differences in ability or success in managing the
boundaries between work and non work-life. Namely, these were beliefs about boundary
flexibility, work control factors and psychological and behavioral detachment.
Beliefs About Boundary Flexibility. Several interviewees mentioned that they felt the
boundaries between work and personal life were asymmetrical; that they had a much easier time
flexing work into home life but a harder time flexing non work back into the boundaries
traditionally defined as “work”. To quantitatively capture this concept, I used a scale developed
by Matthews, Barnes-Farrell and Bulger (2010) that assesses individual’s beliefs about both their
ability to flex work into home (and home into work) and their willingness to do so. The Life
Flexibility Ability scale measures an individual’s belief that they are able to flex work into
personal life—that is that the boundaries around personal life time are permeable to work. The
Life Flexibility Willingness scale measures an individual’s willingness to allow this same flex to
occur. The other side of the coin is the Work Flexibility Ability scale. This scale measures an
individual’s belief that they are able to flex personal life into work. The Work Flexibility
Willingness scale measures their willingness to allow this particular flex to occur.
The goal of these measures is to more closely capture what some of the interviewees
were describing as their experiences with asymmetry of flexibility. They appeared to have
beliefs about the flexibility of boundaries and their willingness to cross them in one direction and
not the other. People who see the boundaries as being flexible in that they can flex personal life
120

into work and are willing to do so, are likely to experience lower work-life interference (H24.3
and 24.5). On the other hand, people who see work as being more able to flex into non work
time and are more willing to allow this are likely to experience greater work-life interference
(H24.7 and 24.9). These variables will also interact with workload such that they will moderate
the effect of high workload on work-life interference (H24.4, 24.6, 24.8 and 24.10). See Table 1
for a complete list of hypotheses.
Work Control Factors. Another aspect of work-life interference and boundary
management that came up during the interviews had to do with work control. Several
interviewees mentioned that frequent “emergencies” and “fire drills” at work made it difficult to
get their planned tasks done during working hours. This meant that they needed to push their
“normal” work tasks into evenings and weekends in order get things done, resulting in work
interference with life. Not all jobs or roles seem to have these types of disruptions, however, so
it appeared that control over one’s work day may vary from person to person.
To assess this, I included the measures of job autonomy and control of work. These
measures are related but distinct. Past research in the area supports this, showing that perceived
job control and autonomy have differential relationships to outcomes such as motivation,
performance, satisfaction and commitment (Spector, 1986; Breaugh, 1989). Autonomy is more
about the individual’s belief that they can arrange their work as they see fit without a lot of
oversight from others (Breaugh, 1989). Control of work, however, has to do more with feeling
in control of one’s workload and time (Karasek, 1978). People who feel less in control of their
work and who feel less autonomous would likely experience greater work interference with life
due to the consequences these work factors would have on pushing workload from the work day
into personal life time (h24.11 and 24.13). For people with greater control of work, they will be
121

buffered against the effects of high workloads on work-life interference. This will be played out
in an interaction effect where those who have high control of work will experience lower levels
of work life interference than those with similar levels of workload but a lower sense of control
(H24.12). Similarly, autonomy will function like control of work and also moderate the effect of
workload on work-life interference, such that those with more autonomy will experience lower
levels of work-life interference, particularly at high workloads (H24.14).
Psychological and Behavioral Detachment. When questioned, a few interviewees
specifically stated that it was not the workload necessarily that was interfering with their
personal lives, it was the preoccupation with work during non work time. Some reported having
difficulty psychologically detaching from work while at home even when not actively engaged in
work. To examine this, a measure of psychological detachment developed by Sonnentag and
Fritz (2005) is included. It is proposed that people who are better at psychologically detaching
from work will have lower work family interference (H24.15) and this will moderate the effect
of workload on work life interference (24.16).
In addition to psychological detachment, several interviewees mentioned that the
technology available to them made it possible to work from home at all hours and this resulted in
quite a bit of afterhours work on the weekends, evenings and on vacations. In order to get a
measure of behavioral detachment during non work hours, I developed three scales to determine
how much participants were working during traditionally non work times: the weekends,
evenings and vacations. As these behaviors are clear examples of work occurring during
personal time, I hypothesize that they will be related to higher work-life interference (H24.17,
24.19, 24.21). For individuals who are able to behaviorally detach and not work in the evenings,
weekends or on vacations, this off-hours work style will moderate the effect of workload on
122

work-life interference such that they will have lower work-life interference than those people
with similar workload who do not behaviorally detach (H24.18, 24.20, 24.22).

Procedure
Like Studies 1 and 2, Study 3 was conducted at Genericorp. This study was survey based
and used a stratified random sample of employees. In order to reduce survey strain on
employees, my contacts at Genericorp recommended I only approach 10% of employees for this
project rather than sample the entire company. Although the original intent of the proposal of
this study was to focus exclusively on North American Genericorp employees, my contacts at
Genericorp requested that a smaller subsample of employees at their global offices be surveyed
as well so that they could review the results across the company. However, although they
supported sampling intact teams in North America, we all agreed that team based sampling
globally would introduce too many layers of complexity to the survey and require a disruptively
large sample size. Due to the global request and differing sample requirement depending on
region, the sampling method used in Study 3 was conducted in two ways.
North American sample. The North American sample was gathered with a focus on
intact teams. Teams were pulled from across all North American offices with roughly 1/3 of the
sample falling into each of the major three functions: Engineering, Sales and General and
Administrative. Teams were defined as all individuals reporting up to the same manager. The
demographic make-up of the teams does not reflect Genericorp as a broader company because
the teams were not selected entirely randomly. Rather, teams were randomly selected based on
gender strata with the goal to have teams with roughly equal gender distribution. This was done

123

to address both the privacy requests of Genericorp and to ensure that demographic variables of
interest were adequately represented.
Another strata within which the teams were randomly selected is that of manager status.
Teams were selected where the majority of the team members were not themselves managers.
This decision was made based on research that suggests that when people manage others, they
tend to associate questions references their “team” to the people they manage rather than to the
other people who report up to the same manager as them. The lower level teams tend to interact
more with each other than teams of managers and are more appropriate to the types of variables
studied in this paper. In addition to the team members, the managers of the teams were also
surveyed. Teams were excluded if they had fewer than 3 members or more than 10 members.
Teams with team members located in another country outside of North America could be
included as long as the majority of the team was located in North America. Finally, Genericorp
requested that no one above a specific job level be surveyed as they have an internal policy to
avoid over surveying their most senior employees.
Because the teams were not purely randomly selected, the resulting demographic makeup of the sample, including gender, age, tenure, job level and manager status are not
representative of Genericorp as a whole. In all, 1272 individuals in North America were
approached for participation and 706 responded (56% response rate). See Table 31 for a
summary or response rates. See Table 32 for a summary of sample characteristics.
Non North American sample. For the global sample, selecting intact teams to sample
was not important, so individuals were sampled independent of their teams. It was still
important to obtain adequate representation for demographic variables, however, so participants
were selected randomly with the goal of roughly equal representation from each of the three
124

functions and with roughly equal gender dispersion. Age was also specifically selected for in the
global sample, with roughly equal numbers of participants selected from each region in the
following age groups: under 30 years old, 30 to 40 years old and older than 40 years old. In all,
This sampling method resulted in a slightly older sample for the non North American sample
[one way ANOVA, F(1,1200)=9.80, p<.01)] with slightly higher job levels [one way ANOVA,
F(1, 1139)= 25.7, p<.01], and slightly longer tenure [one way ANOVA, F(1, 1140)=8.79, p<.01].
See Table 32 for summary of sample characteristics.
Employees were approached for participation through email in a manner similar to that of
Study 2. Within the email, a link to an external survey site was included. This site included
informed consent materials as well as confidentiality assurances. An incentive for participation
was offered in the form of a chance to win one of four $100 Amazon.com gift certificates for
completing the survey.
Measures.
Where at all possible, existing scales with demonstrated psychometric properties were
used. Study 2 informed the development of three scales relating to after-hours work, and led to
the inclusion of several additional extant scales not proposed in the introduction aimed at
examining more closely the individual’s beliefs about work-life boundaries, work control, and
boundary spanning behaviors. Unless otherwise noted, scales were on a 1 to 5 agreement scale,
with 1 being “Strongly disagree” and 5 being “Strongly Agree”. See Appendix C for complete
measures.
Subjective Workload. Workload was assessed using a 3 item scale adapted scale from
Janssen (2001). An example item is “My workload is high”. Cronbach’s alpha for this scale was
.75 and it had a mean of 3.78 and a standard deviation of .69.
125

Quantitative Workload. Quantitative workload was assessed by asking participants to
provide the average number of hours worked per week. Respondents selected one of 10 response
options that listed hours on a 1 to 10 point scale in 5 hours increments, ranging from less than 35
hours per week, 35 to 40 hours per week, on up to more than 75 hours per week. The mean for
this item was 4.79, with a standard deviation of 1.68. The scale point 4 corresponded to the
hours range “46 to 50” per week.
Work-life Interference. Work interference with life was assessed using 5 items adapted
from O’Driscoll, Ilgen, and Hildreth’s (1992) work interference with life scale. And additional
item “Overall, I am satisfied with the balance between my work-life and my personal life” was
added to this scale at the request of Genericorp. An example item includes “I am worried that my
work interferes with my non-work activities and interests”. Cronbach’s alpha for this scale was
.86. The mean of this scale was 2.88, with a standard deviation of .85.
Negotiated Flexibility (I-deal flexibility). Individual flexibility i-deal provisions was
assessed using Hornung, Rousseau, Weigle, Glaster and Angerer’s (2009) scale. Participants are
asked to indicate on a 1 to 5 scale (1= not at all to 5 = to a great extent) the extent to which they
have asked for and successfully received flexible work arrangements for 6 items. A sample item
is “A work schedule suited to me personally”. Cronbach’s alpha for this scale was .93. The mean
for this scale was 2.93, with a standard deviation of 1.14.
Satisfaction with Flexibility. Satisfaction with flexibility was assessed using a scale
developed by Civian et al (2008) and Lu, Kao, Chang, Wu and Cooper (2008) and included 4
items. An example item is “I have the flexibility I need at work”. Cronbach’s alpha for this
scale was .73. The mean for this scale was 3.31, with a standard deviation of .44

126

Used Flexibility. Used flexibility was measured using 2 items from Jones, Scoville, Hill,
Childs, Leishman et al, 2008. These two items ask participants to indicate how often they are
able to arrange work from home and non standard start and end time in their jobs. Internal
consistencies for these items are not available as it is a two item scale. The inter-item
intercorrelation was relatively low at .40, which may indicate people use flexibility in different
ways. The mean value of these two items was 2.76 with a standard deviation of 1.08.
Preferences for Segmentation. Segmentation preferences was assessed using a scale by
Desrochers, Hilton and Larwood (2005). This 4 item scale included items like “I prefer to keep
work-life at work”. Cronbach’s alpha for this scale was .88. The mean of this scale was 3.68
with a standard deviation of .90.
Actual Integration. Actual integration was measured using a scale by Desrochers, Hilton
and Larwood (2005). This 3 item scale included items like “It is often difficult to tell where my
work-life ends and my non-work-life begins.” Cronbach’s alpha for this scale was .80. The mean
of this scale was .80, with a standard deviation of .97.
Emotional Exhaustion. Emotional exhaustion was assessed using items adapted from
the emotional exhaustion facet of Maslach and Jackson’s (1984) Burnout Inventory. Items that
directly reference “people centered” work were omitted (e.g. “working with people all day is
really a strain for me”). The remaining 4 item scale included items like “I feel burned out from
my work”. Cronbach’s alpha for this scale was .81. The mean of this scale was 2.63, with a
standard deviation of .83.
Retention. The same scale from Study 1 was used in Study 3. An example item is “I
plan to be working at Genericorp one year from now.” Cronbach’s alpha for this scale was .84.
The mean for this scale was 3.98, with a standard deviation of .79.
127

Performance. Performance is assessed using the same derived metric as in Study 1.
Average performance rating was 3.43, with a standard deviation of .25.
Manager Emotional Support for WLB. Manager support for WLB was measured using
the emotional support subscale from Hammer, Kossek, Yragui, Bodner & Hanson (2009)’s
Family Supportive Supervisor Behaviors (FSSB) scale and consists of 4 items. Cronbach’s alpha
for this scale was .93. The mean for this scale was 3.78, with a standard deviation of .84.
Team Emotional Support for WLB. Team support for WLB was measured using an
adapted version of the manager emotional support for WLB scale from the FSSB scale from
Hammer et al (2009). A sample item is “My teammates are willing to listen to my problems in
juggling work and non-work-life.” Cronbach’s alpha for this scale was .92. The mean for this
scale was 3.67, with a standard deviation of .76.
Perceived Manager Norm for WLB. Perceived manager norms for WLB was
measured through a subscale from Hammer et al (2009)’s FSSB scale that looks at instrumental
support and consists of three items. An example item is “I can depend on my manager to help
me with scheduling conflicts if I need it.” Cronbach’s alpha for this scale was .85. The mean for
this scale was 3.64, with a standard deviation of .79.
Perceived Team Norms for WLB. Team support for WLB was measured using an
adapted version of the manager instrumental support for WLB scale from the FSSB scale from
Hammer et al (2009). A sample item is “I can depend on my teammates to help me with
scheduling conflicts if I need it.” Cronbach’s alpha for this scale was .85. The mean for this scale
was 3.74, with a standard deviation of .72.
Team Task Interdependence. Team task interdependence was assessed using a 5 item
measure from Langfred (2005). A sample item is “I depend on other people for information I
128

need to do my work”. Cronbach’s alpha for this scale was .76. The mean for this scale was 3.56,
with a standard deviation of .67.

Measures Added. Measures added based on findings from Study 2
Life Flexibility Ability. Life flexibility ability was assessed using a slightly adapted
scale from Matthews, Barnes-Farrell and Bulger (2010) and consisted of 5 items. The scale was
adapted to more accurate reflect personal life in general rather than family in specific. A sample
item is “If the need arose, I could work late without affecting my personal life responsibilities”.
Cronbach’s alpha for this scale was .85. The mean for this scale was 3.40, with a standard
deviation of .86.
Life Flexibility Willingness. Life flexibility willingness also used a scale from
Matthews, Barnes-Farrell and Bulger (2010) and, again, this 4 scale was slightly modified to
reflect personal life flexibility in general rather than family flexibility in specific. A sample item
is “I am willing to change plans with my friends and family so that I can finish a job
assignment”. Cronbach’s alpha for this scale was .74. The mean for this scale was 3.06, with a
standard deviation of .80.
Work Flexibility Ability. Work flexibility ability was measured using a 4 item scale
from Matthews, Barnes-Farrell and Bulger (2010). A sample item is “If the need arose, I could
leave work early to attend to personal life issues.” Cronbach’s alpha for this scale was .79. The
mean for this scale was 3.88, with a standard deviation of .67.
Work Flexibility Willingness. Work flexibility willingness was measured using a 3 item
scale from Matthews, Barnes-Farrell and Bulger (2010). A sample item is “I am willing to take
an extended lunch break so that I can deal with responsibilities related to my personal life.”
129

Cronbach’s alpha for this scale was .68. The mean for this scale was 3.87, with a standard
deviation of .65.
Psychological Detachment. Psychological detachment was assessed using a 4 item scale
from Sonnentag and Fritz (2005). A sample item is “During after work hours, I forget about
work”. At the request of Genericorp, an additional item was added to this scale: “I am able to
detach from work during non-work time (i.e. when I choose not to be working)”. Cronbach’s
alpha for this scale was .85. The mean for this scale was 2.70, with a standard deviation of .82.
Control of Work. Control of work was assessed using a 5 item scale from Claessens,
van Earde, Rutte and Roe (2004). A sample item is “I often have little control over what is
happening at work”. Cronbach’s alpha for this scale was .72. The mean for this scale was 3.43,
with a standard deviation of .65.
Autonomy. Autonomy was assessed using a 3 item scale from Spreitzer (1995). A
sample item is “I have significant autonomy in determining how I do my job”. Cronbach’s alpha
for this scale was .87. The mean for this scale was 4.12, with a standard deviation of .71.
New Scales Developed Based on Study 2
One thing that came up routinely in the interviews during Study 2 was that people tended
to do a lot of work away from the office during non-work time. Even though people were
physically leaving the office, the technological resources people had access to allowed them to
continually monitor email, access work files and do substantive work from any location. To
assess behaviors associated with this work flexing into personal life time, I created a series of
items assessing whether respondents reported working in the evenings, on the weekends or on
vacations.

130

Evening Work Style. Four items were developed to assess evening work style. All were
measured on a 5 point frequency scale 1= Never and 5 = Always or almost always. The items
were “Once I leave the office, I almost never log back on to do more work in the evenings.”, “I
check work email in the evenings but do not usually do work beyond emailing.”, “In the
evenings, I typically take a break from work for a while, but then log back on for an hour or two
before I go to bed”, and “ I do not get much of break in the evenings – I am pretty much logged
on to work from the time I get home to the time I go to bed.” I reverse coded the first item.
When examining scale reliabilities and the factor structure of the scale, it was revealed that the
reliability was improved if the second item about emailing was removed (from .41 to .72). the
mean for this item is higher than the other items (3.14 as compared to 2.14, 3.06, and 2.14,
respectively) so it is likely that checking email during the evening is not indicative of doing more
substantive work and is a more ubiquitous behavior. Therefore, the final scale consisted of three
items, with the email item omitted. The Cronbach’s alpha for this .72. The scale mean is 3.00,
with a standard deviation of .90.
Weekend Work Style. Four items were developed to assess weekend work style. All
were measured on a 5 point frequency scale, 1= Never to 5 = Always or almost always. The
items were “I almost never do work on weekends.”, “I check work email on weekends but do not
usually do work beyond emailing.”, “I generally try to keep on day (e.g. Saturday) work free, but
tend to do some work on the other weekend day”, and “ I often work both days of the weekend.”
I reverse coded the first item. Again, as with Evening Work Style, when examining scale
reliabilities and the factor structure of the scale, it was revealed that the reliability was improved
if the second item about emailing was removed (from .45 to .69). As with the Evening Work
Style scale, the mean for this item is higher than the other items (3.38 as compared to 3.10, 2.74,
131

and 1.98, respectively) so it is likely that checking email during the weekend might be a specific
behavior not indicative of more substantive work during the weekend. Therefore, the final scale
consisted of three items, with the email item omitted. The Cronbach’s alpha for this .69. The
scale mean is 2.54, with a standard deviation of .89.
Vacation Work Style. Five items were developed to assess vacation work style. These
items were “While on vacation, I am able to completely detach from the office.”, “While on
vacation, I frequently check work email.”, “While on vacation, I do some work most day.”, “I
wish I was able to detach from the office more effectively while on vacation.” And “While on
vacation I worry about the amount of work waiting for me when I get back to the office”.
Although the item about email checking did not load strongly with the Weekend and Evening
Work Style scales, this was not true for the Vacation work Style scale. This suggests that
checking email while on vacation may be similar to other work behaviors while on vacation in
ways that email checking during evenings and weekends is fundamentally different than other
weekend and evening work. Perhaps, checking email after and on weekends is ubiquitous in this
organization, while more employees are likely to fully disengage while on vacation. Because
email checking appears to be a part of vacation work style, all five items were retained for the
scale. Cronbach’s alpha for this scale was .82. The mean of the scale was 2.88, with a standard
deviation of .92.

Demographic Variables. A variety of job and personal demographic variables were assessed.
Job Variables. Job tenure, job level, manager status, function, region and team size
were assessed at job level demographic variables. See Table 32 for a summary of these
demographics. The average job level (minimum: 1, maximum: 7) was 4.40, with a standard
132

deviation of 1.44. The average tenure was 2.99 years, with a standard deviation of 2.11. The
2

average team size was 5.40 with a standard deviation of 2.71. The sample was made up of 621
men (54%), 521 women (46%), 242 managers (20%), 962 non managers (80%), 419 people from
Engineering (35%), 395 people from Sales (33%), and 388 people from General and
Administrative (32%). Most (706, 59%) of the respondents were from North America, with 245
(20%) from Europe, 194 (16%) from Asia/Pacific Islands and 59 (5%) from Latin America.
Personal Variables. Age, gender, marital status, number of children, age of children, and
hours spent on childcare and housework were assessed as personal life variables. Average age
was 33.92, with a standard deviation of 7.39. 621 men (51%) and 521 women (43%) and 82
(7%) participants of undeclared gender responded. Most participants were married or partnered
(661, 54%), with 366 identifying as single (30%). Seven (1%) reported they were separated, 26
(2%) were divorced and 4 (.3%) were widowed. Because so few respondents reported
relationship arrangements other than married/partnered and single, this variable was
dichotomized for further analysis to read as either married/partnered or single.
The number of children respondents had ranged from zero to 5, with a mean of .68 and a
standard deviation of 1.00. The age of children ranged from newborn to 33, with a mean of 6.49
and a standard deviation of 6.04. I also looked at whether individuals had young children, as
young children are frequently cited in the work-life research as being particularly demanding in

2

Average team size calculated on N. America subset only since that was the data

collected with team size in mind rather than the full global sample.

133

terms of time for new parents. 280 (23%) of respondents had children under 6 years old and a
subset of those, 216 (18%) had children under 3 years old.
Household Demands. To assess household demands, the following question was asked
of participants “How many hours do you spend per week on housework, including cooking,
cleaning, laundry, picking up dry cleaning, shopping, etc, or arranging for any of these types of
tasks to be done by others?” The responses to the question ranged from zero to 60, with a mean
of 9.64 and a standard deviation of 7.13. The qualities of the distribution of this variable indicate
that it is not normally distributed, so further exploration into the shape of the data was conducted.
A chart of the responses reveals that the data dispersion appears to be multi-modal, with peaks
around the 5, 10, 15 and 20 hours per week time period (see Figure 19) with only 4 individuals
falling in the extreme outlier group (over 40 hours per week). It appears that people may have
very different relationships with home care (e.g. some may hire help, some may spend a lot of
time cooking, some may be single and thus spend less time, some may be caring for multi family
homes) so these numbers may reflect reality rather than represent outlier exaggerations (Lee &
Waite, 2005; Kitterod & Lyngstad, 2005).
Child/eldercare Demands. To assess household demands, the following question was
asked of participants “How many hours do you spend per week on child and/or parent
care, including transportation of children and/or parents or arranging for child and/or parent
care?” The responses to the question ranged from zero to 105, with a mean of 9.94 and a
standard deviation of 13.89. Similarly to the findings for household demands, the when
investigating the distribution of child/eldercare demands, the distribution appears to be
multimodal rather than normal (see Figure 20), with many people spending 0 hours and others
spending 10, 20, 30 hours in groups. Again, there were only 4 individuals who represented true
134

outliers (e.g. over 70 hours per week) and the majority of data seemed to reflect real dispersions
of child/elder care responsibilities so the data was left as is (minus the outliers) for analysis.

Results Study 3
Descriptive Statistics
Functional contrasts. Like Study 1, this data set was made up of participants from three
major occupational functional groups: engineering, sales and general and administrative. Three
one-way, between subjects ANOVAs were conducted to compare the effects of function on the 5
variables that are made up the core variables addressed in this study: work-life interference,
emotional exhaustion and retention, with Bonferroni post hoc test of differences. Please see
Table 33 for the results. There were no significant differences between the functions in
emotional exhaustion scores [F(2, 1162)= .65, n.s.]. For scores on work-life interference, the
function General and Administrative scored significantly lower than the function Engineering
[F(2, 1151)=3.46, p<.05]. For the scores on retention, the function Sales scored significantly
lower than the function Engineering [F(2, 1162)=3.62, p<.05]. Compared to the results from
study 1 for functional comparisons in scores on outcome variables, there are both similarities and
differences. For the outcome of work-life balance/interference, the pattern of results is exactly
the same: Engineering scores show that that function is significantly more satisfied than the
Administrative function in both Study 1 and Study 3, with no other difference between functions
being significant. For the retention outcome, however, Study 3 shows that Sales is significantly
lower than the Administrative function, a finding that was not replicated in Study 1 which found
no functional differences in retention intentions.
135

I conducted two more ANOVAs in which I looked at whether it was expedient to
consider the functions as dichotomous by collapsing the Sales and General and Administrative
functions together as they tend to differ from the Engineering function in similar ways. This
collapsing increased the ANOVA for work interference with life [F(1, 1163) = 4.91, p< .01) but
not for retention [F(1, 1152) = .50, n.s.]. Using the Sales group as the comparison and collapsing
the other two functions together increased the ANOVA fit for retention, however [F(1, 1152) =
4.11, p< .05). Therefore, for regressions with retention as an outcome, the function control will
be dichotomized with Sales as the comparison group while for regressions with work
interference as the outcome, the function control will be dichotomized with Engineering as the
comparison group.
Regional Comparisons. As with function, I conducted regional comparisons of the three
core variables. The four main regions samples were North America, Europe, Asia/ Pacific
Islands, and Latin America. Three one way ANOVAs were conducted to examine possible
regional difference in work-life interference, retention and emotional exhaustion. There were no
regional differences for the work-life interference variables [F(3, 1150)= .71, n.s.]. There were
significant differences in retention [F(3, 1161) = 4.34, p<.01] and emotional exhaustion [F(3,
1161)= 8.12, p<.01]. Bonferroni post hoc tests revealed that North America scored significantly
higher than Europe on emotional exhaustion (mean difference .28, p<.01) and that Latin America
scored significantly higher than Asia Pacific/Islands and Europe on Retention (mean differences
.31, p<.05 and .41, p<.01, respectively). When using region as a control for emotional
exhaustion, region will be dichotomized with North America being compared to all other regions
(ANOVA is more strongly significant in this configuration [F(1,1163) = 17.73, p<.01]). When
using region as a control for retention, Latin America will be the region used as the comparison
136

for region (the ANOVA is more strongly significant in this configuration [F(1, 1163) = 7.39, p<
.01]). A MANOVA was also conducted looking at Region and Function in relation to emotional
exhaustion, retention and work-life interference as an alternative way of looking at the data. The
results are quite similar and are included in Appendix F.
Correlations. See Table 34 for correlations. The correlation matrix reveals quite a few
significant correlations. Some of these are relatively high. Work-life interference in particular
correlates with several variables at the .5 to .6 level (e.g. emotional exhaustion, psychological
detachment, control of work, actual integration, workload). Although these correlations trend
slightly high, they are not high enough to indicate that these constructs are indistinguishable
from one another. Indeed, correlations of this magnitude between work-life balance and work
targeted variables such as emotional exhaustion, work autonomy and workload are often found to
be in the .4 to .6 range (Adams & Jex, 1999; Hill et al, 2001) In addition, these levels of
correlations between these types of variables are relatively typical in the work-life balance
literature, with researchers routinely finding correlations between work-life conflict and
emotional exhaustion in the .7-.79 range (e.g. Hall, Dollard, Tuckey, Winefield, & Thompson,
2010; Singh, Suar & Leiter, 2011 ).
There is one area of concern in the correlation matrix where the high correlation indicates
that two constructs are not distinct from one another. Manager instrumental and emotional
support are correlated at .76. To examine how these functioned as a scale, I conducted an
exploratory factor analysis using maximum likelihood extraction and a Varimax rotation. The
factor analysis returned one factor with eigenvalues greater than one, indicating that all the items
from the two scales are more appropriately tapping into a single construct of manager support for
balance. These factors matched up with the emotional and instrumental factors of the scales.
137

Although some of my hypotheses depend on being able to differentiate manager emotional and
instrumental support for balance due to theoretical differences in where such support should
buffer WLI relationships the most, the data indicates that manager support for balance is more
appropriately treated as a single scale rather than two. For further analyses, the items in the
instrumental and emotional support for balance scales were collapsed into a single scale of
general manager support for balance. Interestingly, when an exploratory analysis was conducted
on the parallel items for team instrumental and emotional support, two factors with eigen values
greater than 1 were returned (Team instrumental and emotional support are correlated at .66).
Therefore the two distinct scales for team support were retained for further analyses.
The correlation matrix did provide support for some distinctions made between similar
constructs in this study. Workload was measured by both hours work and through a survey
measure of subjective workload. Although these measures are correlated (as one would expect),
the correlation of .42 is not high enough to indicate that they cannot be considered distinct
constructs going forward with analyses.
The different types of flexibility also showed low enough correlations that it would
appear that they are not tapping into a single construct. Flexibility satisfaction and used
flexibility are correlated at only .12 and negotiated flexibility (idiosyncratic deals) and used
flexibility are correlated at .37. It is justifiable that these three treatments of flexibility be treated
as distinct.
There were also four variables that examined the permeability of work and life
boundaries, both the respondents’ perceptions of the ability of these boundaries to be flexibile
and their willingness to allow flexibility. The correlations between these four constructs (Life
flexibility ability, Life flexibility willingness, Work flexibility ability, Work flexibility
138

willingness) were not so high as to be concerning for construct distinctness. Correlations ranged
between .06 and .56 for these scales. A confirmatory factor analysis was also conducted to
ensure that these four factors emerged from the Study 3 sample(see Table 35). Values under .25
were suppressed for ease of readability. There is some cross loading for two items on the Work
Flexibility Willingness scale onto the Work Flexibility Ability scale, but in general the factors
fall as they should according to subscale categorization.
The means and standard deviations of the scales were, in general, not worrisome from a
psychometric perspective. Like in Study 1, performance ratings may be range restricted give that
the standard deviation is so small (.24) which may affect the potential findings for hypotheses
involving this variable. Additionally, the mean for the retention scale may be slightly high
(3.98), which might indicate some ceiling effect range restriction for this variable as well.
Control Variables. When testing the hypothesized relationships, it is important to take
into account variables that may affect the variables of interest, and thus control for their potential
effects. The correlation matrix (Table 5) reveals that team size and function have a relationship
to the work-life balance variables, both for the individual and for the manager work-life balance
scores. These scores indicate that people on larger teams experience greater balance (as do their
managers) and people in the functions of Sales and Engineering experience better balance than
those in General and Administrative. Interestingly, tenure and job level are not strongly related
to work-life balance. Originally, job level was to be used as a proxy for age in the control
variables. Age is a variable often used in work-life balance research (Eby, Casper, Lockwood,
Bordeaux, & Brinley, 2005) that I was unable to collect for this data set, but given that job level
is unrelated to the work-life balance variables, it is unnecessary to use it as control, and it cannot

139

serve as a proxy. For hypotheses relating to work-life balance as an outcome, the control
variables of team size and function will be utilized.
Retention is related to different demographic variables. As evident from the ANOVA
performed above, retention intentions have no relationship to function. The correlation matrix
also reveals that they have no relationship to team size. Instead, retention intentions are related
to job tenure and, less strongly, to job level, with people showing decreasing intentions to stay
with the company the longer they have been there and the higher positions they have obtained.
For hypotheses that address retention, tenure will be used as a control. Only tenure will be
controlled for considering tenure and job level are correlated with each other and controlling for
both could result in problems with multicollinearity. Additionally, tenure may be logically
related to retention in that new employees are less likely to be looking for different career
opportunities outside the company and therefore may show artificially inflated retention
intentions. Some research indicates that for highly technical and/or specialized employees (such
as those employed by Genericorp), employees may be more likely to jump between companies
for promotion and/or career development opportunities (Barley & Kunda, 2004) This suggests
that longer tenure at a single company may be related to increased external job searching, thus
linking tenure to retention. As this link is irrelevant to the current study, the relationship of
tenure and retention will be controlled.

Hypothesis Testing
A Note on Statistical Models. Like Study 1, Study 3 takes place at 2 levels (the
individual and the work group). Because the interdependencies of groups can result in overly
permissive statistical tests when group data is analyzed without taking groups into account, it
140

was important to assess the variables for “groupiness” before any analyses were conducted. As
in Study 1, Hierarchical linear modeling (HLM) was used to check if the variables of retention,
work-life interference or emotional exhaustion met the criteria for group level variation. In order
the check for this “groupiness”, unconditional means models were conducted for each of the
3

three variables and ICC(1)s were calculated based on these models . Once a variable (work-life
balance) was identified as existing at the group level (level two), HLM was also used to test the
effects of other variables on this level two variable. HLM is used as both as test of “groupiness”
of variables and also as a way to statistically account for the interdependence of that
“groupiness” in subsequent analyses. See Table 36 for a summary of these findings from Study
1.
None of the three variables have significant Wald Z statistics for the intercepts, and
work-life interference and retention in particular have ICC(1) values that indicate that they do
not function as group variables (less than .05). Bliese (2000) discusses that there are no strong
guidelines about what ICC(1) should be used as a cut off to indicate a group level variable and
mentions that values between .05 and .2 are commonly used. The ICC(1) for emotional
exhaustion is .07, so it does fall in this range cited by Bliese, but considering the non-significant
Wald statistic it appears that emotional exhaustion should not be considered a group variable.

3

When conducting HLM tests, only a subset of the full data set was used. Only groups

with that had 3 or more of the group providing survey responses were considered in analysis. In
addition, managers were excluded from analysis because managers exist at level two rather than
level one.

141

Bliese also cites that .12 is the median value of ICC(1)s, which is considerably higher than .07.
Because of the non significant Wald statistics and the low ICC(1)s, I believe the potential for
overpowered statistics for analysis with emotional exhaustion are minimal to nonexistant and
therefore emotional exhaustion will not be treated as a group variable..
HLM will be used in later analyses to justify the aggregation of group level variables but
since the variables here are not group level, comparison of HLM models will not be necessary
for hypothesis testing. Subsequent hypotheses will be tested using linear regression. The choice
was made to use regression rather than other statistical techniques (e.g. SEM) for two main
reasons. The first is that a method was chosen that could address the potential nested effects of
group variables and SEM is not compatible with HLM nested testing (Raudenbsh & Bryk, 2000).
Because of nesting, some of the models in this paper used a reduced data set that only focuses on
the portion of the respondents who are part of teams with 60% or more respondents or those who
have managers responding. This reduces the dataset considerably, and could affect power path
modeling in SEM. Second, the focus on this paper is on parsimony and how the close
examination of each variable and way help explain pieces of how individuals experience worklife interference. This more piecemeal approach allows for focused attention on each variable
which can help illuminate specific aspects of individual experiences that contribute to
satisfaction with balance and emotional exhaustion. The choice was made in this paper to focus
on the smaller pieces with hope that this tighter view can be expanded in the future into a more
systems perspective and other methods such as SEM can be employed.
Role Theory and Conservation of Resources Theory Hypotheses. Hypothesis 1 stated
that workload will be positively related to work-life interference such that those with higher
workloads will report higher work-life interference. In Study 3, workload was measured both in
142

terms of hours work and through a subjective scale measure. To test this hypothesis, both types
of workload were entered into regression equations as the third step predicting the variable of
work-life interference, after the controls of job level, team size and team task interdependence
were entered and the types of flexibility were entered as second steps. See Tables 37 and 38 for
the results. Both subjective workload and work hours explained additional variance in retention
beyond the control variables. The effect size for subject workload was higher than that of work
2

hours (ΔR = .24 and .19, respectively, p<.01), although both show moderate effects. Thus,
Hypothesis 1 was supported and workload was related to higher levels of work interference with
life.
Hypothesis 2 stated that workload and performance would be negatively related such that
those with higher workload would be rated lower on performance. To test this, two hierarchical
linear regressions were conducted with the two workload variables entered at step two after the
2

controls. See Tables 39 and 40. The ΔR for both types of workload was significant, albeit quite
small (.01 for work hours and .02 for subjective workload, p<.01). However, the relationship was
in the opposite direction of the hypothesized relationship. This finding, while not in the direction
hypothesized, is consistent with findings from Study 1. People with high workloads were
significantly more likely to score higher on performance ratings although the effect was quite
small. Therefore, Hypothesis 2 was not supported.
Hypothesis 3 stated that performance would be positively related to work-life
interference such that those who were rated as high performers would have greater work
interference with life. To test this, a hierarchical linear regression was conducted with job level,
team size and team task interdependence used as control variables. See Table 41. Performance
143

did not additional explanatory power beyond what was accounted for control of work to work
2

interference with life as evidenced by a non significant ΔR . This relationship is different from
what was found in Study 1, where work-life balance was found to be negatively related (albeit
weakly) with performance. The measures used to assess work-life balance/interference differ
greatly between Study 1 and Study 3 and could account for this difference in findings. This
measure in particular is measuring different things than in Study 1. The Study 1 measure focused
on just three items, and these items included satisfaction with manager support, detachment and
general work-life balance—ultimately a broad spectrum of balance satisfaction. Study 3 was able
to more clearly define each of these construct spaces. In particular, work life interference was
defined as times when work interfered with life, rather than just global satisfaction. It may be
that the measure used in Study 1 tapped into a different construct than that in Study 3. Thus,
Hypothesis 3 was not supported.
Hypothesis 4 stated that workload and retention would be negatively related such that
those with higher workloads would be less likely to indicate intentions to stay with the company.
To test this hypothesis, both types of workload were entered into regression equations as the
second step predicting retention, after the controls of function (Sales/all others) region (Latin
America/all other) and tenure were entered. See Tables 42 and 43 for the results. Both
2

2

subjective workload (ΔR = .02, p<.01) and work hours (ΔR = .01, p<.01) explained additional
variance in retention beyond the control variables. Thus, Hypothesis 4 was supported.
Hypothesis 5 stated that performance and retention intentions would be positively
correlated such that those with higher performance ratings would be more likely to indicate
intentions to stay with the company. A hierarchical linear regression was conducted to test this
144

hypothesis and is presented in Step 2 of Table 44. Performance did not significantly alter the
2

value of R and thus is not related to retention intentions. Therefore, hypothesis 5 was not
supported.
Hypothesis 6 stated that work-life interference would be negatively related to retention
intentions such that those who report higher work interference with life will also report lower
intentions to stay with the company. To test this hypothesis, a two step hierarchical linear
regression was conducted and is presented in Step 3 of Table 44. Work-life interference explains
an additional 7% of the variance in retention beyond the control variables, supporting the
contention that those who experience more work-life interference report lower intentions to stay
with the company. Hypothesis 6 was supported.
Hypothesis 7 stated that work-life interference would moderate the relationship of
performance with retention intentions such that those with lower work-life interference and
higher performance would be more likely to indicate intentions to stay with the company. A
three step hierarchical linear regression was conducted to examine this interaction (with the
interaction computed through multiplying the mean-centered performance work interference
with life variables) and is presented in Step 4 of Table 44. The interaction term did not result in
2

significant change to R , therefore hypothesis 7 was not supported.
Hypothesis 8 stated that workload would be positively related to emotional exhaustion
such that those with higher workloads would report higher emotional exhaustion. This
hypothesis was tested using both measures of workload (work hours and subjective workload) in
a hierarchical linear regression. See Tables 45 and 46. Both subjective workload and work hours

145

2

resulted in significant increases in R and explained 23% and 12% of emotional exhaustion,
respectively. Therefore, Hypothesis 8 was supported.
Hypothesis 9 stated that work-life interference would be positively related to emotional
exhaustion such that those with higher work-life interference would also experience higher
emotional exhaustion. A hierarchical linear regression was performed to test this hypothesis and
2

is presented in Step 2 of Table 47. Work-life interference resulted in a significant change to R ,
2

explaining a large amount (ΔR = .40, p<.01) of the variance in emotional exhaustion beyond the
control variables. Therefore, Hypothesis 9 was supported.
Hypothesis 10 stated that work-life interference would partially mediate the relationship
of workload and emotional exhaustion such that when work-life interference was included in the
model, the effect of workload on emotional exhaustion would be diminished. To test this, two
three step hierarchical regressions were conducted with the controls entered as step 1, work-life
interference as step 2, and workload (one regression for each type of workload) as step 3. See
Tables 47 and 48. For subjective workload, the inclusion of work-life interference reduces the β
from .46 to .20, although it is still contributing significant explanatory power to emotional
exhaustion beyond work-life interference. For work hours, however the inclusion of work-life
interference reduces the β from .30 to .02 and work hours no longer contributes a significant
2

amount of explanatory power to emotional exhaustion (ΔR is reduced from .12, p<.01, to .00,
n.s.). Therefore, work-life interference fully mediates the relationship between work hours and
emotional exhaustion. This is notable in comparison to the mediation of work-life interference
on subjective workload as in this case it is only partial, such that the inclusion of work-life
146

interference on the model of subjective workload on emotional exhaustion only reduces the
effects of subjective workload rather than removing it entirely. Hypothesis 10 was supported for
subjective workload as a partial mediation while a full mediation hypothesis was supported for
work hours.
Coping Support Theory Hypotheses. Hypothesis 11 stated that manager emotional
support for balance would moderate the relationship between work-life interference and retention
such that those who report more supportive managers will report greater intentions to stay with
the company despite higher levels of work interferences with life. To test this hypothesis, a three
step hierarchical linear regression was performed and is presented in Table 49. Although there
was a direct effect of manager emotional support on retention (i.e. those with supportive
managers show greater intentions to remain with the company), the interaction term of manager
support and work-life interference was non significant. Therefore, Hypothesis 11 was not
supported.
Hypothesis 12 was the team analogue to Hypothesis 11 and stated that perceptions of
team emotional support for balance would moderated the relationship between work-life
interference and retention such that those with more emotionally supportive teams would report
higher retention intentions regardless of levels of work interference with life. See Table 50 for
the regression that tested this hypothesis. Like in Hypothesis 11, there was a main effect for team
support for balance (those with supportive teams reported higher intentions to remain with the
2

company, ΔR = .07, p<.01) but the interaction term was non-significant. Therefore, Hypothesis
12 was not supported.

147

Hypothesis 13 stated that manager emotional support for balance would moderate the
relationship between work-life interference and emotional exhaustion such that those with more
manager emotional support would report lower emotional exhaustion regardless of level of worklife interference. See Table 51 for the hierarchical linear regression that tested this hypothesis.
There is a main effect for manager emotional support (those with supportive managers
experience less emotional exhaustion, β = -.19, p<.01), but the interaction term is non significant.
Therefore, Hypothesis 13 was not supported.
Hypothesis 14 was again the team analogue to Hypothesis 13 and stated that team
emotional support for balance would moderate the relationship between work-life interference
and emotional exhaustion such that those with more team emotional support would report lower
emotional exhaustion regardless of level of work-life interference. See Table 52 for the
hierarchical linear regression that tested this hypothesis. There is no main effect for team
support and the interaction term is non significant. Therefore, Hypothesis 14 was not supported.
Border and Idiosyncratic Deal theory hypotheses
A Note on Used Flexibility. This study examined the use of flexible arrangements in two
ways. One scale, based on the flexibility dimension of idiosyncratic deal provision by Rousseau
(2005), consisted of 6 types of work flexibility that participants reported specifically negotiating
for. During the course of Study 2 it became apparent that some individuals did not specifically
negotiate for flexibility; rather, they just appeared to take it. Therefore, it seemed appropriate to
include a measure of used flexibility that did not rely on specific negotiation. The second
measure consisted of two items that asked respondents the frequency of use of work from home
options or flexing the start and stop times of their work day. To test used flexibility hypotheses,
both types of flexibility, negotiated flexibility and used flexibility will be examined.
148

Hypothesis 15 stated that the frequency of use of flexible arrangements would be
negatively related to work-life interference. See Tables 53 and 54 for the regression testing this
2

hypothesis (Step 2). The change in R was significant for negotiated flexibility, although the
2

effect size was quite small (ΔR = .01, p<.01). This provided some support for Hypothesis 15,
but suggests that there might be some differences between simply using flexible options and
specifically negotiating flexibility in terms of effects on work-life interference, although the
effects are small.
Hypothesis 16 stated that satisfaction with flexible arrangements would be negatively
related to work interference with life such that those with higher satisfaction with flexibility
would report lower work interference with life. See Table 55 (Step 2) for a summary of the
2

hierarchical linear regression testing this hypothesis. The change in R for flexibility satisfaction
predicting work-life interference was significant, with those reporting greater satisfaction
2

reporting less work-life interference (ΔR = .17, p<.01). Therefore, Hypothesis 16 was
supported.
It is important to note that the effect size for flexibility satisfaction on work interference
with life is quite large (predicting 17% of the variance), which is in noted contrast to the very
small or nonexistent effect sizes of the used or negotiated flexibility effect sizes of Hypotheses
14 and 15. It would appear that the subjective feeling of satisfaction with flexibility is much
more impactful to experiences of work interference with life than actual use of flexibility
supports.

149

Hypothesis 17 stated that use of flexible arrangements would moderate the relationship
between workload and work-life interference such that those who report more use of flexible
work arrangements would also report lower work-life interference regardless of work load.
Since this study included two measures each of workload and used flexibility, four regression
equations were conducted to examine this hypothesis. See Tables 53, 54, 56, and 57 and
Appendix F for the results of these regression equations. For interactions, mean centered
variables were multiplied with each other. None of the interaction terms were significant.
Therefore, Hypothesis 17 was not supported.
Hypothesis 18 stated that satisfaction with flexible arrangements would moderate the
relationship between workload and work-life interference such that those with more satisfaction
with flexibility would report lower work-life interference regardless of workload. Since there
were two measures of workload in this study, two regressions were performed to test this
hypothesis. The interaction terms were computed through multiplying the mean-centered
variables. See Tables 55 and 58. The interaction terms in both regressions were non significant.
Therefore, Hypothesis 18 was not supported.
Hypothesis 19 stated that use of flexible arrangements would be positively related to
retention intentions such that those who used more flexible arrangements would be more likely
to indicate an intention to stay with the company. Two regressions using both types of flexibility
(negotiated flexibility and used flexibility) were conducted. See Tables 59 and 60. Used
flexibility was not significantly related to retention but negotiated flexibility (idiosyncratic deals)
2

was related to retention, although the change in R was quite small (.004). Therefore,

150

Hypothesis 19 was supported for idiosyncratic deals (although very weakly) but not for used
flexibility more generally.
Hypothesis 20 stated that satisfaction with flexible arrangements would be positively
related to retention intentions such that who were more satisfied would be more likely to indicate
greater intentions to stay with the company. See Table 61 for the regression testing this
hypothesis. Flexibility satisfaction explains a significant amount of the variance in retention
beyond the control variables and those more satisfied with flexibility are more likely to indicate a
greater intention to stay with the company. Therefore, Hypothesis 20 was supported. A note
about this effect size: although an effect size of .07 may not seem unduly large, in the context of
employee retention, these findings are very important. Even small effect sizes can have
substantial cost implications if it means a more employees are leaving the company for
preventable reasons. This is particularly true for Genericorp, a company that hires mainly high
talent professionals at limited supply.
Hypothesis 21 stated that use of flexibility would be negatively related to emotional
exhaustion such that those who report using flexibility will report lower emotional exhaustion.
Two regressions were conducted to examine this hypothesis using both measures of used
flexibility. See Tables 62 and 63. Neither used flexibility nor negotiated flexibility were
significantly related to emotional exhaustion. Hypothesis 21 was not supported.
Hypothesis 22 stated that satisfaction with flexibility will be negatively related to
emotional exhaustion such that those who are satisfied with flexibility will report lower
emotional exhaustion. See Table 64 for a regression testing this hypothesis. Flexibility
satisfaction showed a significant relationship to emotional exhaustion beyond the control

151

variables, with those who are more satisfied with flexibility showing lower levels of emotional
2

exhaustion (ΔR = .16, p<.01). Therefore, Hypothesis 22 was supported.
Like the effect flexibility satisfaction has on work interference with life, the effect
flexibility satisfaction has on emotional exhaustion is relatively large. This is especially
compelling in light of the fact that Hypotheses 20 and 21 were not supported, such that actually
enacted or negotiated flexibility had no effect on emotional exhaustion. Again, it appears that
the subjective experience of satisfaction with flexibility rather than the objective behavior of
flexibility drives some of the outcomes that are important to work-life balance, such as emotional
exhaustion.
Hypothesis 23 stated that a greater preference for segmentation will moderate the
relationship between flexibility and work-life interference such that those who prefer
segmentation will report higher work-life interference with greater used flexibility. To test this,
the two types of used flexibility were entered into hierarchical regression equations. See Tables
65 and 66. There is a main effect of segmentation preferences on work-life interference such that
those who prefer greater segmentation tend to have greater work interference with life. The
interaction term of used flexibility and segmentation preferences is not significant, however.
Therefore, is not supported.
Hypothesis 24.1 stated that actual integration will moderate the relationship of workload
on work-life interference such that those who integrate work and life will experience more worklife interference regardless of workload. Two regressions were performed to test this hypothesis
using both types of workload (subjective workload and work hours). See Table 67 and Table 68.

152

4

Both the direct effect of actual integration and the interaction term of integration and subjective
workload were significant predictors of work-life interference, although the interaction was not
significant for work hours. To test for this interaction, the variables were centered around the
mean for each variable, crossed with each other. Simple slopes analysis, where overall fit was
regressed on both control variables, confirmed that interaction was significant such that those
who tend to integrate work and home life more had higher work life interference scores
(M=3.42, sd=.75) as compared to those with lower levels of work-life integration and these
effects were stronger at higher levels of workload (M=3.04, sd=.83, b =.244, t(991) = 2.720, p =
2

.007, ΔR = .01). See Figure 21 for a graph of this interaction.
Hypothesis 24.2 stated that an individual’s preference for segmentation would interact
with their actual integration behavior such that those who had a match between segmenting
behavior and preferences would show lower levels of work-life interference. The interaction
effect was tested by crossing the mean centered variables with each other. See Table 69 for a

4

The scope of this paper mainly centers around boundary management behaviors and

satisfaction which is why actual integration is included in the several hypotheses but preferences
for integration is not. Preferences can be an important determinant in behavior. However, given
the relative higher strength of the relationship for the actual behaviors related to boundary
management to the outcomes of interest, focusing on reported actual integration behaviors rather
than the preferred state made sense within the context of this paper. Preferences likely would
make a larger impact in studies where behavior change is being elicited and would help explain
whether individuals were able to change or impact their segmenting or integrating behavior.

153

summary of the results. The interaction term was not significant. Thus Hypothesis 24.2 was not
supported.
Hypothesis 24.3 stated that life flexibility ability would be positively related to work-life
interference such that those who see their personal lives as being less able to flex to
accommodate to work will experience less work-life interference. See Table 70 and 71 for the
two regressions testing this hypothesis (Step 2). Life flexibility ability does significantly predict
work-life interference, but it does so in the opposite direction as hypothesized, although the
2

relationship is weak (β=-.07, ΔR = .01, p<.01). Therefore, hypothesis 24.3 is not supported.
Hypothesis 24.4 stated that life flexibility ability would moderate the relationship
between work-life interference and workload such that those who are less willing for their
personal lives to be permeable to work would experience less work-life interference regardless of
workload. See Tables 70 and 71 for the two regressions (Step 4). The interaction term is not
significant. Therefore Hypothesis 24.4 is not supported.
Hypothesis 24.5 stated that life flexibility willingness would be positively related to
work-life interference such that those who see their personal lives as being less able to be
permeable to work would experience less work-life interference. See Tables 72 and 73 for
regressions of this hypothesis. The main effect of life flexibility willingness is significant,
indicating that those who are more willing to let work permeate the boundaries of their personal
2

life experience more work interference with life (ΔR = .01, p<.01). Therefore, Hypothesis 24.5
was supported.
Hypothesis 24.6 stated that life flexibility willingness would moderate the relationship
between work-life interference and workload such that who are less willing for their personal
154

lives to be permeable to work would experience less work-life interference regardless of
workload. See Tables 72 and 73 for regressions of this hypothesis. The interaction term was not
significant. Therefore, Hypothesis 24.6 was not supported.
Hypothesis 24.7 stated that work flexibility ability would be negatively related to worklife interference such that those who see work as being more permeable to their non work lives
would experience less work interference with life. See Tables 74 and 75 for a summary of the
two regressions that tested this hypothesis (Step 2). The direct effect of work flexibility ability
was significant, explaining 21% of the variance in work-life interference with those who saw
their work as being more permeable to their personal lives as having less work interference with
life. Therefore, Hypothesis 24.7 was supported.
Hypothesis 24.8 stated that work flexibility ability would moderate the relationship of
work-life interference and workload such that those who believe work is flexible to
accommodate to personal life would experience lower work-life interference regardless of
workload. See Tables 74 and 75 for a summary of the two regressions that tested this hypothesis
(Step 4). The interaction term was non significant. Therefore, hypothesis 24.8 was not
supported.
Hypothesis 24.9 stated that work flexibility willingness would be negatively related to
work-life interference such that those who are more willing to allow their non work lives to
permeate their work lives would experience less work interference with life. See Tables 76 and
77 for a summary of the two regressions that tested this hypothesis (Step 2). The direct effect of
work flexibility willingness was significant, indicating that those who are willing to allow their
personal lives to permeate their work roles experiences less work interference with life, although
2

this effect size was small (ΔR = .05, p<.01). Therefore, Hypothesis 24.9 was supported.
155

Hypothesis 24.10 stated that work flexibility willingness would moderate the relationship
of work-life interference and workload such that those who believe work is flexible to
accommodate to personal life would experience lower work-life interference regardless of
workload. See Tables 76 and 77 for a summary of the two regressions that tested this hypothesis
(Step 4). The interaction term was non significant. Therefore, hypothesis 24.10 was not
supported.
Hypotheses 24.11 and 24.12 stated that control of work would be negatively related to
work interference with life such that those with greater control of work will report less work
interference with life and that control of work would moderate the relationship of work-life
interference and workload. See Tables 78 and 79 for the regressions testing these hypotheses.
The direct effect of Control of Work (Step 2) was significant, indicating that people who feel
more in control of their work experience less work interference with life. It is important to note
that the effect size for control of work on work-life interference is quite large, accounting for
35% of the variance in that outcome. Thus, Hypothesis 24.11 was supported. The interaction
term was significant for the interaction of subjective workload and control of work, but not for
work hours and control of work. Simple slopes analysis, where overall fit was regressed on both
control variables, confirmed that interaction was significant such that those who had higher
control of work had lower work life interference scores (M=2.08, sd=.68) as compared to those
with lower levels of control and that these results were stronger at higher levels of workload
2

(M=3.70, sd=.70, b =-.045, t(990) = 4.348, p< .01, ΔR = .01). See Figure 22 for a graph of the
significant interaction. Although this interaction may be small, it fits with Karasek’s (1989)
demand/control model of work, such that those who report higher control over their work are
156

more equipped to deal with higher workloads without it negatively impacting their work-life
balance (e.g. they report lower work-life interference) than their low control counterparts. This
supports Hypothesis 24.12 for subjective workload, but not for work hours.
Hypotheses 24.13 and 24.14 stated that autonomy would be negatively related to work
interference with life such that those with greater autonomy would report less work interference
with life and that autonomy would moderate the relationship of work-life interference and
workload. See Tables 80 and 81 for the regressions testing these hypotheses. The direct effect
of autonomy (Step 2) was significant, indicating that people who felt they had more autonomy at
work experienced less work interference with life. Thus, Hypothesis 24.13 was supported. The
interaction term was significant for the interaction of subjective workload and autonomy but not
for work hours and autonomy. Simple slopes analysis confirmed that interaction was significant
such that those who had higher autonomy had lower work life interference scores (M=2.68,
sd=.94) as compared to those with lower levels of autonomy and that these results were stronger
2

at higher levels of workload (M=3.12, sd=.88, b =-.052, t(991) = 6.941, p< .01, ΔR = .01). See
Figure 24 for a graph of the significant interaction. The effect size for this interaction may be
trivial, as the main effect of workload is quite clear from Figure 23. However, the fact that the
lines diverge for those with low and high autonomy fits with Karasek’s demand/control model
such that those with higher autonomy (similar to those with greater job control) appear to be
better equipped to deal with high workloads such that their work-life balance is less effected by
large volumes of work. This supports Hypothesis 24.14 for subjective workload, but not for work
hours.

157

Hypotheses 24.15 and 24.16 stated that psychological detachment would be negatively
related to work interference with life such that those with greater psychological detachment
would report less work interference with life and that psychological detachment would moderate
the relationship of work-life interference and workload. See Tables 82 and 83 for the regressions
testing these hypotheses. The direct effect of psychological detachment (Step 2) was significant,
indicating that people who felt they had more psychological detachment from work experienced
less work interference with life. Thus, Hypothesis 24.15 was supported. The interaction term
(Step 4) was significant for the interaction of subjective workload and psychological detachment
but not for work hours and psychological detachment. Simple slopes analysis confirmed that
interaction was significant such that those who had higher psychological detachment had lower
work life interference scores (M=2.30, sd=.74) as compared to those with lower levels of
detachment and that these results were stronger at higher levels of workload (M=3.64, sd=.74, b
2

=-.145, t(991) = 4.285, p< .01, ΔR = .01). See Figure 24 for a graph of the significant
interaction. Again, although the effect size of this interaction is small, it represents an important
insight into the protective effects of psychological detachment, even in the fact of high
workloads. Those who are able to detach from work during non work time show greater ability
to sustain high levels of workload without negative effects to their balance. This supports
Hypothesis 24.16 for subjective workload, but not for work hours.
Hypotheses 24.18 and 24.19 stated that evening work style would be positively related to
work interference with life such that those who work more in the evenings will report more work
interference with life and that evening work style would moderate the relationship of work-life
interference and workload. See Tables 84 and 85 for the regressions testing these hypotheses.
158

The direct effect of evening work style (Step 2) was significant, indicating that people worked
more in the evenings experienced more work interference with life. Thus, Hypothesis 24.17 was
supported. The interaction term (Step 4) was significant for the interaction of subjective
workload and evening work style but not for work hours and evening work style. Simple slopes
analysis, where overall fit was regressed on both control variables, confirmed that interaction
was significant such that those who tend to work in the evenings higher work life interference
scores (M=3.4, sd=.80) as compared to those with lower levels evenings work and that these
results were stronger at higher levels of workload (M=2.50, sd=.79, b =.069, t(984) = 8.048, p<
2

.01, ΔR = .01). See Figure 25 for a graph of the significant interaction. This supports
Hypothesis 24.18 for subjective workload, but not for work hours.
Hypotheses 24.19 and 24.20 stated that weekend work style would be positively related
to work interference with life such that those who work more on the weekends will report more
work interference with life and that weekend work style would moderate the relationship of
work-life interference and workload. See Tables 86 and 87 for the regressions testing these
hypotheses. The direct effect of weekend work style (Step 2) was significant, indicating that
people worked more on the weekend experienced more work interference with life. Thus,
Hypothesis 24.19 was supported. The interaction term (Step 4) was significant for the
interaction of subjective workload and weekend work style but not for work hours and weekend
2

work style (ΔR = .02, p<.01). Simple slopes analysis confirmed that interaction was significant
such that those who work on the weekends had higher work life interference scores (M=3.42,
sd=.83) as compared to those who do not work on the weekends and that these results were
2

stronger at higher levels of workload (M=2.49, sd=.85, b =.100, t(991) = 8.254, p< .01, ΔR =
159

.01). See Figure 26 for a graph of the significant interaction. This supports Hypothesis 24.20 for
subjective workload, but not for work hours.
Hypotheses 24.21 and 24.22 stated that vacation work style would be positively related to
work interference with life such that those who work more on vacation will report more work
interference with life and that vacation work style would moderate the relationship of work-life
interference and workload. See Tables 88 and 89 for the regressions testing these hypotheses.
The direct effect of vacation work style (Step 2) was significant, indicating that people worked
more on vacation experienced more work interference with life. Thus, Hypothesis 24.21 was
supported. The interaction term (Step 4) was significant for the interaction of subjective
2

workload and vacation work style but not for work hours and vacation work style (ΔR = .01,
p<.01). Simple slopes analysis confirmed that interaction was significant such that those who
work on vacations had higher work life interference scores (M=3.49, sd=.77) as compared to
those who do not work on vacations and that these results were stronger at higher levels of
2

workload (M=2.31, sd=.79, b =.090, t(991) = 8.333, p< .01, ΔR = .01). See Figure 27 for a
graph of the significant interaction. This supports Hypothesis 24.22 for subjective workload, but
not for work hours.
The interaction terms for all three behavioral detachment styles (evening, weekend and
vacation) had small effect sizes such that they indicate the main effect of workload on work-life
interference is quite strong. It is also important to note that the supplemental analysis done with a
more integrated regression analysis in which all the detachment and workload variables were
considered within the same model did not yield the same result for vacation and evening work
style interactions in that these were no longer significant (see Appendix F). However, it is
160

important to note that there do appear to be small differences between people who disengage
from work while not on work time in being able to sustain larger workloads without detriments
to their balance. This is a theme seen throughout the interactions in this section and points to an
interesting trend that may hold promise for individuals looking for techniques to improve their
work-life balance.
Group Normative Theory Hypotheses
Hypothesis 25 stated that manager work-life interference would moderate the relationship
of subordinate workload and work-life interference such those with managers with more worklife interference would experience more work interference themselves regardless of level of
workload. It is important to note that the power for this hypothesis was greatly reduced as the
number of participants who also had managers who provided responses was much smaller than
the overall response rate (N=376). See Tables 90 and 91 for the regressions examining these
hypotheses. Manager work interference with life did not have a main effect on individual work
interference with life nor was the interaction of manager work interference with life significant.
Therefore, Hypothesis 25 is not supported.
Hypothesis 26 stated that perceptions of manager expectations about work-life balance
will moderate the relationship between work-life interference and workload. To test this
hypothesis I looked at manager support as moderating the relationship of the two types of
workload on work-life interference. See Tables 92 and 93 for summaries of the regressions.
Although there is main effect for manager support of work-life interference such that those with
2

more supportive managers experience lower levels of work-life interference (ΔR = .11, p<.01).,
the interaction term is not significant. Therefore, Hypothesis 26 was not supported.
161

Hypothesis 27 stated that team work-life interference will moderate the relationship of
individual workload and work-life interference. However, as the HLM models presented earlier
(see Table 36) reveal, work-life interference does not demonstrate qualities of a group level
variable as the ICC(1) value is too low. Therefore, it is impossible to test this hypothesis with
this data set.
Hypothesis 28 stated that perceptions of team work-life balance norms will moderate the
relationship between individual workload and individual work-life interference such that those in
teams with norms that are more supportive of balance will have lower work-life interference. In
order to assess group norms for balance, I assessed three variables for group level agreement of
work-life balance norms: Team instrumental support of balance, team emotional support for
balance and Work flexibility ability. As with the prior group level analyses, I only used a subset
of the data—only groups with 60% or more of member providing surveys were included in
analysis (n=510). See Table 94 for the HLM unconditional means models of these scales. Each
of these measures met criteria for aggregation of ICC(1) values of greater than .08. However,
only Team emotional support and Work flexibility ability had significant Wald statistics. As
Bliese’s (2000) guidelines for ICC(1) thresholds indicate that an ICC of .08 is marginal, and
given the non-significant value of the Wald statistic, Team instrumental support does not meet
criteria to be aggregated. For the other two variables, however, each participant was assigned the
aggregated, average value of their entire team for each of these scales to get at the team norms
for balance.
To test how team norms might interact with workload, six regression analyses were
conducted with the two aggregated scales and the two types of workload predicting individual
level work-life interference. See Tables 95, 96, 97, and 98. Aggregated work flexibility ability
162

2

produced a main effect on individual work-life interference (ΔR = .06, p<.01) as did team
2

emotional support (ΔR = .01, p<.01). Additionally, those in teams where the group agreed that
there were able to flex personal life into work time had lower work-life interference. The
interaction terms in all 6 regressions were not significant, however. Thus, hypothesis 28 was not
supported.

Discussion Study 3
Study 3 investigated many targeted questions-- 48 hypotheses in all-- and found effects of
varying sizes for 28 of the total hypotheses. For a summary of the findings from Study 3 and
Study 1, see Table 1.
Study 3 supported many direct effect hypotheses, several with moderate to strong effects.
Hypotheses having to do with workload, emotional exhaustion and work-life interference
showed particularly strong effects (Hypotheses 1, 8 and 9). Workload, both in terms of
subjective workload and average work hours reported, was strongly related to both work-life
interference and emotional exhaustion, although subjective workload was more strongly related
to both of the variables than work hours. The strongest observed relationship, however, was
between work-life interference and emotional exhaustion, with work-life interference explaining
40% of the variance in the emotional exhaustion variable. The general discussion with delve
more into this relationship and discuss possible reasons that this population in particular may be
more likely than others to show particularly strong linkages between work-life imbalance and
emotional exhaustion.

163

Additional support was found for some supplementary hypotheses, particularly regarding
detachment. Psychological detachment, the tendency for an individual to psychologically
disconnect from work during non-work times, was highly related to work-life interference such
that those who reported greater detachment also reported lower work-life interference.
Behavioral detachment, the extent to which individuals do or do not report engaging in work
related tasks during non-work times, was also investigated and this showed moderate to strong
direct effect sizes as well on work-life interference.
In addition to direct effects, psychological and behavioral detachment indicators also
showed significant, albeit small, interactions with subjective workload. The consistent finding of
these interactions is that detachment appears to act as a buffering mechanism for workload such
that those who are able to psychologically detach and/or physically from their work during nonwork times are better able to tolerate relatively higher levels of workloads without it impacting
their satisfaction with work-life balance than those who are less able to detach. Although these
interaction findings are small, they are important in that they indicate a potentially valuable
intervention point for individuals who wish to improve their work-life balance but do not wish to
reduce their workload.
These findings on detachment are in line with the COR theory (Hobfoll, 1989) and also in
line with other research in the work-life balance field that emphasizes the importance of rest and
recovery on work-life balance satisfaction (Sonnentag & Bayer, 2005; Zijlstra & Sonnentag,
2006). It is possible that detaching from work enables a replenishing of resources for individuals
that those who stay constantly “plugged in” do not benefit from, even given similar subjective
feelings of workload. However, these findings represent a key contribution of this study, as

164

previous research has not demonstrated the interactive effects of psychological or behavioral
detachments and workload on work-life balance outcomes.
Another interesting set of findings from Study 3 that represents a key contribution of this
paper are those that have to do with life and work flexibility ability and willingness. Although
the hypothesized relationships (Hypotheses 24.2, 24.4, 24.6, and 24.8) predicted that there would
be complementary and mirror effects of work and life flexibility willingness and ability on worklife interference, the findings suggest that it is mainly work flexibility perceptions that matter,
with the ability to flex work commitments having a stronger relationship to work-life
interference outcomes than the willingness to flex work and both of these having a stronger
effect than family flexibility perceptions. This finding is not entirely surprising given the
unidirecitonality of the outcome measure that looks at the impact of work on life and not the
impact of life on work. Future studies should investigate whether the life ability or willingness to
flex to accommodate work is more related to life interference with work as an outcome.
However, to date, this is one of the first studies to investigate these four flexibility perceptions at
the same time with particular emphasis on the work domain and provides insight into the
importance of perceptions of ability vs. willingness that will be further explored in the general
discussion (Jin, Ford & Chen, 2013; Winkel & Clayton, 2010).
Related to perceptions of ability to be flexible were the hypotheses tested in Study 3 that
directly investigated satisfaction with flexibility and actual use of flexibility supports. In an
effort to tease apart the relative importance of behavior flexibility and the more subjective
satisfaction with flexibility, Study 3 asked participants to rate both general flexibility satisfaction
as well as indicate whether they partook in a variety of flexible work arrangements. The findings
from these hypotheses (Hypotheses 15, 16, 17, 19, 20) show that satisfaction with flexibility
165

appears to be the most important factor in driving the relationships between flexibility and
outcomes like work-life interference and retention intentions. Actual use of flexibility work
arrangements show very small effects in comparison.
The hypotheses in Study 3 that were unsupported mainly had to do with performance or
were hypothesized interactions. The findings related to performance are similar to those in
Study 1, where relationships between variables and performance were also not supported or were
very small. The general discussion will dive deeper into some of the reasons for this, but this
may be partly due to the face that the measure of performance used in Study 3 shows little
variance and is likely range restricted, like the measure used in Study 1. With a standard
deviation of just .25, there may not have been enough variance in the measure itself to
adequately distinguish people along a performance spectrum such that effects of other variables
could be determined. However, given the fact that other researchers have investigated this area
and have also found null or ambiguous results (Beauregard & Henry, 2009), it appears that work
life balance may contribute very little explanatory information to performance variables.
Another area where the hypotheses were generally unsupported in Study 3 had to do with
the a priori hypothesized interactions of manager and team support. These hypotheses
(Hypotheses 11, 12, 13 and 14) suggested that perceptions of support from the team or manager
would moderate the effect of individual workload on individual work-life interference. It is
important to note, however, that although it was not specifically hypothesized, perceptions of
both manager and team support for work-life balance did have a direct effect on individual worklife interference beyond what could be attributed to workload. So although the relationships
between these types of support and workload were not interactive (e.g., this support was not
more or less valuable at different levels of workload), it appears to be a direct effect that may
166

help add up to a more satisfied individual. One of the reasons these interactions may not exist is
that the main effect of workload on work-life interference is quite strong, such that this effect
may swamp any of the potential mitigating crossed effects of supportive managers or teams. This
finding is in line with what other researchers in the area have found, in that workload can act as
the driving force behind work-life balance and interference such that at higher levels of
workload, other effects are diminished because it is just the workload that appears to be
contributing the difficulties with work-life balance (Fritz & Sonnentag, 2006).

Strengths and Limitations of Study 3.
Study 3 had several key strengths as well as a few important limitations. First, for
strengths, the sampling strategy allowed for team analyses of some of the variables. Although the
moderated team hypotheses were not supported, there were direct effects for team level cultural
work-life balance support variables, and such findings would be impossible without sampling
intact teams. These teams also frequently had managers reporting information as well, so that a
more comprehensive picture of how work-life interference functions within the context of a work
group could be better teased apart.
A second strength is the focus on attempting to tease apart specific mechanisms for both
detachment and flexibility. Often studies rely on a single measure of flexibility or detachment
when investigating these phenomenon (Hill, Erickson, Holmes, & Ferris, 2010; Richman,
Civian, Shannon, Hill, & Brennan, 2008). This can make it difficult to try to understand how
different types of flexibility (e.g. used, negotiated or satisfaction with) can have different effects
on desired outcomes. By coming at the constructs of flexibility and detachment from a variety of

167

perspectives, it was possible for Study 3 to help provide insight into a more nuanced
interpretation of what exactly is important for work-life balance boundary management.
Although the ability to tease apart these relationships is a key strength of Study 3, it also
opened a potential problem from an analytical standpoint. The sheer volume of individual
statistical tests needed to test each effect and hypotheses was large and with so many analysis
conducted, the possibility of making a Type I error increases. If 100 analyses were conducted,
using a significance standard of .05 could result in five supported hypotheses by pure chance.
All attempts were made to decrease the number of individual analyses run by combining
analyses where appropriate, so in total 69 regressions were performed. Of these, 32 were
significant, a much larger proportion than we would expect to see by chance. The hypotheses
were also all based on other research, theory or interview data, so all attempts were made to
make the questions asked as focused as possible despite the layers of complexity. However, the
statistical implications of significance are important to remember in large investigations such as
this and one reason why fishing through the data for significant findings is ill advised.
Given this, however, it is important to keep in mind the design of this paper was
intentional and the multitude of statistical tests arose from the goal of understanding how
variables that are similar but distinct affect outcomes in different ways. Without examining three
different types of flexibility, for instance, it would have been impossible to pinpoint the
particular importance of flexibility satisfaction in the array of types. In all projects, decisions
surrounding parsimony and consolidation are made. For this paper, the goals were centered
around attempting to understand a broader contextual space as the potential cost of consolidating
constructs and testing methods.

168

A third area where Study 3 has added to the literature is in the area of segmentation and
integration. Study 3 investigated both preferences for segmentation (how much a person prefers
drawing a clear line between work and personal life) and how much a person feels they actually
integrate work in their lives. Previous research has suggested that whether one segmented or
integrated their work and personal lives was not important, it only mattered whether there was a
match between preferences and behaviors. Study 3 did NOT support this hypothesis, however.
Instead, the direct effect of actual integration on work-life interference was much more powerful
than preferences (See Table 1 Hypothesis 23, 24.1, 24.2), and this direct effect seemed to swamp
any effect preference had on work-life interference. For integrating work and personal life, it
appears to be less about what you prefer and more about what you actually do.
The main weakness of Study 3 is a weakness shared by all three studies, and that is that
they were all conducted with employees from the same company. The benefits of having a
cooperating company provide access is that many interesting and potentially difficult to obtain
variables (such as performance or team dynamics) are possible to measure with organizational
support. However, this decreases the possibility for generalizability beyond the specific
organizational context. Genericorp is also unique in that is represents a specific employment
market niche, on that focuses employing professional, educated and high skilled workers.
Therefore the findings of Study 3 (and all three studies) need to be interpreted cautiously and not
over applied to other organizations or working conditions to which they may not apply.
Study 3 also uses a cross sectional, survey design which makes it impossible to infer
causal direction of effects. There are many relationships hypothesized in the Study 3 that are
likely reciprocal in nature (such as the relationship between work-life interference and emotional
exhaustion, and the relationship between workload and these two outcomes). By just asking
169

people to report information at a single point in time, it is impossible to discern what causes
what. Future studies in the area should make an attempt to study these variables more
longitudinally to ascertain whether work-life interference leads to emotional exhaustion or if
effects are more reciprocal, or if a third variable is contributing to both. Some researchers in the
field have begun this work (Sonnentag & Bayer, 2005; Zijlstra & Sonnentag, 2006, Binneweis et
al, 2009), but also call for further research with longer time horizons than typically studied in
experience sampling studies to better capture potential causal relationships.
Another potential weakness of this study is in the methods used to analyze the data.
Bivariate regressions were largely utilized so that a parsimonious understanding of each
relationship could be parsed and this led to a more focused study of each individual relationship.
This process also allowed for the use of HLM where appropriate and to test for nested effects.
However, this process can lose the systems viewpoint other statistical processes may lend and
future studies may be encouraged to analyze similar data using an SEM or other path model
approach.

170

CHAPTER 5: DISCUSSION

General Discussion
Taken together, these three studies make a number of important contributions to the
literature on work-life balance, boundary management and flexibility and the importance of
manager and teams in individual work-life balance experiences. To fully examine what can be
learned from the synthesis of all three studies, this discussion section will be organized around
themes. These themes mirror those proposed in the introduction and are tied to the hypotheses
that thread throughout all three studies and are comprised of Flexibility, Border Theory and
Boundary Management, Control and Autonomy, Social Support, Culture and Team Norms, and
Emotional Exhaustion and Retention. The final section of this discussion will touch on
weaknesses to bear in mind during interpretation of these results, as well as suggestions for
future research and practical implications.

Flexibility
One of the areas where this paper has the opportunity to provide the greatest contribution
to the literature is in the area of flexibility and its relationship to work-life balance satisfaction.
Previous researchers have demonstrated that flexibility is indeed important to work-life balance
satisfaction, although there is debate over what kinds of flexibility organizations should seek to
maximize to help employees achieve the best work-life balance outcomes. Some researchers
believe that flexible behaviors should be maximized for those who need them (Kossek & Ozeki,
1999), while others believe that explicit contracts should be made with managers about flexible
arrangements (Rousseau, 2005), still others are focused on general satisfaction with flexible
171

options (Richman et al, 2008) and a culture of flexibility support is emphasized by others
(Kossek & Lambert, 2012).
This paper sought to gain a complete picture of how individuals experience a flexible
workplace and to that end approached flexibility from a variety of standpoints. In the first part of
this section, the relative effects of flexibility satisfaction will be compared to used flexibility and
what those implications may be for organizations and individuals. The second part of this section
will focus on individual perceptions of flexibility access and willingness to use flexibility, and
what can be learned from how these differences play out with regard to the outcomes of interest.
Flexibility Satisfaction vs. Used Flexibility. Study 3 included several hypotheses
(Hypotheses 14 – 20) related to flexibility and investigated both individual satisfaction with their
levels of flexibility and their actual use of flexible arrangements. Used flexibility looked at both
reported use of flexible options such as working from home and explicit negotiation with
managers for these flexible arrangements (I-DEALS). Satisfaction with flexibility asked
individuals to report in general how satisfied they were with their work flexibility. Although the
moderating effects of flexibility were not found, the different effect sizes for satisfaction with
flexibility and use of flexibility is revealing. Used flexibility (both negotiated and reported) has a
very small effect on work-life interference as compared to satisfaction with flexibility.
Study 2 can help provide context for this finding. As flexibility was an area that was
explicitly called out in the interview script, all interviewees were asked about their flexible work
arrangements and whether they felt they had the option to work from flexible locations and if
they took advantage of these opportunities. Findings from Study 2 show that satisfaction with
and use of flexibility work arrangements did not necessarily go hand in hand. Some of the
interviewees who felt satisfied with their flexibility and felt as if working from flexible locations
172

and schedules would be supported actively chose not to, preferring more regimented schedules
and locations to do their work. Managers in particular felt that the environment was supportive of
flexible work arrangements, but that their role may not be the best fit for taking advantage of
options such as working from home, something one manager in particular said was reasonable
and part of the job responsibilities he had agreed to.
The stronger effects of satisfaction with flexibility than actual use of flexibility may be
related to the findings of Hypotheses 24.10-24.13 which have to do with control. Satisfaction
with flexibility shares a moderate correlation with control of work, while actual use of flexible
arrangements is uncorrelated with the variable. If one is satisfied with their flexibility at work,
then they may also feel they have control over how they organize their work time, even if they
choose not to exercise specific flexible work arrangements. Indeed, these two variables are
correlated. Further findings related to control and autonomy will be explored later in the
discussion.
Perceptions of Flexibility Access. In addition to asking participants whether they were
satisfied with or used flexible work arrangements, Study 3 sought to more fully understand how
individual’s perceived their ability and willingness to flex their personal and work lives to
accommodate the needs of the other role. Hypotheses 24.2-24.9 investigated the relationships of
life and work flexibility ability and willingness on work-life interference. Although moderation
hypotheses were not supported, the direct effects of life flexibility and work flexibility
perceptions are interesting and asymmetrical when considering work-interference with life.
Hypothesis 24.2 found that when individuals feel as if their life does not have the ability
to flex for work demands, they actually experience slightly lower work-life interference than
those who perceived a greater ability for flexibility. Although this is in the opposite direction to
173

the hypothesized relationship, it may make sense when considering the variables involved. The
extant research in this area is divided in how it predicts perceptions of ability and willingness to
be related to work-life balance outcomes. Some researchers suggest that if roles are perceived as
more able to flex to the needs of the other role, then the individual will experience less conflict
because intrusions of one role into the other are not so disruptive (Winkel & Clayton, 2010).
Others suggest that those that perceive that the boundaries are less permeable, or less able to flex
to accommodate the needs of the other role, will experience less work-life interference because
they set up firmer boundaries between work and life and prevent the intrusions from occurring
(Matthews, Barnes-Farrell & Bulger, 2010). However little research has been done in this area to
date (Matthews, Swody & Barnes-Farrell, 2012 ). It would appear that for Study 3, the second
option is the most likely explanation. Those who perceive their personal lives as being less able
to flex to work demands may actually be preventing intrusions from work into the life domain
which may decrease work to life interference. This may also be related to increased detaching
behaviors, which are strongly related to satisfaction with work-life interference and will be
discussed more in depth below.
The other direct effects of flexibility willingness and ability to flex boundaries between
work and life were in line with hypothesized relationships. Individuals who felt more willing to
flex their personal lives to accommodate work demands experienced more work interference
with life, although this effect was small. The stronger effects had to do with ability and
willingness to flex work for personal life demands. Individual perceptions about how possible it
was for their work to accommodate their personal life were particularly relevant to work-life
interference, with those who felt that they worked in situations that could not flex for personal
demands reporting higher levels of work-life interference. An individual’s willingness to stretch
174

the work boundary to accommodate personal life demands was also related to work-life
interference, with those less willing to flex this boundary reporting more interference, although
this effect size was smaller.
Although Study 3 only considered one direction of work-life life interference (e.g. the
work to life direction), it is interesting that an individual’s perceptions of the permeability of the
life domain have such relatively small effects on work-life interference compared to an
individual’s perceptions of work-boundary permeability. Much of the theory in this area treats
work and life as symmetric roles with similar boundary structures (Greenhaus & Beutell, 1985).
However, this appears not to be the case. Even looking at mean differences between the
willingness and ability measures, participants in Study 3 feel work is more able to flex than
personal life, and are also more willing to do so. The effects on feelings of imbalance on not
feeling able to flex work, however, are stronger than those of not being able to flex personal life.
The interview findings from Study 2 support this result. It appeared that for many
interviewees, work frequently flexed into non-work time, but there were rarely times when non
work flexed back. For times when non-work did flex into the work domain (mainly in the case of
parents), these interviewees generally were not as bothered by at as they were when work would
unexpected intrude on non-work time. One interviewee expressed this clearly in the following
quote:
“So if I have a project, especially with these 10-hour [self-guided trainings] that we have
to take, it’s a nighttime thing. I’m never going to do that during the work day. And
occasionally, once in a blue moon I’ll say to myself “Okay because I’m going to do this
[training program] at night, I’m going to go off campus during lunch and do an errand.”
175

Once in a blue moon I actually make that happen, usually it ends up being a full work
day and then three hours at night. I think it might be because work stuff is always more
pressing. There are many more deadlines in your work life than in your personal
life… And also I think that there’s a mindset that if it’s between 9 and 5 that you
shouldn’t be doing something that’s kid focused or personal focused.” Participant 29

This finding adds to the literature because it calls out what may be a fundamental
difference between how people view the boundaries of work and non-work roles. The non-work
outcomes such as general life or family satisfaction, personal life stressors or other personal
goals were outside the scope of this paper so were not explored and could be considered a
drawback of this study. Further research is needed to more clearly tease apart how perceptions of
boundary permeability in the work and non-work domain may affect work specific outcomes or
non-work specific outcomes so see if asymmetric effects can be observed for those outcomes.
For this study, the finding that perceptions work flexibility ability is most strongly tied to
work-life flexibility fits in with findings related to flexibility satisfaction and control (to be
discussed later in this paper). It stands to reason that individuals who feel as if their work roles
allow for more freedom of choice to exercise flexibility would also feel more in control over
their schedules and work, as well as feel more satisfied with their flexibility.

Border Theory and Boundary Management
Another area where this paper offers contributions to the literature is further clarifying
the relationship between individual preferences for segmentation or integration and how
individuals are able to actually enact these preferences. Several researchers have theorized that it
176

is not the actual behavior of segmenting or integrating that results in problems with work-life,
but rather a mis-match between preference and reality (Ashforth, Kreiner, & Fugate, 2000,
Desrochers & Sargent, 2004). Empirically, this hypothesis has been supported in a couple of
research studies (Kreiner, 2006, Rothbard, Phillips, & Dumas, 2005).
In this paper, however, match between preference and actual segmentation/integration
behavior did not predict work-life interference or emotional exhaustion. What mattered with
regard to these two outcomes was actual segmentation behavior. People who reported that they
were able to leave work at work, that is, those who were actually enacting segmenting behaviors,
reported lower levels of work-life interference and emotional exhaustion regardless of their
stated preferences, compared to those who integrated their work and their lives. Another
compelling finding is that actual integration moderated the relationship between subjective
workload and work-life interference such that those who integrated their work and non-work-life
experienced higher levels of work-life interference at both high and low subjective workload
levels.
Although Study 3 was able to provide the most nuanced exploration of segmentation
preferences and behaviors, the thread of more detachment linking to greater work-life balance
satisfaction was seen in Study 2 as well. The interviewees who reported greater problems either
mentally or physically detaching from work also tended to report greater dissatisfaction with
balance, whereas those with set routines that included times that they were fully disengaged from
work generally had better work-life balance outcomes. Also interesting to note is that several on
the interviewees who were struggling with work-life balance satisfaction mentioned that one way
they thought they might be able to improve their satisfaction was through more effective

177

detaching behaviors (e.g. not checking emails on weekends or evenings, closing the computer at
night, detaching on vacation, etc.).
Another key contribution of this paper is that detachment was investigated in similar
ways to flexibility, with efforts being made to look at detachment from a variety of angles. In
Study 3, detachment was investigated through probing at psychological detachment (the degree
someone can mentally disengage from work when not working), preferences for segmentation
(how much they prefer to separate their work and personal lives), actual segmentation (how
much they report they actually do separate their work and personal lives), and then behavior
indicators of work and personal life satisfaction, as indicated by working during evenings,
weekends and vacations. The behavioral detachment measures were added explicitly due to
findings from Study 2 interviews that indicated a large proportion of interviewees were engaging
in work during non-work time, particularly those who were experiencing dissatisfaction with
their balance.
All types of detaching had effects on work interference with life, although it is clear that
boundary management preferences have smaller effects on the outcomes of interest than the
more behaviorally or psychologically anchored ones. Unsurprisingly, people who report
difficultly psychologically detaching from work also report higher behaviors that would indicate
that they are physically not detaching from work as well, such as checking email during personal
time, working while on vacation, etc.
Taken together, these findings provide support for an alternative theory of
integration/segmentation than the dominant one in the current literature. The alternative
proposition is that fit between preferences and actions matter less than behaviors themselves
when it comes to detaching from work. The findings from Study 3 and Study 2 support a
178

conclusion that integrators are, in general, more at risk of experiencing problems with work-life
balance than those who segment work and life. This is consistent with some recent findings by
other researchers, who have found similar results (Olson-Buchanan & Boswell, 2006). This is
important because it suggests a possible point of intervention for people who may be struggling
with achieving the work-life balance they want. Rather than advising people to try to match their
integration/segmentation behaviors to their preferences, it may be better to advise them to try to
draw firmer boundaries between work and non work to achieve a more satisfactory balance.
This suggests that organizations wishing to help employees better manage work-life balance
could institute programs that explicitly encourage psychological detachment behaviors and
stopping the physical work engagement behaviors that many also display, even if the employee
does not explicitly prefer segmentation to integration as a work management style.
When considering segmentation/integration, detachment and work-life balance it is
important to keep in mind, however, that it is impossible to discern the causal direction of the
variables within the context of a cross sectional study such as this one. It is unlikely that the
ability to detach is a characteristic that is entirely within behavioral control. There are likely
personality aspects to it as well job role constraints that keep some employees more “plugged in”
than others (Sonnentag & Kruel, 2006). The findings of this paper indicate that detachment can
serve as a buffer between higher workloads and negative outcomes such as work-life interference
and emotional exhaustion, but other researchers in the field have found that at the highest levels
of workload, the workload itself swamps all efforts to disengage and thus no benefits from rest
and recovery can be observed (Benneweis, Sonnentag & Mojza, 2009). This is definitely
something to keep in mind if looking to implement a program with employees with very high
workloads.
179

From the individual difference standpoint, there is also no research to indicate that
training those who prefer to integrate to become segmentors in their work-life styles would reap
the same benefits that are observable in Study 3. It is likely that those who currently segment
work and life, at least to some extent, have preferences for those behaviors regardless of what
they report as their preferences as they are actually carrying out the detachment behaviors.
Future research could measure the effectiveness of an experimental intervention in which those
who prefer to integrate and those who prefer to segment are encouraged to follow behavioral
plans that either encourage segmentation or integration. This would be a better method for
observing whether an intervention encouraging more segmentation would work across all
employees, regardless of individual difference preferences.
Given the very strong relationships between psychological detachment, work-life
interference and emotional exhaustion, it is important to continue to investigate the reasons why
and how individuals do and do not detach from work. The interviews from Study 2 illuminated
that many people felt the desire to detach but were having difficulty turning that desire into
actual behavior. This question drove the investigation into control and autonomy in Study 3.

Control and Autonomy
Control and autonomy over one’s work emerged as an important aspect of individual’s
experience of work-life balance during the interviews in Study 2. Many of the interviewees
discussed how managing work-life balance was their responsibility as individuals, but that they
were experiencing difficulty gaining control over their schedules and the demands of their jobs.
This led to the development of supplementary hypotheses 24.10-24.13 in Study 3 to specifically
investigate the relationship of feelings of control over one’s work on work-life interference.
180

The findings indicate that control of work had a large and autonomy had a small
relationship with work-life interference, such that those who felt they had more control over their
work experienced less work-life interference. This is consistent with the qualitative findings
from Study 2, where those who felt more in control of their work appeared to experience better
outcomes. During the interviews in Study 2, several interviewees mentioned how unexpected
work, or “firedrills”, would often come up during their workdays, which would throw off their
work plans and schedules and make it so that they would have to work late or overhaul their
schedules to get the work done that they had originally planned. Those that had more predictable
schedules, or felt that the boundaries and limits they placed on their schedules were more
respected, were more satisfied both with their work-life balance and with their work in general.
The aspect of control is a very interesting one in how it is related to detachment as well.
For those that feel that their schedules are out of their control and that there is little they can do
about it, it would be difficult to feel comfortable fully disengaging from work on the chance a
crisis popped up that they would be responsible for solving.
One of the more interesting findings around control of work is that it is not correlated
with job level. Although it seems logical that as a person gains organizational prestige and
responsibility that their control over their work would increase, that is not the case at Genericorp.
This leaves the reasons behind why individuals feel that they do or do not have control over their
work unclear. What makes it possible for the lowest level employee to feel that they have equal
control over their schedule as a higher level manager, who in essence should be arranging the
schedules and workloads for others? Research in other areas has indicated that perceived work
control (as compared to actual control) of situations has strong relationships to desired
outcomes, including decision-making, engagement and stress (Kahnweiler &Thompson, 2000;
181

Sprung & Jex, 2012), although there has been little research to date linking control to work-life
balance variables (Hsu, 2011). Future research should more closely examine the role control over
one’s work plays both in being able to detach from work and in satisfaction with work-life
balance.
Another open question is to what degree is control of work is a personality variable or is
tied to job responsibilities. The findings from Study 2 would suggest that it is a little bit of both,
with those showing the greatest control over their work adopting an attitude that is markedly
different from their less controlled counterparts. The individuals who exerted more control over
their work and lives (and generally erected rather strict boundaries between the two), took a
perspective that their work quality would speak for itself in a positive way and that they did not
have to compensate with quantity, that they had made a conscious decision to live their life in a
certain way with certain priorities and that they were arranging their time and schedules
accordingly. Those with less control appeared to have not made such conscious choices and
were more reacting to whatever stimuli they experienced in their environment, so whatever new
crisis came up in their work-life derailed their plans and impacted their satisfaction with balance.
For some of the interviewees, this reactiveness was likely unnecessary for their roles, although it
seemed necessary at the time.
It is possible this reactivity is related to proactive personality, a personality variable that
is associated with identifying and acting on opportunities, showing initiative and showing a
disposition toward movement and growth (Seibert, Crant & Kraimer, 1999). Researchers have
shown proactive personality to be related to high achievers and those seeking strong career
trajectories (Seibert, Crant & Kraimer, 1999), which given the sample used in this paper may
make this personality trait unusually prevalent. Most of the research to date on proactive
182

personality has explored the positive outcomes of this personality trait (e.g. increased
occupational citizenship behaviors (Gan & Ceung, 2010), greater career commitment (Seibert,
Crant, & Kraimer, 1999), greater creativity (Kim, Hon & Lee, 2010). It would be interesting to
investigate this further in the future looking at proactive personality as a possible precursor for
reduced work-life balance satisfaction or increased emotional exhaustion.
However, for two or three of the interviewees, being reactive was explicitly tied to job
responsibilities that could be critical to Genericorp and could have far ranging implications if a
particular crisis was not dealt with in a timely manner. This is where there is likely an
interaction of job responsibility and individual difference occurs. Future research in this area
should more closely examine the drivers of perceptions of work control, and how to encourage it
in individuals who might be experiencing different barriers to control. Practical implications
from the control findings are similar to those for detachment and include encouraging individuals
to be more mindful about how they allocate their time and their energy to ensure that it aligns
with they want to spend their time.

Social Support
Study 2 revealed, and Study 3 supported, that for some individuals both detachment and
work control appeared within the ability of the individual to regulate while others felt more at the
mercy of their schedules and work demands. One of the reasons for the differences in these
perceptions could be due to the differences in environments individuals experience in work and
life settings. An environmental aspect that impacts individuals differently is social support. All
three studies looked at manager social support and Study 2 and Study 3 took a closer look at
team and personal life social support.
183

Managers. Studies 1 and 3 investigated whether having a manager with poor work-life
balance had an additive effect on individual work-life balance. The theory behind this was that
managers as leaders for a team set the norms for the group, which can contribute to group effects
for poor work-life balance. Neither Study 1 nor Study 3 found an interaction effect for manager
work work-life balance effecting subordinate work-life balance, however.
In trying to understand this non effect, the results from all three studies pointed to a
possible explanation. Being a manager is associated is having lower satisfaction with work-life
balance in general, across all three studies. Even in the qualitative interviews, the majority of
those interviewed mentioned that managers tended to not be good role models for a wellbalanced life and that it was difficult to find leadership examples of work-life arrangements they
would want to emulate. It is possible that any cross-level effects are being swamped by the
stronger direct effect of the fact that managers themselves just have lower satisfaction with
balance. Additionally, from the interviews in Study 2 it became clear that satisfaction with
balance was, at least to some extent, driven by personal preferences and mindset. This variation
at the individual level could further decrease any effects manager work-life balance may have on
their subordinates.
One potential weakness of this paper is that in both Study 1 and 3 manager work-life
balance was measured through self-report of the manager and linked downward to their team
through hierarchy reporting relationships. That is, how managers themselves perceive their
balance is being linked to how their subordinates perceive their own balance. When thinking
about norms, a potentially more illuminating question to ask is how do subordinates view their
manager’s work-life balance and look at the resulting effects on subordinate work-life balance.
As Study 2 revealed, individual differences in preferences for work style and hours can vary
184

greatly by individual so what seems like a reasonable balance to one person may seem
unsatisfactory to another. Norms research supports this approach as norms are often constructed
based on internally formed perceptions of the environment (Ashforth & Mael, 1989), and it
would be interesting for future research to more closely examine if there is a difference between
perceptions of manager balance and self-reports of that balance.
Related to perceptions, Study 3 looked at individual perceptions of manager support for
work-life balance (Hypothesis 26) and those effects on work-life interference. Although the
proposed moderation hypothesis was not supported, there was a moderate effect for manager
support on work-life interference, such that those who felt more supported by managers, felt less
work-life interference. This finding is consistent with other findings in the work-life balance
area that show that manager support is integral for employees to feel positively about their worklife balance and to feel more able to arrange their work and personal lives schedules to match
their role needs (O’Driscoll, Poelmans, Spector, Kalliath, Allen, Cooper, & Sanchez, 2003;
O’Neill, Harrison, Cleveland, Almeida, Stawski & Crouter, 2009).
Team. Related to manager support for balance is team support for balance. The nature of
work is changing such that it is becoming more team based (Mathieu, Maynard, Rapp & Gilson,
2008), and employees in companies such as Genericorp typically spend a lot of their working
time collaborating with teammates. Hypothesis 28 proposed a moderating effect of team worklife balance culture that was not supported in Study 3. However, the direct effect of team support
and team culture were found.
Aggregated to the team level, team instrumental support (e.g. the degree to which
teammates step up to help others with practical help when they need it to meet personal life
demands) was related to individual work-life interference, such that those who were on teams
185

with higher aggregated team instrumental support for work-life balance experienced lower worklife interference. This effect was small, but still interesting, given the already noted tendency for
individual variation in work-life balance.
Potentially more interesting than the culture of support variable was one that had to do
with perceptions of workplace flexibility. Individual perceptions of work flexibility ability
(previously discussed in the Flexibility section of the discussion) met conditions to be aggregated
to the team level as well, and can be thought of as a team level variable. This aggregated team
level perception of work flexibility ability can be thought of as a normative perception held by
the team at large of how possible it is for work to flex to accommodate personal life demands.
The effect for this team level variable had cross level effects on individual work-life balance,
such that those with teams with norms that showed more ability to flex for personal life demands
showed lower work-life interference.
Although the moderated teams hypotheses were not supported in this paper, the direct
cross level effects represent strong contributions as very few studies have previously focused on
team or local effects on work-life balance (Ford, Heinen, & Langkamer, 2007). As Study 2 made
clear, the effects of local culture for work-life balance are strong, with many extrapolating that
how their team operates with regard to workload and balance norms is the same as how all teams
within the company operate. This area is ripe for future research, with more targeted studies
more fully examining what types of support or norms are supported at the team level.
One of the potential weaknesses of this study, and one that is often felt in teams research,
had to do with how team was defined for analysis (Guzzo & Dickson, 1996). A team was defined
as all employees who report up to a specific manager. However, it is entirely possible the
individuals within the Genericorp population who responded to Study 3 or Study 1 define their
186

teams differently than this or are on more than one well defined work team and more closely
identify with project teams and management-chain teams. This makes it challenging when
attempting to define the group that has the most local influence on an individual in terms or norm
setting and expectations. Future research could potentially improve on this design by having
individuals define their own teams, or use a population that has very well defined teams to
investigate the relationships.
Family. A third type of social support that emerged through the data was that of family.
Although this relationship was not specifically hypothesized, a very interesting relationship was
found in the correlations of Study 3. Typically in work-life balance research, work-life balance
dissatisfaction (in both directions) increases as more role responsibilities are added to a person’s
life. So parents tend to have more unsatisfactory work-life balances than non-parents, and those
who are married tend to have more unsatisfactory balances than single individuals (Banjeree &
Perrucci, 2012). This is hypothesized to be due to extra role load with the same amount of time
resources such that the individual is increasingly spread thin as responsibilities increase
(Kinnunen & Mauno, 2008). In Study 3, this relationship was actually reversed. Married people
and parents actually showed lower scores on work-life interference than their un-partnered and
childless counterparts, as well as a higher ability to detach.
The interviews from Study 2 add some explanation to why this might be. As mentioned
in the results for Study 2, all 10 parents interviewed mentioned having the same nightly schedule
after work, one that included several hours of time that was explicitly devoted to family with no
work interruptions. Although they also mentioned checking in to work again late in the evening,
it may be that this period of detachment that parents appear to be using to be more fully present
for their children could be serving as a protective force for work-life interference. It seems
187

logical to suppose that individuals with spouses may be engaging in similar detaching behaviors
while at home. Several of the interviews mentioned talking with spouses or partners as a routine
part of their evenings. In marked contrast to this were some of the single interviewees who
spoke more of never turning off from the office.
It is possible that the finding that having a family may serve as a protective factor against
work-life balance through more effective detachment may be a phenomenon that is specific to
Genericorp. The workstyle at Genericorp is frequently described as “fast-paced”, with more
work than any one person could possibly finish. Perhaps in environments such as this, it is
difficult for some individuals to impose their own hard lines for when it is appropriate to detach
from work. When external pressures like family come into play, individuals may find it easier to
draw this line between work and non-work time (Winkel & Clayton, 2010). It is also possible
that for some people, work serves as a protective factor for unusually stressful home lives.
Research suggests that this is particularly likely for those who live in lower SES situations or
have stressful home lives (Grzywacz & Fuqua, 2000; Zabkiewicz, 2010). This population is
likely not captured in this sample, so these relationships would have to be observed in other
studies with more diverse samples to be considered generalizable. That said, although
Genericorp is not indicative of all organizations, this mindset of high velocity and innovation has
become increasingly prevalent in professional and competitive for-profit organizations, so it is
likely that these findings may generalize to workers in similar industries or roles.

Culture, Team Norms and Expectations
One of the main goals of this paper was to further understand how environmental context
helps explain the barriers and resources individuals have in seeking a satisfactory work-life
188

balance. Social support from managers, teammates and family appears to be an important aspect
of this, but the hypotheses proposed in the main body of this paper were focused on explicit
norms as well. However, In Study 1 and 3, the moderation hypotheses (Hypothesis 25, 26, and
28) around team norms and workload were unsupported, although there was a small direct effect
of perceived team support on individual work-life interference (as discussed above).
In attempting to interpret these non-results, the interviews from Study 2 provided
invaluable insight and context. Entering into this research project and into the interviews, the
hypothesized assumption was that teams would operate under more explicit espoused norms
regarding work-life balance and workload. These espoused norms may have included things like
explicit competition between teammates, face time norms (e.g. always be at work between the
hours of 8 and 6), explicit overtime assumptions for teammates (e.g. always check email on
weekends), and similar “out in the open” team assumptions that all team members felt compelled
to conform to. The interviews revealed very different team dynamics, at least on the surface.
Interviewees were more likely to say that their teammates and managers explicitly discouraged
all of these behaviors, encouraging them to leave early, rest, think of their balance, take time off,
etc. rather than explicitly encouraging them to work harder. And yet, many still struggled with
achieving a satisfactory balance. So the question remained, what was driving this?
The team effects revealed in Study 2 were much more subtle than anticipated and
occurred within the minds of the individuals being interviewed. The context effect of norms
appeared to be layered at several different levels. First, the individual had specific beliefs about
what sort of performance was expected for themselves. This expectation belief was generally
based on beliefs about what other teammates or peers were doing, and calibrated against their
performance. These beliefs were embedded in larger cultural context where individuals had
189

beliefs about how they would be treated by their leaders or Genericorp in general (e.g. with
career development or role responsibilities) if they did not meet their personal levels of
performance expectations. These beliefs generally seemed to be formed internally to the
individuals, based on watching peers perform, seeing what was being rewarded within the
system, and attempting to calibrate one’s own performance expectations to the model.
An interesting open question highlighted by these three studies is whether or not the
choices some individuals make to attempt to maximize performance at the cost of work-life
balance satisfaction actually pay dividends for them in performance evaluations. In Studies 1 and
3, Hypothesis 2 did find a significant relationship between workload and performance such that
those who worked longer hours had slightly higher performance ratings, but this effect was very
small. Study 1 and Study 3 were split on support for Hypothesis 3 with Study 1 finding some
support for the link between performance and work-life balance (those with higher performance
had lower balance), while Study 3 did not find this relationship. These are intriguing findings
because in the interviews in Study 2, there was a strong belief that putting in extra time and
effort and sacrificing satisfaction with work-life balance is directly linked to performance gains
for individuals.
Future research could more fully explore the relationship between performance, success,
workload and work-life balance to unpack the complicated relationships between the three.
Because Studies 1 and 3 are correlative and not longitudinal, it is impossible to see if there are
causative or reciprocal relationships between high workloads and increasing or diminishing
performance, or an inverted U relationship that Janssen (2001) suggests. The detachment
findings would also suggest that those who are able to psychologically detach from work during
non-work times may also be able to tolerate higher levels or workloads and potentially higher
190

performance levels. A future study in this area could explore experience sampling methodology
to get real-time information about individual detaching behaviors, balance satisfaction and
performance to see if there is an optimal equilibrium between these variables. Some initial work
in this area has been done by Sabine Sonnentag (Binnewies et al, 2009; Sonnentag & Ziljistra,
2006) who has discovered that performance is often the lagged variable, ultimately effected by
psychological stress. She also calls for more research in this area, citing longer time series to
designs to better infer causality and reciprocal relationships.

Emotional Exhaustion and Retention: Why Organizations Should Care
All three studies investigated how work-life balance and boundary management impacted
outcomes that have real importance to organizations, namely emotional exhaustion and retention.
In addition to the individual problems high levels of emotional exhaustion and stress can cause
(including increased mental and physical health complaints, substance abuse and decreased wellbeing; Schaufeli et al, 2009; Shirom & Melamed, 2005; Um & Harrison, 1998), emotional
exhaustion can take a toll on organizations through increased costs associated with highly
stressed employees (in the U.S. the costs are estimated at $300 billion per year; Rupprecht &
Grawitch, 2010). Organizations also have a clear stake in decreasing attrition, particularly in
industrial sectors where hiring is difficult and competition between companies is strong (Trevor,
Gerhart & Boudreau, 1997). This paper adds value to organizations by more clearly defining the
relationships between work-life balance and boundary management techniques to these
organizationally relevant outcomes and helps lead to the development of resources and
interventions that could decrease emotional exhaustion and attrition.

191

Emotional Exhaustion. The strongest observed relationship in Study 3 was between
emotional exhaustion and work-life interference, with effect sizes so large that they deserve a
closer examination. Although large, this connection is in line with theory in the field that
suggests that emotional exhaustion will more prevalent in occupations where there is a high
amount of personal job involvement as these individuals are investing more in their work role
than those who are less involved in their jobs (e.g. Bakker, Demerouti, & Schaufeli, 2002,
Maslach, Schaufeli & Leiter, 2001). The interviews from Study 2 would confirm that job
involvement for this population appears to remarkably high.
Additionally, burnout theory indicates that there can be many causes of job burnout (of
which emotional exhaustion is an important component), and these causes are likely additive
(Maslach & Leiter, 2008). These causes include lack of control, unclear job expectations, reward
imbalance, values misfit between individual and company, poor fit with job, lack of social
support, large job demands or work-life imbalance (Vallerand, 2010). Across all three studies,
many of these causes appear to be relative non-issues, with employees reporting generally high
levels of control, intrinsic interest and engagement in their work, and a general sense of fit with
and commitment to their job and company. Therefore, for the majority of Genericorp employees
who might be feeling emotionally exhausted, the two main components of the burnout equation
that they are likely to experience are work demands (in the form of workload) and work-life
interference.
Future research in this area should more closely examine how the relationship between
work-life interference, work demands, job involvement and emotional exhaustion plays out in
other populations. It would be intriguing to investigate the differential effects of the burnout
“causes” on both work-life interference and emotional exhaustion. In the Genericorp population,
192

work-life interference and workload appear to be the main drivers of emotional exhaustion given
the relative mitigation of the other causes of burnout in the general population. Other work
situations may show different results.
One potential weakness of the findings linking emotional exhaustion and work-life
interference is in the way the items were worded. Work-life interference was anchored as only
occurring in the work-to life direction and emotional exhaustion was worded to be specifically
grounded in the work sphere as well, rather than a general assessment of stress. While
situationally grounding variables is recommended for studies that are focusing on one domain in
particular, as this paper did (Shaffer & Postlewaite, 2012), the fact that both variables are
specifically focused on the work environment may be showing stronger relationships between
the two than if more global measures of work-life balance or stress were utilized.
Retention. Retention of qualified employees is a strong priority for organizations as
losing employees represents a loss not just of their current performance but also of the costs of
searching for, interviewing, hiring and training new employees (Trevor, Gerhart, & Boudreau,
1997). Because the costs to the organization are so high for losing employees, even reducing
voluntary turnover by a small amount is likely worth a great deal to an organization.
Across all three studies, the findings related to work-life balance and boundary variables
and retention were smaller than those found for emotional exhaustion but were still significant.
In Studies 1 and 3, those who reported lower satisfaction with work-life balance or greater worklife interference reported decreased intentions to stay with the organization. This sentiment was
echoed in Study 2 as well. During the interviews, a substantial proportion felt that they would
need to leave their current role in order to achieve the work-life balance they desired.

193

One key weakness of this study is that as it was a series of cross sectional studies, it was
impossible to observe actual attrition behavior, e.g. people actually leaving the organization.
Future studies should examine the effect of work-life balance on attrition longitudinally so as to
be able more concretely link attitudes toward work-life balance to actual leaving behaviors rather
than just retention intentions (as this study measured).

Weaknesses
Within each section in the General Discussion, specific strengths, weaknesses and
research directions were discussed. However, there are a few weaknesses that span across all
three studies that should be called out specifically. These have to do with team definition,
organizational setting and the measurement of performance.
In Study 1 and Study 3 teams were used as a contextual unit of analysis for several of the
hypotheses. These teams were defined as all individuals who report up to the same manager.
This definitional decision was made for two reasons. One, this is consistent with how teams are
defined in field studies elsewhere in teams research as teams under managers (Stewart, 2006)
unless specific teams such as flight crews or emergency room teams are being investigated
(Guzzo & Dickson, 1996). Two, there were no better ways to designate team membership
available for categorizing individuals. However, it is possible and likely that many individuals
who were part of the studies defined their teams differently or were part of different teams. It is
likely that teams within Genericorp more closely resemble project teams than management
teams, but classifying people into these groups was impossible with data available.
This may have been particularly problematic for the research conducted in this paper,
since many of the effects investigated in Studies 1 and 3 have to do with the normative and social
194

effects of team. If an individual was assigned to a team for analysis that they personally did not
feel a social connection to, then the social effects of the normative behavior the people in that
group would not have an effect on them (Ashforth & Mael, 1989). The inability for individual
participants to self-identify team membership and be analyzed within those groups may have
hindered analytical attempts to find social or normative effects of teams.
A second weakness is that all three studies were conducted within the same organization.
The benefits of this approach included a high degree of access that allowed a very in depth and
multi-part study to be conducted in a field setting with full cooperation of Genericorp. The
drawback, however, is in generalizability. The Genericorp population likely represents a specific
type of professional, highly skilled employee with high job importance and these findings likely
do not apply to dissimilar industries or employee types. That said, it is likely that these finding
have some generalizability to individuals who are similar to Genericorp employees or work in
similar settings, so the findings have some utility beyond the localized context.
The final weakness that spans Study 1 and Study 3 is that the measure of performance is
not optimal. As the employees at Genericorp generally do not produce something that can be
measured in objective output, the rating consists of various subjective inputs that are calibrated
for fairness. This results in a score that is very range restricted with the majority of individuals
falling around the midpoint with very little variation. This makes it difficult to find any
relationships between performance and anything else.

Future Research Directions
In addition to the research directions outlined in each section of the General Discussion
above, there are several research directions that emerged from this paper that go beyond the
195

initial scope of the project and thus need their own section. These research directions have to do
with team task interdependence, protective factors in an individual’s environment, and
developmental aspects of the work-life interface.
Team task interdependence was a variable that was collected as a control variable for the
purposes of this paper. How teams dynamically interact with one another is outside the scope of
this paper which focuses more on team culture and its effects on individual satisfaction and
boundary management so further investigation into how task interdependence affects individual
and group outcomes was not explored. However, team task interdependence shows some
intriguing relationships to other variables including those that are of very high interest to
organizations, such as performance. Although this correlation is small, any relationship to the
range restricted performance variable is notable as most variables are uncorrelated to
performance in Study 3. Future research should further explore team task interdependence and
the potential relationship to both individual performance and team level performance and to
investigate whether these relationships hold in a study focused on these variables.
Perceptions of team task interdependence also had an effect on individual levels of
psychological detachment, with those with more interdependent teams showing lower levels of
psychological detachment. This finding opens up a different lens through which to consider how
individuals interact with their teams in constructing their own satisfaction with work-life balance
and boundary management techniques. This paper mainly dealt with perceptions of ability and
willingness to exercise detachment and flexibility behaviors, but the task interdependency
measure highlights how individual’s roles within the organization may function differently such
that those in roles that require less interdependency may be better able to detach. Future research

196

should more closely investigate the role and workload dynamics and sharing among teams, and
how this can affect detaching and flexible arrangements.
Another area that is open for further research has to do with protective factors at the team
and individual levels. Study 3 found no effect for gender on work-life interference and also
found that having a child or a spouse served as a protective factor against work-life interference.
These findings are in contrast to what is usually found in the literature. It would be important to
replicate these findings again to confirm as they were unhypothesized and to ensure that they
were not sample specific. Future research could more clearly illuminate in what sectors or job
types does having a family help individuals define clearer boundaries such that the increased
demands of the role are mitigated by the benefits achieved from having a reason and an impetus
to detach.
The final research direction that arose from this paper has to do with considering worklife balance from a developmental lens. In Study 2, it was clear that for some of the interviewees,
they viewed poor work-life balance as a necessary and (to some degree) acceptable sacrifice for
career advancement. Some of the interviewees were clearly sacrificing their work-life balances
in exchange for quick promotion and advancement, while others were content to do solid work
and maintain a satisfactory balance without as rapid career growth. It would be interesting for
future researchers to more closely examine work-life balance and career aspiration decisions in
tandem from a developmental perspective such that are there phases in an individual’s life where
they are more willing (and able) to double down on work while focusing less energy on their
personal life and other phases where they choose to be more focused on their personal life. This
area could be particularly illuminating in looking at the switch points, the time periods where
individuals choose to shift focus from work to personal life or back again. Some researchers have
197

begun to investigate this area, looking to see whether loyalty patterns in employees shift over
different life stages. Roehlings, Rhoelings and Moen (2001) found that employee loyalty and
preferences for organizational work-life balance supports shifted over personal life phase
changes, including singlehood, parenthood, and older adulthood. Roehlings et al (2001) do call
for further research into this area to explain patterns of employee loyalty and whether
organizations should strive to meet different needs of different employees as the progress
through different life stages.

Practical Implications
Several practical implications are discussed above in the specific sections of the general
discussion, but there are some overarching ways the findings of this paper can help organizations
help employees better balance work and personal lives. The importance of sense of control and
autonomy cannot be understated, and organizations should design work and roles with an eye
toward maximizing employee decision latitude wherever possible. Within the interviews and
then confirmed in Study 3, individuals who felt they were responding to more reactive tasks than
proactive ones felt more dissatisfied with their balance and more out of control of their schedules
in general. This control appears to be important for psychologically detaching from work, which
can allow individuals to adequately rest and recover.
Organizations should also be aware that satisfaction with flexibility seems to be the most
important aspect of flexibility, rather than actual flexible behaviors (e.g. working from home or
flexing hours) or having manager make explicit agreements with their subordinates about
flexible arrangements. Monitoring employee flexibility satisfaction over time could help
organizations keep a pulse of how employees are feeling with regards to flexibility and intervene
198

if satisfaction declines. It would appear that options for flexibility may not even have to be used
very often for employees to feel satisfied with them; they just have to know that they option to
use them is there.
These are important findings because they help organizations help employees improve
work-life balance while maintaining high workloads. It is often impractical to suggest that
organizations decrease workloads as a way to deal with work-life balance as organizations exist
in a competitive world where a certain amount of work must get done, so solutions that can
capitalize on behavioral or cultural change are more likely to be adopted.
In addition to things organizations can do, this paper also suggests practical interventions
for managers and individuals. Manager support had a direct effect on work-life interference such
that those who had managers who they felt were both emotionally supportive and instrumentally
supportive had lower work-life interference. Managers should be mindful of this and take some
targeted time to ask their subordinates about their workloads and balance and take specific steps
to help them reprioritize or redistribute work if balance is a continued problem. This kind of
support can go a long way in increasing work-life balance satisfaction for an individual
employee.
Even absence organizational and manager support, this paper suggests that there are
things individuals can do to help with their own work-life balances. The findings suggest that
actively detaching from work, including psychologically as well as physically detaching (e.g. not
checking email or working on the weekends), can help those with even high workloads have
sustainable work-life balances. Another way individuals can help themselves is through gaining
as much control over their work as possible. One way to do this is to be as mindful about their

199

career and life choices as possible, being honest about benefits and tradeoffs so that decisions
about time spent in work vs. non work roles are more aligned with personal beliefs and values.

200

APPENDICES

201

Appendix A: Selected Scales Used in Study 1

Workload. Which of the following are the biggest barriers to your productivity? Please select up
to 3 from the list below.
1. None: I don't face any major barriers to productivity right now.
2. Workload: I have too much work to get things done.
3. Skills: I don't have the necessary skills to do my job.
4. Autonomy, not enough: I don't have enough control over my work.
5. Manager involvement: I don't get enough guidance/direction from managers/leaders.
6. Low Performers: I'm dependent on people who don't get the job done.
7. Tools / Systems: Software I need to do my job is lacking or inefficient.
8. Equipment: Physical resources I need to do my job are lacking or inefficient.
9. Meetings: I spend too much time meeting rather than getting work done.
10. Interviewing: I spend too much time interviewing/evaluating job candidates.
11. Politics: I'm dependent on people who don't collaborate effectively.
12. Risk aversion: Culture/management doesn't approve or support riskier ideas/projects.
13. Decision-making takes too long.
14. Decision-making involves too many people.
15. Other (please explain below)
Retention Items assessed on 1-5 agreement scale
1. I plan to be working at Genericorp one year from now.
2. I plan to be working at Genericorp five years from now.
Work-Life Balance. Items assessed on a 1-5 agreement scale
1. I am able to detach from work during non-work time (i.e., when I choose not to be
working).
2. I am satisfied with the balance between my work-life and my personal life.

202

Appendix B: Structured Interview Script

Hi! My name is Megan Huth. I’ve worked with the People Analytics since last summer/fall as an
intern, and I’ll be joining the group full-time in January. In the meantime, I’m working on my
PhD at Michigan State University and this project is a component of my dissertation research.
Thank you for agreeing to participate!
I’m meeting one-on-one with Genericorp employees as part of a broader data collection going on
this fall. This session will last an hour, and I’m excited to hear what you have to say. Through
these interviews, I’m trying to get a better understanding of how Genericorp employees are
experiencing work-life balance and in what ways they feel supported (or not) by their managers
and teams. When I talk about work-life balance, I’m referring to how satisfied you are at
balance of time and energy you get to put toward work and other roles you care about, like
family, friendships, volunteerism, sports, or really anything else you care about.
This will be a guided interview with specific questions I’ll be asking, but if you have anything to
add, or a different perspective you think is important, please speak up! The purpose of these
interviews is to really capture what Genericorp employees are experiencing and the best way get
at that is through your contribution. All interviews will remain confidential—the transcripts will
be provided to People Analytics to study, but identifying information will be removed.
Do you have any questions before we begin? To jump start the interview I want you to tell me
about your day yesterday. Can you walk me through what it was like? What your morning was
like at home, how you got to work, your work day, things like that?
…..

Workload
1. Tell me a little about your workload. How many hours do you estimate you work per
week? Do you feel like your workload is too much, just right or not enough? How would
you describe your ideal workload?
2. Tell me a little about how you structure your work. How would you describe the
flexibility of your work? Do you spend the same hours in the office each day or is your
schedule more flexible?
3. Tell me a little about your team’s workload. How is work distributed on your team?
4. Can you give me specific examples of times you felt you were able to successfully
structure your workload to be able to achieve satisfactory balance? How about a time you
were less successful?
Non work responsibilities and conflict
1. Would you say that overall you are satisfied with your balance between work and
nonwork?
203

2. How frequently do you have demands on your personal time from family, friends,
hobbies or interests? Can you tell me a little bit about how much time you prefer to spend
on non-work per week? What kinds of activities do you like to do during non-work times
3. In what ways (if at all) does work conflict with the things you’d like to do in the rest of
your life? If so, can you describe an incident and tell me how you dealt with it?
4. Can you give me specific examples of times you felt you were able to successfully
demands from both work and non-work to satisfactory balance? How about a time you
were less successful?

Manager support
1. How does your manager support your efforts to balance your work and non-work-life?
2. In what ways does your manager model ways to successfully balance work and nonwork?
3. Can you give me specific examples of times you felt more supported by your manager in
balance work and life? How about a time you felt less supported?
4. What kind of support from your manager do you think would be ideal to help you
balance? What kind of support would you like but currently don’t have?
5. Have you made any arrangements with your manager that have helped you balance your
work and family such as increased flexibility in hours or a work-from home option? If so,
can you describe how you made these arrangements? Did you approach your manager,
did she/he approach you or did it just arise organically?
6. How easy do you find it easy to talk to you manager about work-life balance issues?
Why is this easy/hard?

Team support
1. What are the norms on your team for work-life balance? Describe the balance of your
team members as you see it.
2. Do you feel that other members of your team are able to successfully balance their work
and life? Why or why not?
3. In what ways does your team support you? In what ways could they be better supports?
4. In what ways do you feel supported by your team members to take steps to balance your
work and non-work-life?
5. What team norms are for hours worked? For face time?
6. Do other members on your team have more or less flexible work schedules than you? If
yes, why do you think there are these differences? What types of flexible arrangements
do you team members have?
Barriers and Strategies
1. Do you use flextime? If yes, then how do you use it? If no, then why not?
2. Do you ever work remotely? If so, where do you work and how often?
3. What do you think are good strategies for balancing work and non-work-life?
4. What sorts of barriers do you think stand in the way of better balance? Are there barriers
from your manager? Team? Yourself?
204

Additional Information
1. In an ideal world, how would you structure your work and non-work time to achieve
optimal balance?
2. Is there anything else you want to tell me about how work-life balance is treated here at
Genericorp and within your team?

205

Appendix C: Survey Items or Study 3

Unless otherwise noted, all scales are on a 1 to 5, Strongly disagree to Strongly agree scale.

Involvement
1. I'm excited about the work I'm doing.
2. I make good use of my skills and abilities in my current role.
3. My work gives me a sense of personal accomplishment

Autonomy
1. I have significant autonomy in determining how I do my job.
2. I decide on my own how to go about doing my work.
3. I have considerable opportunity for independence and freedom in how I do my job.

Control of work
1. I feel in control of my time.
2. I find it difficult to keep to my schedule because others take me away from my work.
3. I feel that I have my work under control.
4. I feel confident in that I am able to complete my work on time.
5. I often have little control of what is happening at work.

Retention (selected items only- 4 items total)
1. I plan to be working at [Genericorp] one year from now.
2. I plan to be working at [Genericorp] five years from now.

Emotional Exhaustion
1. I feel fatigued when I get up in the morning and have to face another day on the job
2. I feel burned out from my work
3. I feel frustrated by my job
4. This job puts too much stress on me
5. When I get up in the morning, I feel like going to work.
6. At work, I feel bursting with energy.
7. I am enthusiastic about my job.

Job interference with life
1. Overall, I am satisfied with my work-life balance
2. I am worried that my work interferes with my non-work activities and interests.
206

3. Things I want to do outside of work can’t get done because of the demands my job puts
upon my time.
4. Due to emergencies at work, I have to make last minute changes to my plans for activities
off the job.
5. I have to put off non-work things I would like to do because of my work requirements.
6. I can’t sleep because of thinking about things at work that I have to get done.
7. Being in a positive mood at work helps me to be in a positive mood at home.
8. Being happy at work improves my spirits at home.
9. Values developed at work make me a better person in my nonwork life.
10. When things are going well at work, my outlook regarding my nonwork life is improved.

Family Flexibility-Ability
1. Because of my personal life responsibilities, I cannot change my work schedule (for
example going in early or staying longer to finish work related responsibilities) .
2. If the need arose, I could work late without affecting my personal life responsibilities
3. My personal life responsibilities would not prevent me from going into work early if the
need arose.
4. My personal life responsibilities would not prevent me from going into work an extra day
in order to meet work responsibilities.
5. From a personal life standpoint, there is no reason why I cannot rearrange my schedule to
meet the demands of my work.

Family Flexibility-Willingness
1. I am willing to change plans with my friends and family so that I can finish a job
assignment.
2. I am willing to change vacation plans that I have made with friends and family to meet
work related responsibilities.
3. While at home, I do not mind stopping what I am doing to complete a work related
responsibility.
4. I am not willing to cancel plans with my friends and family to deal with work related
responsibilities.

Work Flexibility-Ability
1. I am able to arrive and depart from work when I want in order to meet my personal life
responsibilities.
2. If the need arose, I could leave work early to attend to personal life issues.
3. If something came up in my personal life, it would be all right if I arrived to work late.
4. While at work, I can stop what I am doing to meet responsibilities related to my personal
life.

207

Work Flexibility-Willingness
1. I am willing to take an extended lunch break so that I can deal with responsibilities
relating to my personal life.
2. Assuming it was alright with my manager, I would not mind arriving to work late so that
I could meet my personal life responsibilities
3. If it became necessary in order to meet my personal life responsibilities I would be
willing to change the shift, or start stop times, that I normally work.
4. I am not willing to take time off from work to deal with my personal life responsibilities.
5. I would be willing to compress my normal work week into 4 days rather than 5 if it meant
I could better deal with my personal life responsibilities.

Manager Support
1. My manager is willing to listen to my problems in juggling work and non-work life.
2. My manager takes the time to learn about my personal needs.
3. My manager makes me comfortable talking to him or her about my conflicts between
work and non-work.
4. My manager and I can talk effectively to solve conflicts between work and non-work
issues.
5. My manager expects me to respond to email during nonstandard work hours (e.g.,
evenings and/or weekends).
6. I can depend on my manager to help me with conflicts in my schedule between my
personal and my work if I need it.
7. I can rely on my manager to make sure my work responsibilities are handled when I
have unanticipated non-work demands.
8. My manager works effectively with me to creatively solve conflicts between work
and non-work.

Team Emotional Support
1. My teammates are willing to listen to my problems in juggling work and non-work life.
2. My teammates take the time to learn about my personal needs.
3. My teammates make me comfortable talking to them about my conflicts between work
and non-work.
4. My teammates and I can talk effectively to solve conflicts between work and non-work
issues.
5. My teammates expect me to respond to emails or other work requests during nonstandard
work hours (e.g., evenings and/or weekends).

Team Instrumental Support
1. I can depend on my teammates to help me with conflicts in my schedule between my
personal and my work if I need it.
208

2. I can rely on my teammates to help me make sure my work responsibilities are
handled when I have unanticipated non-work demands.
3. My teammates work effectively to creatively help me solve conflicts between work and
non-work.

Team Task Interdependence
1. Most of my work activities are affected by the activities of other people on the team
2. My work cannot be done unless other people on my team do their work.
3. Team members have to work together to get group tasks done.
4. I depend on other people for information I need to do my work.
5. Unsatisfactory performance of my job would delay the work performance of other
people.

Segmentation/Integration
1. It is often difficult to tell where my work life ends and my non-work life begins
2. I tend to integrate my work and non-work duties when I work at home.
3. In my life, there is a clear boundary between my career and my non-work role.
4. I don’t like to have to think about work while I am at home.
5. I prefer to keep work life at work.
6. I don’t like work issues creeping into my home life.
7. I like to be able to leave work behind when I go home.

Psychological detachment
1. During after-work hours, I forget about work.
2. During after-work hours, I don’t think about work at all.
3. During after-work hours, I distance myself from my work.
4. During after-work hours, I get a break from the demands of work.

Evening Work Style
1. Once I leave the office, I almost never log back on to do more work in the evenings.
2. I check work email in the evenings but do not usually do work beyond emailing.
3. In the evenings, I typically take a break from work for a while, but then log back on for a
an hour or two before I go to bed.
4. I do not get much of a break in the evenings-- I am pretty much logged on to work from
the time I get home to the time I go to bed.

Weekend Work Style
1. I almost never do work on the weekends.
2. I usually check work email on weekends but do not usually do work beyond emailing.
209

3. I generally try to keep one day (e.g., Saturday) work free, but tend to do some work on
the other weekend day.
4. I work both days of the weekends.

Vacation Work Style
1. While on vacation, I am able to completely detach from the office.
2. While on vacation, I check work email.
3. While on vacation, I do some work most days.
4. I wish I was able to detach from the office more effectively while on vacation.
5. While on vacation I worry about the amount of work waiting for me when I get back to
the office.

Subjective Workload
1. I have to work fast.
2. I have too much to do.
3. I can do my work in comfort (reverse coded).
4. My workload is high.

Objective Workload
1. How many hours do you work per week on average (including evenings and weekends)
2. less than 35
3. 35-40
4. 41-45
5. 46-50
6. 51-55
7. 56-60
8. 61-65
9. 66-70
10. 70-75
11. more than 75

Perceived flexibility
1. I have the flexibility I need at work.
2. Overall, I am satisfied with the flexibility I have at work.
3. I would like to have more options to work from home or work remotely.
4. I would like greater flexibility in the hours I work.

210

Used flexibility
1. How often do you arrange to work from home in your current position? (5= weekly, 4= A
couple of times a month, 3= once a month, 2= a few time a year, 1= never or almost
never)
2. How often do you arrange non standard start and end times (e.g. coming in late/early
staying late early) in your current position? (5= weekly, 4= A couple of times a month,
3= once a month, 2= a few time a year, 1= never or almost never)

Idiosyncratic Deals
I have asked for and successfully negotiated with my manager:
1. A work schedule suited to me personally
2. Personal choices regarding where I do my work
3. Extra flexibility in starting and ending my work day
4. Possibilities to work from home or remotely
5. Personal discretion in scheduling and timing my work
6. A work schedule customized to my personal needs.

Demographics
[Note: These data will never be shared in any identifiable form with Genericorp]
1. Marital Status
2. # Children at home (if any)
3. Age of children at home (if any)
4. How many hours do you spend per week on housework, including cooking, cleaning,
laundry, picking up dry cleaning, shopping, etc, or arranging for any of these types of
tasks to be done by others: __________
5. How many hours do you spend per week on child and/or parent care, including
transportation of children and/or parents or arranging for child and/or parent care:
________I have to work fast.

211

Role Theory and Conservation of Resources Theory

Appendix D: Tables
Table 1. Summary of paper hypotheses, including whether they were supported
quantitatively by study 1 or 3.
H
Hypothesis
Study 1 Study 3
#
Results
results
1
Workload will be negatively related to WLB and positively
related to work life interference such that with higher
Yes
Yes
workloads will report lower satisfaction with balance and
higher interference.
2
Workload and performance will be negatively related such
No
No
that those with higher workloads will be rated lower on
performance.
3
Performance will be negatively related to WLB and positively
related to work life interference such that those who are rated
Yes
No
as higher performers will indicate less satisfaction with
balance and higher interference
4
Workload and retention intentions will be negatively related
such that those with higher workloads will be less likely to
Yes
Yes
indicate an intention to stay with the company.
5
Performance and retention intentions will be positively related
such that those with higher performance ratings will be more
No
No
likely to indicate an intention to stay with the company than
those with lower performance.
6
Work life balance will be positively related to retention
intentions and interference will be negatively related to
retention intentions such that those who report higher balance
Yes
Yes
will also report higher intentions to stay with the company
and those who report higher interference will report lower
intentions.
7
WLB and work life interference will moderate the
relationship of performance with retention intentions such that
those with higher WLB or lower work life interference and
No
No
higher performance will be more likely to indicate intentions
to stay with the company than those with lower performance.
8
Workload will be positively related to emotional exhaustion
such that those who report higher workloads will also report
Yes
more emotional exhaustion.
9
WLB will be negatively and work life interference will be
positively related to emotional exhaustion such that those who
Yes
report higher balance will report less emotional exhaustion
and those with more interference will report more exhaustion.
Note: For interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation.

212

Table 1 (cont’d)
Hypothesis

10

WLB and work life interference will partially mediate the
relationship of workload and emotional exhaustion such
that when included in the relationship the effect of
workload on emotional exhaustion will be diminished.

11

Manager support for balance will moderate the relationship
between WLB or work life interference and retention such
that those who report more supportive managers will report
higher intention to remain with the company despite levels
of WLB or work life interference
Team support for balance will moderate the relationship
between WLB and work life interference and retention
such that those who report more supportive teams will
report greater intention to remain with the company despite
levels of WLB or work life interference.
Manager support for WLB will moderate the relationship
between work life interference and emotional exhaustion
such that those who receive more support will be buffered
from the negative effects of work life interference on
emotional exhaustion.
Team support for WLB will moderate the relationship
between work life interference and emotional exhaustion
such that those who receive more support will be buffered
from the negative effects of interference on emotional
exhaustion.
Frequency of use of flexible arrangements will be
negatively related to work life interference such that those
who report greater use of flexible arrangements will report
lower work life interference.

Role Theory

H#

12

Border Theory and I-deal
Coping/Support Theories
Theory

13

14

15

16

Study 1
Results

Satisfaction with flexible arrangements will be negatively
related to work life interference such that those who report
greater satisfaction with flexible arrangements will report
lower work life interference.

213

Study 3
results
Yes
(for
subjective
work load
but not
work
hours)

No

No

No

No

Yes
(for
negotiated
flexibility
only)
Yes

Table 1 (cont’d)
H # Hypothesis
17

18

Border Theory and I-deal Theory

19

20

21

22

23

24

Study 1
Results

Frequency of use of flexible arrangements will moderate
the relationship between workload and work life
interference such that those who report more use of
flexible arrangements will also report lower work life
interference despite high workloads.
Satisfaction with flexible arrangements will moderate the
relationship between workload and work life interference
such that those who report more use of flexible
arrangements will also report lower work life interference
despite high workloads.
Frequency of use of flexible arrangements will be
positively related to retention intentions such that those
who report more use of flexible arrangements will also
report greater intentions to stay with the company.
Satisfaction with flexible arrangements will be positively
related to retention intentions such that those who report
more satisfaction with flexibility will also report greater
intentions to stay with the company.
Frequency of use of flexible arrangements will be
negatively related to emotional exhaustion such that those
who report more use of flexible arrangements will also
report lower levels of emotional exhaustion.
Satisfaction with flexible arrangements will be negatively
related to emotional exhaustion such that those who report
more satisfaction with flexibility will also report lower
levels of emotional exhaustion.
A preference for segmentation will moderate the
relationship between flexibility and work life interference
such that those who prefer segmentation will not report
lower work life interference with increases in flexibility.
WLB management techniques will moderate the
relationship between workload and work life interference
such that those who use more techniques will be buffered
from the negative effects of high workload on work life
interference.

214

Study 3
results

No

No

Yes
(for
negotiated
flex. only.)
Yes

No

Yes

No

See below

Table 1 (cont’d)
H # Hypothesis
24.1

24.2

Supplemental Hypotheses: Border Theory

24.3

24.4

24.5

24.6

24.7

24.8

Study 1
Results

Actual Integration will moderate the relationship with
workload and work life interference such that those who
integrate work and life more will experience more
interference regardless of workload.
Preferred segmentation will moderate the relationship of
actual integration and work-life interference such that
when there is a match between segmentation preference
and segmentation behavior, the relationship between
actual segmentation and work-life interference will be
reduced or eliminated.
Life flexibility ability will be positively related to work
life interference such that those who see their personal
lives as being less able to be permeable to work will
experience less work life interference.
Life flexibility ability will moderate the relationship
between work life interference and workload such that
those who are less willing for their personal lives to be
permeable to work will experience less work life
interference regardless of workload.
Life flexibility willingness will be positively related to
work life interference such that those who see their
personal lives as being less able to be permeable to work
will experience less work life interference.
Life flexibility willingness will moderate the relationship
between work life interference and workload such that
who are less willing for their personal lives to be
permeable to work will experience less work life
interference regardless of workload.
Work flexibility ability will be negatively related to work
life interference such that those who see work as being
more permeable to their non work lives will experience
less work interference with life.
Work flexibility ability will moderate the relationship of
work life interference and workload such that those who
believe work is flexible to accommodate to persona life
will experience lower work life interference regardless of
workload.

215

Study 3
results
Yes
(subj.
workload
only)

No

No
(significant
in the
opposite
direction)

No

Yes

No

Yes

No

Table 1 (cont’d)
H#

Hypothesis

24.1

Work flexibility willingness will be negatively related to
work life interference such that those who are more
willing to allow their non work lives to permeate their
work lives will experience less work interference with
life.
Work flexibility ability will moderate the relationship of
work life interference and workload such that those who
are more willing to allow their non work lives to
permeate their work lives will experience lower work
life interference regardless of workload.
Control of work will be negatively related to work
interference with life such that those with greater control
of work will report less work interference with life.
Control of work will moderate workload and work
interference with life such who have greater control of
work will report less work interference with life
regardless of workload.

Supplemental Hypotheses: Border Theory

24.10

24.11

24.12

24.13

24.14

24.15

24.16

Study
1
Results

Autonomy will be negatively related to work
interference with life such that those who report more
autonomy will report less work interference with life.
Autonomy will moderate workload and work
interference with life such that who report more
autonomy will report less work interference with life
regardless of workload.
Psychological detachment will be negatively related to
work life interference such that those who are more
psychologically detached will report lower work life
interference
Psychological detachment will moderate the relationship
between workload and work life interference such that
those who are more psychologically detached will
experience less work interference with life regardless of
workload.

216

Study 3
results

Yes

No

Yes
Yes
(for
subjective
workload
only)
Yes
Yes
(for
subjective
workload
only)
Yes
Yes
(for
subjective
workload
only)

Table 1 (cont’d)
H#

Hypothesis

24.17

Evening work style will be positively related to
work interference with life such that those who
work more in the evenings will report more work
interference with life.
Evening work style will moderate workload and
work interference with life such that who work in
the evenings will report more work interference
with life regardless of workload.

24.18

Supplemental Hypotheses: Border Theory

24.19

24.20

24.21

24.22

Study 1
Results

Weekend work style will be positively related to
work interference with life such that those who
work more on the weekends will report more work
interference with life.
Weekend work style will moderate workload and
work interference with life such that who work on
the weekends will report more work interference
with life regardless of workload.
Vacation work style will be positively related to
work interference with life such that those who
work more on vacation will report more work
interference with life.
Evening work style will moderate workload and
work interference with life such that who work on
vacation will report more work interference with
life regardless of workload.

217

Study 3
results
Yes
Yes
(for
subjective
workload
only)
Yes
Yes
(for
subjective
workload
only)
Yes
Yes
(for
subjective
workload
only)

Table 1 (cont’d)
H#

Hypothesis

25

Manager work life interference will moderate the
relationship of subordinate workload and work life
interference such that those with managers with more
work life interference will also have high work life
interference.
Perceptions of manager expectations about work-life
balance will moderate the relationship of subordinate
workload and work life interference such that those
with mangers what expect balance will also have lower
work life interference, regardless of level of workload.
Team WLB will moderate the relationship between
individual workload and individual work life
interference, such that those in teams with higher
balance will be buffered from the negative effects of
workload on work life interference.
Perceptions of team WLB norms will moderate the
relationship between individual workload and
individual work life interference such that those in
teams with norms for higher balance will be buffered
from the negative effects of workload on work life
interference.

Group Normative Theory

26

27

28

Study
1
Results

218

Study 3
results

No

No

No

Could not
be tested

No

Table 2: Team Summary for Study 1
Number of Team
Members
4
5
6
7
8
9
10
Average

Number of
Teams in Data
Set of That Size
32
28
15
21
20
10
8
6.78

Average
Response Rate
Per Team
0.95
0.95
0.91
0.90
0.85
0.91
0.95
.91

Table 3: Summary of Sample Characteristics

Job level
(1 low to 9 high)

Function

Engineering

Individuals
298

Teams
41

Sales

292

44

General and
Admin
Tenure in Years
(Avg)
Level 1-3

220

37

Level 4-5

350

Level 6-7

202

Level 8-9

22

Average
Performance
Score

3.44

3.14
227

219

Table 4: Bonferroni Comparison for Work Life Balance Scores (n=810)

Comparisons

Mean Function
Std Error
Differences
Engineering : Sales
.04
.07
Engineering : G &A
.21*
.08
Sales: G&A
.17
.07
* The mean difference is significant at the .05 level

95% CI
Lower Bound Upper
Bound
-.13
.22
.02
.40
-.02
.36

Table 5: Study 1 Correlation Matrix.
Job
level

Team Size

Team
size
1

Tenure Perf
rating

Funct

Wrkld

Ret.

WLB

Job level

.12**

1

Tenure (in
days)
Performance
rating
Function
(Sales/Not)

.05

.37**

1

-.07

.20**

.22**

1

- .22**

.21**

-.11**

.07

1

Workload

-.08*

.01

.12**

.11**

.08*

1

Retention

.04

-.08*

-.19**

-.06

.03

-.03

1

Work Life
Balance
Manager WLB

.11**

.01

.08*

.00

-.12**

-.35**

-.10**

1

.13**

-.04

-.03

-.07

-.08*

-.03

.24**

.12**

Mgr
WLB

1

* p< .05, **p< .01
Notes: Function was coded 1 = General & Admin, 0 = Sales and Engineering functions; Manager
WLB scores represent the actual scores of the individual’s manager as mapped to them through
the organizational hierarchy.

220

Table 6: Means and Standard Deviations of Study 1 Variables
N
Team Size
Job level
Tenure (in days)
Performance rating
Function
Workload
Retention
Work Life Balance
Manager WLB

Mean

810
777
805
740
810
810
809
810
688

6.75
4.57
1145.37
3.44
.27
.40
3.73
3.51
3.74

221

Standard
deviation
1.90
1.55
683.92
.25
.45
.49
.80
.94
.89

Table 7: Study 1 Unconditional Means Models
for Work Life Balance and Retention (n=809)
Work Life
Balance

Retention

Intercept

3.76**

3.73**

Standard error

(.05)

(.04)

Intercept

.63**

.64**

Standard error

(.04)

(.04)

Residual

.12**

.04

Standard error

(.03)

(.02)

Model Fit
Statistics
Deviance

1724.55

1679.69

AIC

1728.55

1683.69

BIC

1737.61

1692.75

ICC(1)

.16

.06

Fixed Effects

Random effects

* p< .05, **p< .01

222

Table 8: Hypothesis 1: HLM models for workload predicting WLB. (n=809)
Fixed Effects

Model 1
Unconditional
Means (empty
model to test for
“groupiness”

Model 2
Fixed effect:
Unconditional
Means Model
with Controls

Model 3
Random Effect:
Unconditional
Means Model
with Controls

Model 4
Fixed Effect
Hypothesis
Model Workload
Predictor

Intercept
Standard error
Workload
Standard error

3.72**
.04

3.42**
.16

3.42**
.16

3.49**
.15
-.61**
.06

Control: Team size

.05*

.05*

.04

Standard error
Control: Function
Standard error
Random effects
Intercept
Standard error

.02
-.11
.09

.02
-.11
.09

.02
-.06
.09

.12**
.03

.11**
.03

.11**
.04

.08**
.02

Residual
Standard error
Model Fit Statistics

.67**
.04

.67**
.04

.67**
.03

.60**
.03

Deviance

2070.53

2071.29

2071.29

1974.71

AIC

2074.53

2075.29

2075.29

1978.71

BIC

2083.92

2084.67

2084.67

1988.09

ICC(1)
* p< .05, **p< .01

.10

223

Table 9: Mean comparison of performance scores for those who list workload as a barrier
to performance and those who do not (n=810)

Comparisons

Mean
performance
scores
3.42

Std Deviation

Workload is NOT
.25
listed as a barrier
Workload IS listed as 3.48
.24
a barrier
Total
3.45
.25
* The mean difference is significant at the .05 level

224

95% CI
Lower Bound Upper
Bound
3.40

3.45

3.45

3.50

3.43

3.47

Table 10: Hypothesis 3: HLM models for performance predicting WLB. (n=740)
Model 1
Fixed effect:
Unconditional
Means Model
with Controls

Model 2
Fixed effect:
Model with
Performance at
Predictor

Model 3
Random effects:
Model with
Performance at
Predictor

Intercept
Standard error
Performance
Standard error

3.42**
.16

3.40**
.17
-.23
.14

3.40**
.17
-.23
.14

Control: Team size

.05*

.05*

.05*

Standard error
Control: Function
Standard error
Random effects
Intercept
Standard error
Performance
Standard error

.02
-.11
.09

.02
-.15
.10

.02
-.12
.10

.11**
.03

.12**
.03

.11**
.03
.35
.31

Residual
Standard error
Model Fit Statistics

.67**
.04

.64**
.04

.64
.03

Deviance

2071.29

1887.81

1886.51

AIC

2075.29

1893.81

1894.51

BIC

2084.67

1907.61

1912.92

ICC(1)
* p< .05, **p< .01

.10

Fixed Effects

225

Table 11: Mean comparison of retention scores for those who list
workload as a barrier to performance and those who do not (n=809)

95% CI
Comparisons

Mean
Ret
Scores

Std
Dev

Lower Bound

Upper Bound

Workload is
NOT listed
as a barrier

3.66

0.82

3.72

3.86

Workload IS
listed as a
barrier

3.79

0.77

3.57

3.74

Total

3.74

0.8

3.67

3.79

* The mean difference is significant at the .05 level

226

Table 12: Hypothesis 6: HLM models for WLB and performance
predicting retention. Model 1= Null, Model 2 = WLB Fixed, Model 3
= Performance Fixed, Model 4 = Interaction (n=809)
Fixed Effects
Intercept
Standard error
WLB
Standard error
Performance
Standard error
PerfXWLB
Standard error

Model 1
3.62**
.12

Model 2
2.92**
.17
.22**
.04

Model 3
4.37**
.43

Control: Team size
Standard error

.02
.02

.00
.00

.00
.00

.00
.01

Random effects
Intercept
Standard error

.62**
.04

.55**
.03

.58**
.03

.53**
.03

Residual
Standard error

.02
.05

.03
.02

.19
.21

Interaction Residual
Interaction Std error
Model Fit Statistics
Deviance
AIC
BIC

.46
.28
-.12
.08
.04
.02

1946.56
1952.57
1966.65

1897.71
1928.52
1924.52

1747.92
1755.92
1774.34

1717.05
1725.05
1743.44

-.19
.13

* p< .05, **p< .01

227

Model 4
4.91**
1.66
-.16
.43
-.54
.47
.10
.12

Table 13: Hypothesis 25: HLM models for workload and Manager
WLB predicting WLB, with Manager WLB moderating individual
workload (n=688)
Fixed Effects
Model 1
Model 2 Model 3 Model 4
Intercept
3.76**
3.77**
3.42**
3.42**
Standard error
(.05)
(.04)
(.15)
(.15)
Workload
-.59**
-.59**
-.38**
Standard error
(.06)
(.06)
(.25)
Manager WLB
.10*
.10*
Standard error
(.04)
(.04)
WorkloadXMgrWLB
-.06
Standard error
(.06)
Control: Team size
Standard error
Control: Function
Standard error
Random effects
Intercept
Standard error
Workload
Standard error

.12**
(.03)

.09**
(.03)

.09**
(.03)
.04
(.07)

.09**
(.03)
.04
(.07)

Residual
Standard error

.63**
(.04)

.56**
(.03)

.56**
(.03)

.56**
(.03)

Model Fit Statistics
Deviance
AIC
BIC

2070.53
2074.53
2083.92

1645.24
1649.24
1659.30

1643.58
1649.58
1663.16

1646.33
1653.33
1665.92

* p< .05, **p< .01

228

Table 14: Qualitative Coding Themes
Theme category

# Interviewees
who mention
Detachment
30
Manager
30
Satisfaction
30
Workload
30
Flexibility
29
Team
29
Agency
28
Culture
28
Performance
24
Career
23
Anxiety/ Stress
20
Commute/Global
20
Family
15
Top leaders
14
Managing
14
Rewards
13
Sustainability/Attrition 10

% Interviewees
who mention
100%
100%
100%
100%
97%
97%
93%
93%
80%
77%
67%
67%
50%
47%
47%
43%
33%

Targeted or Emergent Theme
Targeted
Targeted
Targeted
Targeted
Targeted
Targeted
Emergent
Targeted
Emergent
Emergent
Emergent
Emergent
Emergent
Targeted
Targeted
Emergent
Emergent

Table 15: Satisfaction Categories
Satisfaction Category
Individual is not satisfied
Individual is satisfied
WLB is not sustainable

# of Interviewees who mention
18
12
10

Table 16: Detachment Categories
Detachment Category
Does work in the evenings/weekends
Tries to set clear boundaries
Prefers looser boundaries
Hard to find time for personal life
Taking vacation is difficult
Cannot mentally unplug from work

# of Interviewees who mention
26
26
14
12
12
10

229

Table 17: Manager Categories
Manager Category
Manager is supportive
Manager is not a good role model
Manager does not communicate well
about WLB
Manager helps by pushing back/
reprioritizing
Manager is a good role model for WLB
Manager is not supportive
Manager cannot help on WLB
Never have communicated about WLB
with Manager, no data to make
judgments

Table 18: Workload Categories
Workload Category
Work ebbs and flows
Work nights and weekends
More work than anyone can possibly finish
Emails and meetings are hard to manage
Reactive tasks (firedrills, etc.)
Not enough human resources/staff

Table 19: Flexibility Categories
Flexibility Category
Location flexibility is an option
Flexibility means you work all the time
Satisfied with flexibility
Would like more flexibility
Flexibility not always good for
productivity

# of Interviewees who mention
20
16
15
13
10
10
9
5

# of Interviewees who mention
28
23
22
21
15
9

# of Interviewees who mention
23
16
15
13
11

230

Table 20: Agency Categories
Agency Category
Works at creating satisfactory balance
I want WLB, but feel I am "bad" at
getting it.
WLB is personal responsibility
Work is out of my control/reactive
Prioritizing work over balance

Table 21: Team/ Culture Categories
Team/ Culture Category
Team satisfaction is mentioned
Team is supportive
Team is full of high achievers who
expect immediate responses don't want
to be weakest link
Don't want to fall behind, appear weak
At Genericorp, performance is the
highest priority, WLB is not valued
Peers work all hours, expect instant
responses
Want to succeed, get promoted so must
compete w/peers
Culture of WLB support

Table 22: Performance Categories
Performance Category
Need to distinguish oneself,
expectations/competition are high
Performance management structure
rewards imbalance
Always being compared to other people,
have to work extra hard
Work quality suffers with overwork
Can do more work when imbalanced

# of Interviewees who mention
22
18
12
10
5

# of Interviewees who mention
25
25
21

18
17
16
10
5

# of Interviewees who mention
20
20
8
2
2

231

Table 23: Career Categories
Career Category
Concerns WLB is not compatible with
desired career progression
Don't want to appear weak, so don’t
pursue satisfying WLB
Worry about job security

Table 24: Anxiety/Stress Categories
Anxiety/Stress Category
Feeling anxious, frustrated, stressed
Speed of industry cause panic, pressure
Performance expectations, ambiguous
goals cause stress
Feeling tired, losing sleep

# of Interviewees who mention
19
5
4

# of Interviewees who mention
12
12
12
9

Table 25: Commute/Global Categories
Commute/ Global Category
Commute makes satisfaction difficult
Travel/ Global issues

# of Interviewees who mention
16
8

Table 26: Family Categories
Family Category
Family is a part of participant’s life
Mention having children
Not enough time with family
Family is understanding

# of Interviewees who mention
13
10
10
3

Table 27: Top Leaders
Top Leaders Category
Top leaders are not good role models
Top leaders say they support WLB

# of Interviewees who mention
9
8

232

Table 28: Managing Categories
Managing Category
Help by talking about problems
Help by doing something
Struggles with managing WLB of direct
reports

Table 29: Rewards Categories
Rewards Category
WLB is worse if don't feel work is being
rewarded
Work is intrinsically rewarding

# of Interviewees who mention
11
11
7

# of Interviewees who mention
11
8

Table 30: Sustainability/Attrition Categories
Sustainability/Attrition Category
# of Interviewees who mention
would need to leave Genericorp to get
7
better WLB
would need to internally transfer to get
6
better WLB

233

Table 31: Study 3 response rates
Mgr Status

Function

N

North America
Sampled
Responded
response rate
Non N. America
Sampled
responded
response rate
Total

Mgr

Non
Mgr

Tenure Eng
(avg
yrs)

Sales

G&A

1272
706
55.5%

180
123
68.3%

1092
583
53.4%

2.6
2.8

421
244
58.0%

442
232
52.5%

409
230
56.2%

803
498
62.0%
1204

190
119
62.6%
242

613
379
61.8%
962

3.2
3.1

258
175
67.8%
419

301
163
54.2%
395

244
158
64.8%
388

3.0

234

Table 32: Study 3 Sample Characteristics
Gender

Mgr Status

Function

N

North
America
Europe
Asia/Pac
Is.
Latin
America
Total

Men

Wm
en

Mgr

Not
Mgr

Age
(ag
yrs)

Job
level
(ag)

Eng

Sles

GA

32.9

Tenu
re
(ag
yrs)
2.8

706

346

299

123

583

4.3

244

232

230

245
194

135
107

109
87

70
37

175
157

36.4
34.3

3.2
3.4

4.8
4.3

90
71

75
53

79
69

59

33

26

12

47

32.8

2.7

4.5

14

35

10

1204

621

521

242

962

33.2

3.0

4.4

419

395

388

Table 33: Bonferroni Comparison for Retention Scores (n=1165)
Retention
WIL
95% CI
95% CI
Compariso Mean St Lo
Up
Mean Std LoB Up
ns
Fct
d
Bnd Bnd Fct
Err nd
Bnd
Diff. Er
.
Diff.
.
r
Eng :
.11
.0 -.03 .24 -.13
.06 -.27 .02
Sales
6
Eng : G
-.04
.0 -.17 .10 -.15* .06 -.29 -.00
&A
6
Sales:G& -.15* .0 -.29 -.01 -.02
.06 -.16 .13
A
6
* The mean difference is significant at the .05 level

235

Emotional Exhaustion
95% CI
Mean St Low Up
Fct
d
Bnd Bnd
Diff.
Er .
.
r
-.07
.0 -.21 .07
6
-.04
.0 -.18 .11
6
.03
.0 -.11 .18
6

Table 34: Correlation table
Job Demos
Job
Level

Tnre
Yrs

Mgr
Stat

Perf
Rate

Tm
Size

Tm
Tsk
Inter

Sex

Person Demos
Marit
al
Age
Stat

#
Child

Child
Care
Hrs

Person demos

Job demos

Job Level
Tenure in Yrs
0.24
Manager Status
0.57
0.17
Perf Rating
0.08
0.28
0.19
Team Size
-0.03 0.06
n/a
0.09
Team Task
Interdependence
0.15
0.05
0.20
0.09
-0.02
Gender
0.13
-0.05 0.03
-0.05 0.02
0.02
Age
0.53
0.19
0.31
-0.05 0.04
0.04
0.07
Marital Status
0.32
0.15
0.19
-0.01 0.06
-0.02 0.09
0.41
Number of
Children
0.37
0.07
0.21
-0.04 0.06
-0.03 0.12
0.54
0.46
Child Care Hrs
0.15
0.18
0.06
0.05
0.03
-0.07 -0.16 0.12
0.35
0.39
House Work Hrs
0.00
0.12
-0.02 -0.05 0.04
-0.04 -0.16 0.14
0.21
0.18
0.36
Note: Values greater than .06= p< .05, Values greater than .08 = **p< .01 , Sex:0=women, 1= men, Marital Status
0=single(including widowed/divorced), 1=married/partnered

236

Hse
wrk
Hrs

Table 34 (cont’d)
Job demos

Person Demos
Mari #
Chld
-tal
Chil Care
Stat
d
Hrs

Cntrl

Wrkld

Outcome

WLB

Job
Lev
Work Life
Interference
Work Life
Positive Spill
Retention
Emotional
Exhaustion
Subjective
Workload
Work Hours
Autonomy
Control of Work

Tnre
Yrs

Mgr
Stat

Perf
Rate

Tm
Size

Tm
Tsk
Inter

Sex

Age

0.14

0.05

0.14

0.10

-0.07

0.14

-0.03

0.07

-0.03

0.02

0.02

0.04

-0.05

-0.05

-0.01

-0.05

-0.01

0.07

0.00

-0.11

-0.02

-0.01

-0.05

-0.05

-0.02

-0.18

-0.06

-0.01

-0.03

0.03

-0.03

0.03

0.01

0.03

-0.05

-0.03

0.00

0.10

0.02

0.09

-0.09

0.08

-0.03

-0.05

-0.07

-0.10

-0.03

0.02

0.16

0.10

0.16

0.18

-0.10

0.19

-0.02

0.03

0.01

-0.01

0.05

0.02

0.25
0.13

-0.02
0.06

0.24
0.06

0.11
0.12

-0.07
-0.02

0.17
0.04

0.13
0.08

0.05
0.02

-0.07
0.05

0.00
0.05

-0.11
-0.01

-0.09
-0.06

-0.04

-0.06

-0.06

-0.06

0.06

-0.07

0.01

-0.05

0.06

0.02

-0.08

-0.04

-0.11

-0.09

-0.03

0.02

-0.01

-0.04

-0.17

-0.04

0.06

0.14
0.10

0.10
0.02

0.03
0.02

0.09

0.00

0.07

Boundary Management

Psychological
-0.16 -0.06 -0.11 -0.13 0.07 -0.10 0.02 -0.10 -0.01
Detachment
Actual
0.16 0.03 0.11 0.11 -0.07 0.14 0.05 0.03 -0.03
Integration
Segmentation
-0.14 0.04 -0.06 0.01 0.00 0.01 -0.07 -0.17 -0.02
Preference
Eve Work Style
0.21 0.04 0.15 0.11 -0.09 0.09 0.02 0.11 0.01
Wknd Wk Style 0.16 0.01 0.13 0.11 -0.06 0.06 0.03 0.14 -0.02
Vacation Work
0.17 0.00 0.13 0.06 -0.12 0.07 0.00 0.06 -0.03
Style
Note: Values greater than .06= p< .05, Values greater than .08 = **p< .01 , Sex:0=women, 1= men
237

Hse
wrk
Hrs

Table 34 (cont’d)
Job demos

Person Demos
Mari #
Chld
-tal
Chil Care
Stat
d
Hrs

Flexibility

Job
Lev
Flexibility
Satisfaction
Used Flexibility
Negotiated
Flexibility
Life Flex Able
Life Flex Will
Work Flex Able
Work Flex Will

Tnre
Yrs

Mgr
Stat

Perf
Rate

Tm
Size

Tm
Tsk
Inter

Hse
wrk
Hrs

Sex

Age

0.07
0.08

-0.01
0.04

0.01
0.00

0.02
0.00

0.02
0.05

0.07
0.00

0.11
0.09

-0.01
0.05

0.03
0.03

0.03
0.10

-0.09
0.06

-0.12
0.01

-0.01
-0.15
0.06
0.00
0.00

-0.01
-0.17
-0.07
0.05
0.07

-0.08
-0.06
0.07
-0.07
-0.04

0.03
-0.03
0.02
-0.01
0.01

0.03
-0.03
0.00
0.00
-0.04

0.03
0.00
0.05
0.04
0.09

0.00
0.02
0.04
0.07
0.03

-0.03
-0.18
0.09
-0.07
-0.04

0.02
-0.31
-0.04
0.04
0.00

0.05
-0.38
0.09
-0.02
-0.03

0.10
-0.42
-0.01
-0.05
-0.05

0.04
-0.24
-0.01
-0.09
-0.05

Support

Mgr Support
-0.05 -0.01 -0.05 0.04 0.00 0.03 0.02 -0.09 -0.03 -0.03 0.04 -0.02
Team Emotional
Support
-0.07 -0.07 -0.05 -0.06 -0.05 0.07 0.02 -0.14 -0.10 -0.08 -0.03 -0.10
Tm Instrumental
Support
-0.06 -0.01 -0.03 -0.03 0.01 0.13 0.07 -0.09 -0.02 -0.06 -0.07 -0.09
Note: Values greater than .06= p< .05, Values greater than .08 = **p< .01 , Sex:0=women, 1= men, Marital Status
0=single(including widowed/divorced), 1=married/partnered

238

Table 34 (cont’d)
WLB

Wrkld Outcome

WLB

WIL
Work Life Inter.
Work Life
Positive Spill
Retention
Emot. Exh.
Subj. Workload
Work Hours

Outcome
WL
Pos
Spill

Ret

Em.
Exh

Workload

Control

Boundary Management

Subj.
wrk

Cntrl
Auto. of
Wrk

Psy.
Dtch

Wor
k Hrs

Act.
Int.

Seg.
Pref.

Eve

Wkend

-0.18
-0.28

0.28

0.61

-0.24

-0.49

0.53

-0.06

-0.19

0.47

0.48

-0.07

-0.10

0.30

0.42

Boundary Management

Cntrl

-0.18 0.19 0.31 -0.33 -0.08 0.01
Autonomy
Control of Work -0.59 0.15 0.33 -0.60 -0.43 -0.26 0.34
-0.51 0.13 0.14 -0.35 -0.40 -0.46 0.07 0.34
Psych. Detach.
Actual
0.50 -0.08 -0.10 0.33 0.35 0.46 -0.06 -0.31 -0.60
Integration
Segmentation
0.12 0.01 -0.13 0.19 0.04 -0.14 -0.14 -0.10 0.21 -0.25
Preference
Evening Work
0.40 -0.08 -0.09 0.29 0.33 0.51 -0.05 -0.23 -0.61 0.59
Style
Weekend Work
0.39 -0.04 -0.07 0.25 0.28 0.52 -0.09 -0.24 -0.54 0.53
Style
Vacation Work
0.42 -0.10 -0.14 0.33 0.30 0.39 -0.13 -0.25 -0.53 0.45
Style
Note: Values greater than .06= p< .05, Values greater than .08 = **p< .01 , Sex:0=women, 1= men
239

-0.31
-0.26

0.68

-0.09

0.51

0.50

Vaca

Table 34 (cont’d)
WLB

Flexibility

WIL
Flexibility
Satisfaction
Used Flexibility
Negotiated
Flexibility
Life Flex Able
Life Flex Will
Work Flex Able
Work Flex Will

Outcomes
WL
Pos
Spill

Ret

Em.
Exh

Workload
Wor
Subj.
k
wrk
Hrs

Control
Cntrl
Auto
of
.
Wrk

Boundary Management
Psy.
Dtch

Act.
Int.

Seg.
Pref.

Eve

Wkend

Vaca

-0.38 0.13 0.27 -0.39 -0.21 -0.14 0.40
0.06 -0.06 -0.03 0.06 0.05 0.07 0.08

0.30 0.20 -0.18 -0.21 -0.17 -0.17 -0.26
-0.10 -0.05 0.11 -0.11 0.13 0.07 0.00

-0.07
-0.15
0.08
-0.43
-0.18

0.02
0.17
0.03
0.32
0.11

0.00
0.13
0.09
0.19
0.13

0.07
0.11
0.05
0.19
0.11

-0.06
-0.12
-0.02
-0.27
-0.05

0.05
-0.11
0.04
-0.20
-0.02

-0.03
0.08
0.24
-0.22
-0.10

0.12
0.08
0.01
0.33
0.13

0.03
0.10
-0.23
0.26
0.15

0.03
0.02
0.28
-0.19
-0.09

-0.14
-0.09
-0.33
-0.07
0.06

0.06
-0.02
0.31
-0.15
-0.11

0.00
0.05
0.38
-0.26
-0.16

-0.07
-0.07
0.25
-0.27
-0.20

Support

Mgr Support
-0.26 0.24 0.31 -0.32 -0.12 -0.12 0.36 0.26 0.15 -0.12 -0.07 -0.11 -0.13 -0.17
Team Emotional
Support
-0.22 0.18 0.16 -0.17 -0.06 -0.05 0.13 0.22 0.15 -0.08 -0.04 -0.09 -0.10 -0.10
Tm Instrumental
Support
-0.30 0.22 0.18 -0.25 -0.13 -0.10 0.21 0.27 0.21 -0.14 -0.01 -0.16 -0.16 -0.17
Note: Values greater than .06= p< .05, Values greater than .08 = **p< .01 , Sex:0=women, 1= men, Marital Status
0=single(including widowed/divorced), 1=married/partnered

240

Table 34 (cont’d)
Flexibility

Flexibility

Flex
Satis.
Flexibility
Satisfaction
Used Flexibility
Negotiated
Flexibility
Life Flex Able
Life Flex Will
Work Flex Able
Work Flex Will

Support
Used
Flex

Negot.
Flex

Life
Flex
Ability

Life
Flex
Will

Work
Flex
Ability

0.27
0.14
0.10

-0.10
-0.10

Work
Flex
Will

0.56

Mgr
support

Team
Emot
Sup.

0.12
0.19

0.37

0.21
0.03
0.46
0.15

-0.06
0.03
0.17
0.17

-0.02
0.03
0.20
0.14

Support

0.34
0.06
0.21
0.13
0.07
0.35
0.21
Mgr Support
Team Emotional
0.19
0.01
0.05
0.18
0.06
0.24
0.18
0.31
Support
Tm Instrumental
0.29
0.03
0.08
0.17
0.00
0.35
0.26
0.34
0.66
Support
Note: Values greater than .06= p< .05, Values greater than .08 = **p< .01 , Sex:0=women, 1= men, Marital Status
0=single(including widowed/divorced), 1=married/partnered

241

Team
Inst.
Sup.

Table 35: Confirmatory Factor Analysis for Work/Life Flexibility Ability and Willingness Scale (Using Direct Oblimin Rotation).
(n=1140)
Life
Flex
Able
Because of my personal life responsibilities, I cannot change my work schedule (for example
going in early or staying longer to finish work related responsibilities). [R]

.755

If the need arose, I could work late without affecting my personal life responsibilities.

.843

My personal life responsibilities would not prevent me from going into work early if the need
arose.

Life
Flex
Will

.819

My personal life responsibilities would not prevent me from going into work an extra day in order .763
to meet work responsibilities.
From a personal life standpoint, there is no reason why I cannot rearrange my schedule to meet
the demands of my work.

.737

I am willing to change plans with my friends and family so that I can finish a job assignment.

.826

I am willing to change vacation plans that I have made with friends and family to meet work
related responsibilities.

.777

While at home, I do not mind stopping what I am doing to complete a work related responsibility.

.609

I am not willing to cancel plans with my friends and family to deal with work related
responsibilities. [R]

.744

242

Work
Flex
Able

Work
Flex
Will

Table 35 (cont’d)
Life
Flex
Able

Life
Flex
Will

Work
Flex
Able

Work
Flex
Will

I am able to arrive and depart from work when I want in order to meet my personal life
responsibilities.

.788

If the need arose, I could leave work early to attend to personal life issues.

.844

If something came up in my personal life, it would be all right if I arrived to work late.

.801

While at work, I can stop what I am doing to meet responsibilities related to my personal life.

.600

.273

I am willing to take an extended lunch break so that I can deal with responsibilities relating to my
personal life.

.325

.553

Assuming it was all right with my manager, I would not mind arriving to work late so that I could
meet my personal life responsibilities.

.807
.813

If it became necessary in order to meet my personal life responsibilities I would be willing to
change the shift, or start stop times, that I normally work.
Note: loadings below .25 suppressed.

243

Table 36: Study 3 Unconditional Means Models for Work Life
Interference, Retention and Emotional Exhaustion (n=878)
Work Life
Balance

Retention

Emotional
Exhaustion

Intercept

2.81**

4.02**

2.66**

Standard error

(04)

(.03)

(.04)

Intercept

.03

.03

.05

Standard error

(.03)

(.02)

(.03)

Residual

.68**

.59**

.66**

Standard error

(.04)

(.04)

(.04)

Model Fit
Statistics
Deviance

1590.12

1526.41

1616.23

AIC

1594.12

1530.41

1620.23

BIC

1306.02

1539.35

1629.16

ICC(1)

.03

.04

.07

Fixed Effects

Random effects

* p< .05, **p< .01

244

Table 37: Hypothesis 1 and 17. Hierarchical linear regression
predicting work interference with life from used flexibility and the
interaction of used flexibility and work hours (n=991)
2

ΔR
.02**

B

SE

β

.00

.02

-.01

.00

.01

-.02

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Used flexibility

.00
.02

.02

.02

Step 3:
Work hours

.19**
.23

.02

.45**

Step 4: Interaction
Used flexibility X
Work hours
* p< .05, **p< .01

.00
-.01

.01

-.03

245

Table 38: Hypothesis 1 and 17. Hierarchical linear regression
predicting work interference with life from the interaction of
negotiated flexibility and subjective workload (n=991)
2

ΔR
.02**

B

SE

β

.01

.02

.01

.00

.01

.00

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Negotiated flexibility
(i-deals)
Step 3:
Subjective Workload

.01*
-.08

.02

-.10**
.24**

.61

.04

.50**

Step 4: Interaction
Negotiated flex X
Subj. workload
* p< .05, **p< .01

.00
-.04

.03

-.04

Table 39: Hypothesis 2: Hierarchical linear regression of subjective
workload predicting performance (n=991)
2

ΔR
.08**

B

SE

β

.03

.00

.26**

Step 1: controls
Tenure (yrs)
Step 2:
Subjective workload

.02**
.05

.01

.14**

* p< .05, **p< .01

246

Table 40: Hypothesis 2: Hierarchical linear regression of work
hours predicting performance (n=1000)
2

ΔR
.08**

B

SE

β

.03

.00

.28**

Step 1: controls
Tenure (yrs)
Step 2:
Work hours

.01**
.02

.00

.11**

* p< .05, **p< .01

Table 41: Hypothesis 3. Hierarchical linear regression predicting
work interference with life from performance (n=993)
2

ΔR
.02**

B

SE

β

Job level

.03

.02

.05

Team size

-.01

.01

-.05

Team task
interdependence
Step 2:

.13

.04

.10**

Performance

.19

Step 1: controls

.00
.12

.06

* p< .05, **p< .01

247

Table 42: Hypothesis 4: Hierarchical linear regression of subjective
workload predicting retention (n=998)
2

ΔR
.05**

B

SE

β

-.14

.05

-.08**

.37

.11

.11**

-.06

.01

-.15**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Subjective workload

.02**
-.17

.04

-.15**

* p< .05, **p< .01

Table 43: Hypothesis 4: Hierarchical linear regression of work
hours predicting retention (n=995)
2

ΔR
.05**

B

SE

β

-.14

.05

-.08**

.35

.11

.10**

-.06

.01

-.17**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Work hours

.01**
-.05

.01

-.10**

* p< .05, **p< .01

248

Table 44: Hypotheses 5, 6 and 7: Hierarchical linear regression
predicting retention from performance, work interference with life
and the interaction of work interference with life and performance
(n=969)
2

ΔR
.04**

B

SE

β

.37

.10

.11**

-.13

.05

-.07*

-.06

.01

-.17**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Performance

.00
.14

.10

.04*

Step 3:
Work interference
with life
Step 4: Interaction

.07**
-.25

.03

-.27**
.00

Perf X Work Inter Life
-.02

.11

-.01

* p< .05, **p< .01
Table 45: Hypothesis 8: Hierarchical linear regression of subjective
workload predicting emotional exhaustion (n=987)
2

ΔR
.03**

B

SE

β

Region (N. America/
all others)
Tenure

.25

.05

.15**

.02

.01

.05

Team size

.00

.01

-.01

Team task
interdependence
Step 2:

-.03

.04

-.03

Step 1: controls

.23**

Subjective workload
.55

.04

.46**

* p< .05, **p< .01

249

Table 46: Hypothesis 8: Hierarchical linear regression of work
hours predicting emotional exhaustion (n=985)
2

ΔR
.04**

B

SE

β

Region (N. America/
all others)
Tenure

.24

.06

.14**

.04

.01

.11**

Team size

-.01

.01

-.04

Team task
interdependence
Step 2:

-.01

.04

.00

Step 1: controls

.12**

Work hours
.15

.02

.30**

* p< .05, **p< .01
Table 47: Hypothesis 9 and 10: Hierarchical linear regression of
work life interference predicting emotional exhaustion and partial
mediation of work life balance on the relationship of subjective
workload predicting emotional exhaustion (n=986)
2

ΔR
.03**

B

SE

β

Region (N. America/
all others)
Tenure

.26

.05

.15**

.02

.01

.06*

Team size

.00

.00

-.01

Team task
interdependence
Step 2:

-.06

.03

-.05

Work life interference

.52

Step 1: controls

.37**
.03

.52**

Step 3:
Subjective workload

.03**
.24

.04

.20**

* p< .05, **p< .01

250

Table 48: Hypothesis 10: Hypothesis 9 and 10: Hierarchical linear
regression of work life interference predicting emotional exhaustion
and partial mediation of work life balance on the relationship of
work hours predicting emotional exhaustion (n=984)
2

ΔR
.04**

B

SE

β

Region (N. America/
all others)
Tenure

.26

.05

.15

.03

.01

.08

Team size

.00

.01

-.02

Team task
interdependence
Step 2:

-.04

.03

-.03

Work life interference

.60

Step 1: controls

.38**
.03

.61

Step 3:
Work hours

.00
.01

.02

.02

* p< .05, **p< .01

251

Table 49: Hypothesis 11: Hierarchical linear regression of the
interaction of Work interference with life and Manager support
predicting retention (n=1033)
2

ΔR
.04**

B

SE

β

-.14

.05

-.08**

.38

.10

.11**

-.05

.01

-.14**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Work interference
with life
Manager support for
WLB
Step 3: Interaction

.13**
-.19

.03

-.20**

.25

.03

.26**
.00

WIL X Mgr support
-.02

.03

-.02

* p< .05, **p< .01

252

Table 50: Hypothesis 12: Hierarchical linear regression of the
interaction of Work interference with life and team emotional
support predicting retention (n=923)
2

ΔR
.05**

B

SE

β

-.15

.05

-.09**

.35

.11

.10**

-.06

.01

-.16**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Work interference
with life
Team emotional
support for WLB
Step 3: Interaction
WIL X Team
emotional support
* p< .05, **p< .01

.07**
-.23

.03

-.24**

.10

.03

.10**
.00

.03

.04

.02

253

Table 51: Hypothesis 13: Hierarchical linear regression of the
interaction of Work interference with life and team emotional
support predicting emotional exhaustion (n=1033)
2

ΔR
.04**

B

SE

β

Region (N. America/
all others)
Tenure

.27

.05

.16**

.03

.01

.07**

Team Size

.00

.00

-.02

Team Task
Interdependence
Step 2:

-.03

.03

-.02

Step 1: controls

Work interference
with life
team emotional
support for WLB
Step 3: Interaction
WIL X team
Emotional support
* p< .05, **p< .01

.40**
.55

.03

.56**

-.19

.03

-.19**
.00

.00

.03

.00

254

Table 52: Hypothesis 14: Hierarchical linear regression of the
interaction of Work interference with life and team emotional
support predicting emotional exhaustion (n=924)
2

ΔR
.05**

B

SE

β

Region (N. America/
all others)
Tenure

.26

.05

.16**

.03

.01

.08**

Team Size

.00

.01

-.02

Team Task
Interdependence
Step 2:

-.04

.03

-.03

Step 1: controls

Work interference
with life
Team emotional
support for WLB
Step 3: Interaction
WIL X Team
emotional support
* p< .05, **p< .01

.37**
.59

.03

.60**

-.05

.03

-.04
.00

.01

.03

.01

255

Table 53: Hypotheses 15 and 17. Hierarchical linear regression
predicting work interference with life from used flexibility and the
interaction of used flexibility and subjective workload (n=979)
2

ΔR
.02**

B

SE

β

.01

.02

.01

.00

.01

.00

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Negotiated flexibility
(i-deals)
Step 3:
Subjective Workload

.01*
-.08

.02

-.10**
.24**

.61

.04

.50**

Step 4: Interaction
Negotiated flex X
Subj. workload
* p< .05, **p< .01

.00
-.04

.03

-.04

Table 54: Hypotheses 15 and 17. Hierarchical linear regression
predicting work interference with life from negotiated flexibility and
the interaction of negotiated flexibility and work hours (n=991)
2

ΔR
.02**

B

SE

β

.00

.02

.00

.00

.01

-.02

.07

.04

.06

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Negotiated flexibility
(i-deals)
Step 3:
Work hours

.00
-.04

.02

-.05
.19**

.23

.02

.45**

Step 4: Interaction
Negotiated flex X
Work hours
* p< .05, **p< .01

.00
-.02

.01

-.04
256

Table 55: Hypotheses 16 and 18. Hierarchical linear regression
predicting work interference with life from flexibility satisfaction
and from the interaction of flexibility satisfaction and subjective
workload (n=991)
2

ΔR
.02**

B

SE

β

.03

.02

.04

.00

.01

.00

.09

.04

.07*

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Flexibility
Satisfaction
Step 3:
Subjective workload

.17**
-.36

.03

-.31**
.16**

.51

.04

.42**

Step 4: Interaction
Flexibility
Satisfaction X
subjective workload
* p< .05, **p< .01

.00

-.06

.04

-.04

257

Table 56: Hypothesis 17. Hierarchical linear regression predicting
work interference with life from the interaction of used flexibility and
subjective workload (n=991)
B

SE

β

.00

.02

.01

.00

.01

-.01

.05

.04

.04

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Used flexibility

.00
.02

.02

.02

Step 3:
Subjective workload

2

ΔR
.02**

.23**
.60

.04

.49**

Step 4: Interaction
Used flexibility X Subj.
workload
-.03
* p< .05, **p< .01

.00
.03

-.03

258

Table 57: Hypothesis 17. Hierarchical linear regression predicting
work interference with life from the interaction of used flexibility
and work hours (n=991)
2

ΔR
.02**

B

SE

β

.00

.02

-.01

.00

.01

-.02

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Used flexibility

.00
.02

.02

.02

Step 3:
Work hours

.19**
.23

.02

.45**

Step 4: Interaction
Used flexibility X
Work hours
* p< .05, **p< .01

.00
-.01

.01

-.03

259

Table 58: Hypothesis 18. Hierarchical linear regression predicting
work interference with life from the interaction of flexibility
satisfaction and work hours (n=988)
2

ΔR
.02**

B

SE

β

.02

.02

.03

.00

.01

-.02

.10

.04

.08**

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.17**

Flexibility satisfaction
-.40

.03

-.34**

Step 3:

.14**

Work hours
.20

.02

.39**

Step 4: Interaction
Flexibility satisfaction
X Work hours
* p< .05, **p< .01

.00
.01

.02

.02

Table 59: Hypothesis 19: Hierarchical linear regression of the
predicting retention from used flexibility (n=987)
2

ΔR
.05**

B

SE

β

-.15

.05

-.09**

.36

.11

.10**

-.06

.01

-.17**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Used flexibility

.00
-.02

.02

-.02

* p< .05, **p< .01

260

Table 60: Hypothesis 19: Hierarchical linear regression of the
predicting retention from negotiated flexibility (n=975)
2

ΔR
.05**

B

SE

β

-.14

.05

-.08**

.35

.11

.10**

-.06

.01

-.16**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Negotiated flexibility
(i-deals)
* p< .05, **p< .01

.00*
.04

.02

.06*

Table 61: Hypothesis 20: Hierarchical linear regression of the
predicting retention from flexibility satisfaction (n=986)
2

ΔR
.05**

B

SE

β

-.11

.05

-.07*

.31

.11

.09**

-.06

.01

-.17**

Step 1: controls
Function (Sales/all
others)
Region (Latin
America/ all others)
Tenure (yrs)
Step 2:
Flexibility satisfaction

.07**
.28

.03

.26**

* p< .05, **p< .01

261

Table 62: Hypothesis 21: Hierarchical linear regression of the
predicting emotional exhaustion from used flexibility (n=988)
2

ΔR
.04**

B

SE

β

.25

.06

.14**

.04

.01

.10**

Step 1: controls
Region (N. America/
all others)
Tenure
Team size
-.01

.01

-.06

Team task
interdependence
Step 2:

.06

.04

.05

Used flexibility

.05

.00
.03

.06

* p< .05, **p< .01
Table 63: Hypothesis 21: Hierarchical linear regression of the
predicting emotional exhaustion from negotiated flexibility (n=976)
2

ΔR
.04**

B

SE

β

.26

.06

.15

.04

.01

.10

Step 1: controls
Region (N. America/
all others)
Tenure
Team size
-.01

.01

-.05

Team task
interdependence
Step 2:

.06

.04

.05

Negotiated flexibility

-.05

.00
.03

-.06

* p< .05, **p< .01

262

Table 64: Hypothesis 22: Hierarchical linear regression of the
predicting emotional exhaustion from flexibility satisfaction (n=987)
2

ΔR
.04**

B

SE

β

.21

.06

.12**

.03

.01

.08**

Step 1: controls
Region (N. America/
all others)
Tenure
Team size
-.01

.01

-.05

Team task
interdependence
Step 2:

.09

.04

.07*

Flexibility satisfaction

-.45

.16**
.04

-.39**

* p< .05, **p< .01
Table 65: Hypothesis 23. Hierarchical linear regression predicting
work interference with life from the interaction of used flexibility
and segmentation preferences (n=988)
2

ΔR
.02**

B

SE

β

.05

.02

.07

-.01

.01

-.05

.15

.04

.12**

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.00

Used flexibility
.04

.03

.05

Step 3:
Segmentation
Preference
Step 4: Interaction
Used flex X
Segmentation
Preference
* p< .05, **p< .01

.01**
.11

.03

.12**
.00

.02

.03

.03

263

Table 66: Hypothesis 23. Hierarchical linear regression predicting
work interference with life from the interaction of negotiated
flexibility and segmentation preferences (n=976)
2

ΔR
.02**

B

SE

β

.04

.02

.07

-.01

.01

-.05

.15

.04

.12**

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.01**

Negotiated flex
-.05

.03

-.06**

Step 3:
Segmentation
Preference
Step 4: Interaction
Neg. Flex X
Segmentation
Preference
* p< .05, **p< .01

.01**
.11

.03

.11**
.00

-.02

.03

-.03

264

Table 67: Hypothesis 24.1. Hierarchical linear regression predicting
work interference with life from the interaction of subjective
workload and actual integration (n=987)
B

SE

β

-.01

.02

-.01

.00

.01

.01

.02

.03

2

ΔR
.02**

.02

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.25**

Actual integration
.32

.03

.37**

Step 3:

.11**

Subjective workload
.46

.04

.37**

Step 4: Interaction
Actual integration X
Subj. workload
* p< .05, **p< .01

.01**

.10

.03

.08**

265

Table 68: Hypothesis 24.1 Hierarchical linear regression
predicting work interference with life from the interaction of
work hours and actual integration (n=985)
2

ΔR
.02**

B

SE

β

-.01

.02

-.01

.00

.01

.00

.04

.04

.03

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Actual integration

.25**
.34

.03

.39**

Step 3:
Work hours

.06**
.13

.05

.25**

Step 4: Interaction
Actual integration X
Work hours
* p< .05, **p< .01

.00
.00

.01

.02

Table 69: Hypothesis 24.2. Hierarchical linear regression predicting
work interference with life from the interaction of actual integration
and preferences for segmentation(n=1004)
B

SE

β

-.01

.02

2

ΔR
.02**

-.01

Step 1: controls
Job level
Team size
.00

.01

.01

Team task
interdependence
Step 2:

.02

.03

.02

Actual integration

.43

.25**
.03

.50**

Step 3:
Actual integration X
pref for segmentation
* p< .05, **p< .01

.00

.04

.03

.04

266

Table 70: Hypothesis 24.3 and 24.4. Hierarchical linear regression
predicting work interference with life from life flexibility ability and
the interaction of subjective workload and life flexibility ability
(n=991)
2

ΔR
.02**

B

SE

β

.00

.02

.00

.00

.01

-.01

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.01**

Life Flexibility Ability
-.07

.03

-.07**

Step 3:

.22**

Subjective workload
.59

.04

.49**

Step 4: Interaction
Life Flexibility Ability
X Subj. workload
* p< .05, **p< .01

.00
.02

.04

.02

267

Table 71: Hypothesis 24.3 and 24.4. Hierarchical linear regression
predicting work interference with life from life flexibility ability and
the interaction of work hours and life flexibility ability (n=989)
2

ΔR
.02**

B

SE

β

-.02

.02

-.03

-.01

.01

-.03

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.01**

Life Flexibility Ability
-.17

.03

-.18**

Step 3:

.20**

Hours worked
.24

.02

.47**

Step 4: Interaction
Life Flexibility Ability
X Hrs worked
* p< .05, **p< .01

.00
.00

.02

.00

268

Table 72: Hypothesis 24.5 and 24.6. Hierarchical linear
regression predicting work interference with life from life
flexibility willingness and the interaction of subjective workload
and life flexibility willingness (n=991)
B

SE

β

.00

.02

.01

.00

.01

.00

.04

.04

.04

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Life flexibility
willingness
Step 3:
Subjective workload

.01**
.09

.03

.09
.23**

.60

.04

.49

Step 4: Interaction
Life Flexibility
willingness X Subj.
workload
* p< .05, **p< .01

2

ΔR
.02**

.00

.04

.04

.03

269

Table 73: Hypothesis 24.5 and 24.6. Hierarchical linear
regression predicting work interference with life from life
flexibility willingness and the interaction of work hours and life
flexibility willingness (n=989)
2

ΔR
.02**

B

SE

β

Job level

.00

.02

.00

Team size

.00

.01

-.02

.06

.04

.05

Step 1: controls

Team task
interdependence
Step 2:
Life flexibility
willingness
Step 3:
Hours worked
Step 4: Interaction
Life flexibility
willingness X Hrs
worked
* p< .05, **p< .01

.01**
.00

.03

.00
.18**

.23

.02

.44
.00

.01

.02

.02

270

Table 74: Hypothesis 24.7 and 24.8. Hierarchical linear regression
predicting work interference with life from work flexibility ability
and the interaction of subjective workload and work flexibility
ability (n=991)
2

ΔR
.02**

B

SE

β

.01

.02

.02

.00

.01

-.01

.07

.03

.06**

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.21**

Work flexibility ability
-.46

.04

-.37**

Step 3:

.16**

Subjective workload
.52

.03

.43**

Step 4: Interaction
Work flexibility ability
X Subj. workload
-.07
* p< .05, **p< .01

.00
.05

-.04

271

Table 75: Hypothesis 24.6 and 24.7. Hierarchical linear regression
predicting work interference with life from work flexibility ability
and the interaction of work hours and work flexibility ability
(n=989)
2

ΔR
.02**

B

SE

β

.01

.02

.02

-.01

.01

-.03

.09

.04

.07**

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.21**

Work flexibility ability
-.48

.04

-.38**

Step 3:
.11**
Hours worked
.18

.02

.35**

Step 4: Interaction
Work flexibility ability
X Hrs worked
.01
* p< .05, **p< .01

.00
.02

.01

272

Table 76: Hypothesis 24.9 and 24.10. Hierarchical
linear regression predicting work interference with life
from work flexibility willingness and the interaction of
subjective workload and work flexibility willingness
(n=991)
B

SE

β

.00

.02

.00

.00

.01

.00

.07

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Work flexibility
willingness
Step 3:

2

ΔR
.02**

.05**
-.27

.04

.21**
.23**

Subjective
workload
.59
Step 4: Interaction

.04

Work flexibility
willingness X
Subj. workload
* p< .05, **p<
.01

.05

-.09

.49**
.00

-.05

273

Table 77: Hypothesis 24.9 and 24.10. Hierarchical linear
regression predicting work interference with life from work
flexibility willingness and the interaction of work hours and
work flexibility willingness (n=989)
B

SE

β

Job level

.00

.02

.00

Team size

-.01

.01

-.03

.08

.04

2

ΔR
.02**

.06*

Step 1: controls

Team task
interdependence
Step 2:
Work flexibility
willingness
Step 3:
Hours worked
Step 4: Interaction
Work flexibility
willingness X Hrs
worked
* p< .05, **p< .01

.05**
-.22

.04

-.16**
.16**

.21

.02

.42**
.00

.00

.02

.00

274

Table 78: Hypothesis 24.11 and 24.12. Hierarchical linear
regression predicting work interference with life from control of work
and the interaction of subjective workload and control of work
(n=990)
2

ΔR
.02**

B

SE

β

.02

.02

.04

.00

.00

.00

.06

.03

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.35**

control of work
-.61

.04

-.47**

Step 3:

.07**

Subjective workload
.37

.04

.30**

Step 4: Interaction
control of work X
Subj. workload
* p< .05, **p< .01

.00*
-.11

.05

-.06*

275

Table 79: Hypothesis 24.11 and 24.12. Hierarchical linear
regression predicting work interference with life from control of work
and the interaction of work hours and control of work (n=988)
2

ΔR
.02**

B

SE

β

.01

.02

.02

.00

.00

-.01

.06

.03

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.36**

control of work
-.68

.03

-.52**

Step 3:

.08**

Hours worked
.16

.01

.31**

Step 4: Interaction
control of work X Hrs
worked
* p< .05, **p< .01

.00
-.01

.02

-.02

276

Table 80: Hypothesis 24.13 and 24.14. Hierarchical linear
regression predicting work interference with life from autonomy and
the interaction of subjective workload and autonomy (n=991)
2

ΔR
.02**

B

SE

β

.02

.02

.03

.00

.01

-.01

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.05**

autonomy
-.20

.03

-.17**

Step 3:

.21**

Subjective workload
.57

.04

.47**

Step 4: Interaction
autonomy X Subj.
workload
* p< .05, **p< .01

.01*
-.10

.04

-.07*

277

Table 81: Hypothesis 24.13 and 24.14. Hierarchical linear
regression predicting work interference with life from autonomy and
the interaction of work hours and autonomy (n=989)
2

ΔR
.02**

B

SE

β

.01

.02

.02

-.01

.01

-.03

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.05**

autonomy
-.25

.04

-.21**

Step 3:

.18**

Hours worked
.22

.02

.44**

Step 4: Interaction
autonomy X Hrs
worked
* p< .05, **p< .01

.00
-.01

.02

-.02

278

Table 82: Hypothesis 24.15 and 24.16. Hierarchical linear
regression predicting work interference with life from psychological
detachment and the interaction of subjective workload and psych
detachment (n=991)
2

ΔR
.02**

B

SE

β

-.01

.02

-.02

.00

.01

.00

.04

.03

.03

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.26**

psych detachment
-.38

.03

-.37**

Step 3:

.10**

Subjective workload
.44

.04

.36**

Step 4: Interaction
psych detachment X
Subj. workload
* p< .05, **p< .01

.01**
-.14

.04

-.11**

279

Table 83: Hypothesis 24.15 and 24.16. Hierarchical linear
regression predicting work interference with life from psychological
detachment and the interaction of work hours and psych detachment
(n=989)
2

ΔR
.02**

B

SE

β

-.02

.02

-.02

.00

.01

-.01

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.26**

psych detachment
-.41

.03

-.41**

Step 3:

.06**

Hours worked
.14

.02

.27**

Step 4: Interaction
psych detachment X
Hrs worked
* p< .05, **p< .01

.00
-.01

.02

-.01

280

Table 84: Hypothesis 24.17 and 24.18. Hierarchical linear
regression predicting work interference with life from evening work
style and the interaction of subjective workload and evening work
style (n=991)
2

ΔR
.02**

B

SE

β

-.01

.02

-.02

.00

.01

.01

.05

.04

.04

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.15**

evening work style
.24

.03

.26**

Step 3:

.15**

Subjective workload
.51

.04

.42**

Step 4: Interaction
evening work style X
Subj. workload
* p< .05, **p< .01

.01*
.09

.04

.07*

281

Table 85: Hypothesis 24.17 and 24.18. Hierarchical linear
regression predicting work interference with life from evening work
style and the interaction of work hours and evening work style
(n=989)
2

ΔR
.02**

B

SE

β

-.01

.02

-.02

.00

.01

-.01

.07

.04

.06

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.15**

evening work style
.23

.03

.24**

Step 3:

.08**

Hours worked
.16

.02

.32**

Step 4: Interaction
evening work style X
Hrs worked
* p< .05, **p< .01

.00
.01

.02

.02

282

Table 86: Hypothesis 24.19 and 24.20. . Hierarchical linear
regression predicting work interference with life from weekend work
style and the interaction of subjective workload and weekend work
style (n=991)
2

ΔR
.02**

B

SE

β

-.01

.02

-.01

.00

.01

-.01

.04

.04

.04

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.16**

Weekend work style
.26

.03

.27**

Step 3:

.14**

Subjective workload
.49

.04

.40**

Step 4: Interaction
Weekend work style X
Subj. workload
* p< .05, **p< .01

.01**
.12

.04

.09**

283

Table 87: Hypothesis 24.19 and 24.20. Hierarchical linear
regression predicting work interference with life from weekend work
style and the interaction of work hours and weekend work style
(n=989)
2

ΔR
.02**

B

SE

β

-.01

.02

-.01

.00

.01

-.02

.07

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.16**

Weekend work style
.24

.03

.25**

Step 3:

.07**

Hours worked
.16

.02

.32**

Step 4: Interaction
Weekend work style X
Hrs worked
* p< .05, **p< .01

.00
.00

.02

-.01

284

Table 88: Hypothesis 24.21 and 24.22. Hierarchical linear
regression predicting work interference with life from vacation work
style and the interaction of subjective workload and vacation work
style (n=990)
2

ΔR
.02**

B

SE

β

-.01

.02

-.02

.00

.01

.02

.04

.04

.03

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.18**

vacation work style
.29

.03

.31**

Step 3:

.14**

Subjective workload
.51

.04

.41**

Step 4: Interaction
vacation work style X
Subj. workload
* p< .05, **p< .01

.01**
.10

.03

.08**

285

Table 89: Hypothesis 24.21 and 24.22. Hierarchical linear
regression predicting work interference with life from vacation work
style and the interaction of work hours and vacation work style
(n=988)
2

ΔR
.02**

B

SE

β

-.02

.02

-.02

.00

.01

.01

.06

.04

.05

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.19**

vacation work style
.29

.03

.31**

Step 3:

.08**

Hours worked
.17

.02

.33**

Step 4: Interaction
vacation work style X
Hrs worked
* p< .05, **p< .01

.00
-.01

.02

-.01

286

Table 90: Hypothesis 25 Hierarchical linear regression predicting
work interference with life from the interaction of manager work life
interference and subjective workload (n=285)
2

ΔR
.02**

B

SE

β

.05

.04

.07

Step 1: controls
Job level
Team size
.01

.02

.03

Team task
interdependence
Step 2:

.06

.07

.05

Manager work life
interference
Step 3:

.02

.00
.05

.02
.27**

Subjective workload
.62

.06

.51**

Step 4: Interaction
Manager work life
interference X Subj.
workload
* p< .05, **p< .01

.01

.09

.07

.07

287

Table 91: Hypothesis 25 : Hierarchical linear regression predicting
work interference with life from the interaction of manager work life
interference and work hours (n=284)
2

ΔR
.02**

B

SE

β

.00

.04

.00

Step 1: controls
Job level
Team size
.02

.02

.06

Team task
interdependence
Step 2:

.05

.07

.04

Manager work life
interference
Step 3:

.05

.00
.05

.05
.17**

Hours worked
.24

.03

.43**

Step 4: Interaction
Manager work life
interference X work
hrs
* p< .05, **p< .01

.01

.05

.03

.08

288

Table 92: Hypothesis 26. Hierarchical linear regression
predicting work interference with life from the interaction of
manager instrumental support and subjective workload (n=979)
B

SE

.01

.02

.02

.00

.01

-.01

.07

.04

2

ΔR
.02**

β

.06*

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Manager
instrumental support
Step 3:
Subjective workload

.11**
-.26

.03

-.25**
.19**

.56

.04

.46**

Step 4: Interaction
Manager support X
Subj. workload
* p< .05, **p< .01

.00
-.05

.04

-.04

289

Table 93: Hypothesis 26. Hierarchical linear regression
predicting work interference with life from the interaction of
manager instrumental support and work hours (n=980)
B

SE

.00

.02

.00

.00

.01

-.03

.09

.04

2

ΔR
.02**

β

.07*

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Manager
instrumental support
Step 3:
Hours worked

.11**
-.29

.03

-.28**
.15**

.21

.02

.40**

Step 4: Interaction
Manager support X
work hrs
* p< .05, **p< .01

.00
.03

.02

.04

290

Table 94: Study 1 Unconditional Means Models to test for Aggregation
for Team Emotional Support, Team Instrumental Support and Work
Flexibility Ability (n=507)
Team Emotional
Support

Team
Instrumental
Support

Work Flexibility
Ability

Intercept

3.73**

3.78**

3.94**

Standard error

(.04)

(.03)

(.03)

Random effects
Intercept

.07**

.04

.04*

Standard error

(.02)

(.02)

(.02)

Residual

.49**

.45**

.37**

Standard error

(.03)

(.03)

(.02)

Model Fit
Statistics
Deviance

1301.79

1224.22

1350.04

AIC

1305.79

1228.22

1354.05

BIC

1314.51

1236.93

1363.15

ICC(1)

.13

.08

.10

Fixed Effects

* p< .05, **p< .01

291

Table 95: Hypothesis 28. Hierarchical linear regression predicting work
interference with life from the interaction of aggregated team emotional
support and subjective workload (n=507)
B

SE

β

-.01

.03

-.01

.00

.01

.01

.03

.05

.02

Step 1: controls
Job level

2

ΔR
.02*

Team size
Team task
interdependence
Step 2:
Aggregated Team Emot
Support
Step 3:
Subjective workload

.01
-.07

.05

-.05
.21**

.56

.05

.47**

Step 4: Interaction
Aggregated Team Emot
Support X Subj.
workload
* p< .05, **p< .01

.00

-.12

.07

-.06

292

Table 96: Hypothesis 28. Hierarchical linear regression predicting work
interference with life from the interaction of aggregated team emotional
support and work hours (n=507)
B

SE

β

-.03

.03

-.05

.01

.01

.02

.07

.05

.06

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Aggregated Team Emot
Support
Step 3:
Hours worked

.01*
-.06

.06

-.04
.18**

.22

.02

.43

Step 4: Interaction
Aggregated Team Emot
Support X work hrs
* p< .05, **p< .01

2

ΔR
.02**

.00
-.05

.03

-.06

293

Table 97: Hypothesis 28. Hierarchical linear regression predicting work
interference with life from the interaction of aggregated work flexibility
ability and subjective workload (n=507)
B
Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:
Aggregated Work Flex
Ability
Step 3:
Subjective workload
Step 4: Interaction
Aggregated Work Flex
Ability X Subj. workload
* p< .05, **p< .01

SE

β

.01

.03

.01

.00

.01

.00

.04

.05

2

ΔR
.02**

.04
.06**

-.24

.05

-.19**
.19**

.54

.05

.45**
.00

-.11

.07

-.06

294

Table 98: Hypothesis 28. Hierarchical linear regression predicting
work interference with life from the interaction of aggregated work
flexibility ability and work hours (n=507)
2

ΔR
.02**

B

SE

β

-.02

.03

-.03

.01

.01

.02

.09

.05

.07

Step 1: controls
Job level
Team size
Team task
interdependence
Step 2:

.06**

Aggregated Work Flex
Ability
-.24
Step 3:

.05

-.19**
.16**

Hours worked
.21

.02

.41**

Step 4: Interaction
Aggregated Work Flex
Ability X work hrs
.00
* p< .05, **p< .01

.00
.03

.00

295

Appendix E: Figures

Figure 1: Model of the Hypothesized Relationships and How They Fit Within Theoretical
Frameworks
Support
Post conflict factors:
 Manager support
 Team support
Box 1
Role Theory/ COR Theory
WLB
Antecedents:
 Workload
 Performance

Outcomes:
 Retention
 Emotional
Exhaustion

Box 2

Border Theory/ Normative Theory
Pre conflict factors:
 Segmentation/integration
 Satisfaction with flexibility
 Use of flexibility
 WLB management techniques
 Team/manager norms
Box 3

296

Figure 2. Model showing COR Hypotheses

H8+

H1-

Workload

2-

H
H2+

WLB

Emotional
Exhaustion

H9-

H4-

H3-

H6+
H7

Performance

Retention
Intentions

H5+

297

Figure 3. Model Showing Moderation Hypotheses of WLB on Outcomes

Manager
Social
Support
H11

H13

Team
Social
Support
H12

H14

H9-

Emotional
Exhaustion

WLB

H6+

298

Retention
Intentions

Figure 4. Model of Border Management Hypotheses and Moderations

WLB
management
techniques
H24

H1-

Workload

Segmentation
/ Integration

WLB

H9H6-

Emotional
Exhaustion

H23

H17/18

H15/16+

Flexibility
Use/
Satisfaction

H21/22-

H19/20+

299

Retention
Intentions

Figure 5. Model and Manager and Team Hypotheses

Manager
WLB

Team WLB
H

Retention
Intentions

Perceptions
Perceptions
27
26 of Team
of Manager
WLB
WLB
Expectations
Expectations

H25

H26

H27

Workload

H28

WLB
H1-

300

Figure 6: Graph of Proposed Moderation for Hypothesis 7

Hypothesis 7: WLB will moderate the relationship of performance with retention intentions such
that those with higher WLB and higher performance will be more likely to indicate intentions to
stay with the company than those with lower performance.

Retention Intentions
Retention Intentions

5
5
4.5
4.5
4
4
3.5
3.5
3
3
2.5
2.5

High WLB
High WLB

2
2

Low WLB
Low WLB

1.5
1.5
1
1
0.5
0.5
0
0

Low Perf Perf
Low

High Perf
High Perf

301

Figure 7: Graph of Proposed Moderation for Hypothesis 11

Hypothesis 11: Manager support for balance will moderate the relationship between WLB and
retention such that those who report more supportive managers will report higher intention to
remain with the company as compared to those with similar levels of WLB with less supportive
managers

Retention Intentions
Retention Intentions

5
5
4.5
4.5
4
4
3.5
3.5
3
3
2.5
2.5

High Manager
High Manager
High Manager
Support
Support
Support

2
2
1.5
1.5

Low Manager
Low Manager
Low Manager
Support
Support
Support

1
1
0.5
0.5
0
0
Low
Low WLB WLB

High
High WLB

WLB

302

Figure 8. Graph of Proposed moderation for Hypothesis 12
H12: Team support for balance will moderate the relationship between WLB and retention such
that those who report more supportive teams will report greater intention to remain with the
company as compared to those with similar levels of WLB but with less supportive teams.
5

Retention Intentions
Retention Intentions

4.5
4
3.5
3
2.5

High Team
High Team Support
Support
Low Team Support
Low Team
Support

2
2
1.5
1.5
1
1
0.5
0.5
0
0
Low
Low WLB WLB

High WLB
High WLB

303

Figure 9: Graph of Proposed Moderation for Hypothesis 13

H13: Manager support for WLB will moderate the relationship between WLB and emotional
exhaustion such that those who receive more support will be buffered from the negative effects
of low WLB on emotional exhaustion.

Emotional Exhaustion
Emotional Exhaustion

5
5
4.5
4.5
4
4
3.5
3.5
3
3

High Manager
High Manager
Support
Support

2.5
2.5
2
2

Low Manager
Low Manager
Support
Support

1.5
1.5
1
1
0.5
0.5
0
0
Low WLB
Low WLB

High WLB
High WLB

304

Figure 10: Graph of Proposed Moderation for Hypothesis 14

H14: Team support for WLB will moderate the relationship between WLB and emotional
exhaustion such that those who receive more support will be buffered from the negative effects
of low WLB on emotional exhaustion.

Emotional Exhaustion
Emotional Exhaustion

5
5
4.5
4.5
4
4
3.5
3

High Team
Support
High team support

2.5
2
2

Low team support
Low Team
Support

1.5
1.5
1
1
0.5
0.5
0
0

Low Workload
Low WLB
Low WLB

High Workload
High WLB
High WLB

305

Figure 11: Graph of Proposed Moderation for Hypothesis 17

H17: Frequency of use of flexible arrangements will moderate the relationship between workload
and WLB such that those who report more use of flexible arrangements will also report greater
work-life balance despite high workloads.

5
5
4.5
4.5

WLB
Work LIfe Balance

4
4
3.5
3.5
3
3

Uses flexibility
Satisfied with
Flexibility

2.5
2.5
2
2

Does not use
Not Satisfied with
flexibility
Flexibility

1.5
1.5
1
1
0.5
0.5
0
0

Low Workload

High Workload

306

Figure 12: Graph of Proposed Moderation for Hypothesis 18

WLB
Work LIfe Balance

H18: Satisfaction with flexible arrangements will moderate the relationship between workload
and WLB such that those who report more use of flexible arrangements will also report greater
work-life balance despite high workloads.
5
5
4.5
4.5
4
4
3.5
3.5
3
3
2.5
2.5
2
2
1.5
1.5
1
1
0.5
0.5
0
0

Satisfied with
Satisfied with
flexibility
Flexibility
Not satisfied with
Not Satisfied with
flexibility
Flexibility

Low Workload
Low Workload

High Workload
High Workload

307

Figure 13: Graph of Proposed Moderation for Hypothesis 23

H23: A preference for segmentation will moderate the relationship between flexibility and WLB
such that those who prefer segmentation will not report higher WLB satisfaction with increases
in flexibility.

WLB
Work Life Balance

5
5
4.5
4.5
4
4
3.5
3.5
3
3

Preference
Preference for
Segmentation
Segmentation

2.5

2

Preference for
Preference
integration
Integration

1.5
1.5
1
1
0.5
0.5
0
0

Low Flexibility
Low Flexibility

High Flexibility
High Flexibility

308

Figure 14: Graph of Proposed Moderation for Hypothesis 24

H24: WLB management techniques will moderate the relationship between workload and WLB
such that those who use more techniques will be buffered from the negative effects of high
workload on WLB.

5
5
4.5
4.5

WLB
Work Life Balance

4
3.5
3.5

3

Fewer WLB
Less WLB Mgmt
management
Techniques
techniques

2.5
2.5
2
2
1.5
1.5

More WLB Mgmt
More WLB
Techniques
management
techniques

1
1
0.5
0.5
0
0

Low Workload
Low Workload

High Workload
High Workload

309

Figure 15: Graph of Proposed Moderation for Hypothesis 25

H25: Manager work-life balance will moderate the relationship of subordinate workload and
work-life balance such that those with more balanced managers will also be more balanced,
regardless of level of workload
.

5

5

4.5

4.5

WLB

Work Life Balance

4
4
3.5
3.5
3
Low manager

2.5
2.5

Low Manager WLB
WLB

2
2

High Manager WLB
High manager

1.5
1.5

WLB

1

1

0.5

0.5

0

0

Low Workload

Low Workload

High Workload

High Workload

310

Figure 16: Graph of Proposed Moderation for Hypothesis 26

H26: Perceptions of manager expectations about work-life balance will moderate the relationship
of subordinate workload and work-life balance such that those with mangers what expect balance
will also be more balanced, regardless of level of workload.

5
5
4.5
4.5

WLB
Work Life Balance

4
4
3.5
3.5
Manager
Manager expectations
of WLB
Expectations of WLB

3
2.5
2

Manager
Manager expectations
of work first
Expectations of Work
First

1.5
1.5
1
1
0.5
0.5
0
0

Low Workload
Low Workload

High Workload
High Workload

311

Figure 17: Graph of Proposed Moderation for Hypothesis 27

H27: Team WLB will moderate the relationship between individual workload and individual
satisfaction with balance such that those in teams with higher balance will be buffered from the
negative effects of workload on balance.

5
5
4.5
4.5

WLB
Work Life Balance

4
3.5
3
2.5

Low Team WLB
Low Team WLB

2

High Team WLB
High Team WLB

1.5
1
1
0.5
0.5
0
0
Low Workload
Low Workload

High Workload
High Workload

312

Figure 18: Graph of Proposed Moderation for Hypothesis 28.

H28: Perceptions of team WLB norms will moderate the relationship between individual
workload and individual satisfaction with balance such that those in teams with norms for higher
balance will be buffered from the negative effects of workload on balance.

5
5
4.5
4.5

WLB
Work Life Balance

4
4
3.5
3.5
3

Team norms support
Team Norms Support
WLB
WLB
Team norms do Not
Team Norms Do not
support WLB
Support WLB

2.5
2.5
2
2
1.5
1.5
1
1
0.5
0.5
0
0

Low Workload
Low Workload

High Workload
High Workload

313

Figure 19: Graph of Frequency of Hours Spent on Housework

Frequency of endorsement

200

Hours Spent on Housework

175
150
125
100
75
50
25
0
1

5

10

5

20

25

30

35

40

Hours spent on housework

314

45

50

55

60

Figure 20: Graph of Frequency of Hours Spent on Child and/or Elder Care

300

Hours Spent on Child /Elder Care
Frequency of endorsement

250

200

150

100

50

0
1

5

10

15

20

25

30

35

40

45

Hours spent on child/eldercare

315

50

55

60

65

70

Work Life Interference

Figure 21: Interaction of Actual Integration and Subjective Workload Predicting Work
Interference with Life.

5
4
High actual
integration

3
2

Low actual
integration

1
0
Low Workload

High Workload

316

Work Life Interference

Figure 22: Interaction of Control of Work and Subjective Workload Predicting Work
Interference with Life.

5
4
High control of
work

3
2

Low actual
integration

1
0
Low Workload

High Workload

317

Work Life Interference

Figure 23: Interaction of Autonomy and Subjective Workload Predicting Work
Interference with Life.

5
4
3
High autonomy

2
Low autonomy

1
0
Low Workload

High Workload

318

Figure 24: Interaction of Psychological Detachment and Subjective Workload Predicting
Work Interference with Life.

Work Life Interference

4
3.5
3
High psychological
detachment

2.5
2

Low psychological
detachment

1.5
1
.5
0
Low Workload

High Workload

:

319

Figure 25: Interaction of Evening Work Style and Subjective Workload Predicting Work
Interference with Life.

Work Life Interference

4
3.5
3
Works a lot in the
evenings

2.5
2

Does not work a
lot in the evenings

1.5
1
.5
0
Low Workload

High Workload

320

Figure 26: Interaction of Weekend Work Style and Subjective Workload Predicting Work
Interference with Life

Work Life Interference

4
3.5
3
2.5

Works a lot on the
weekends

2
1.5

Does not work a lot on
the weekends

1
.5
0
Low Workload

High Workload

321

Figure 27: Interaction of Vacation Work Style and Subjective Workload Predicting Work
Interference with Life.

Work Life Interference

4
3.5
3
2.5
Works a lot on
vacation

2
1.5

Does not work a lot on
vacation

1
.5
0
Low Workload

High Workload

322

Appendix F: Supplemental Analyses

Combined regression of interactions proposed in Hypotheses 15-18.
Table 99. Supplemental Analyses. Hypotheses 15-18: Hierarchical
linear regression of interaction of flexibly and workload predicting
work life interference (n=812)
2

ΔR
.02**

B

SE

β

Job level

.00

.01

.00

Team size

.01

.01

.01

Team task
interdependence

.06

.03

.05

Step 1: controls

Step 2:

.18**

Flexibility satisfaction

-.32

.03

-.28*

Negotiated flexibility

-.03

.04

-.04**

Used flexibility

.05

.02

.06

Step 3:

.22**

Subjective workload

.40

.04

.33**

Work hours

.14

.02

.28**

Step 4:

.00

Flex Sat X Subj.
Workload
Flex Sat X Wrk Hours

-.05

.05

-.03

.02

.02

.03

Neg Sat X Subj.
Workload
Neg Sat X Wrk Hours

-.03

.04

-.03

.00

.01

-.01

.04

.01

.02

-.03

Used Sat X Subj.
.02
Workload
Used Sat X Wrk Hours -.01
* p< .05, **p< .01

323

Table 100. Supplemental Analyses: Hypotheses 24.1-24.10: Hierarchical
linear regression of interaction of flexibility perceptions and workload
predicting work life interference (n=812)
B

SE

2

ΔR

β

Step 1: controls

.02**

Job level

0.00

0.00

0.00

Team size

-0.01

0.02

-0.02

Team task
interdependence

0.03

0.03

0.02

Step 2:

.39**

Actual Integration

0.24

0.03

0.28**

Life Flex Ability

-0.09

0.03

-0.09**

Life Flex Willing

-0.03

0.03

-0.03

Work Flex Ability

-0.39

0.04

-0.31**

Work Flex Willing

0.04

0.04

0.03

Step 3:

.10**

Subjective workload

0.36

0.04

0.29**

Work hours

0.13

0.05

0.25**

Step 4:

.10**

Act Int X Subj. Wrkld

0.09

0.04

0.08**

Act Int X Wrk Hours

-0.02

0.02

-0.10**

LFA X Subj. Wrkld

0.03

0.04

0.02

LFA X Wrk Hours

-0.02

0.02

-0.04

LFW X Subj. Wrkld

-0.02

0.04

-0.01

LFW X Wrk Hours

0.03

0.02

0.05

WFA X Subj. Wrkld

-0.08

0.06

-0.05

324

Table 100. (cont.)
WFA X Wrk Hours

0.01

0.03

0.02

WFW X Subj. Wrkld

0.02

0.06

0.01

WFW X Wrk Hours

0.03

0.03

0.04

* p< .05, **p< .01

325

Table 101. Supplemental Analyses. Hypotheses 24.11-14:
Hierarchical linear regression of interaction of control and autonomy
and workload predicting work life interference (n=812)
2

ΔR
.22**

B

SE

β

Job level

0.00

0.00

0.01

Team size

0.01

0.02

0.01

Team task
interdependence

0.04

0.03

0.03

Step 1: controls

Step 2:

.36**

Autonomy

-0.03

0.03

-0.03

Control of Work

-0.56

0.04

-0.43**

Step 3:

.12**

Subjective workload

0.27

0.04

0.22**

Work hours

0.13

0.01

0.25**

Step 4:

.00

Autonomy X Work hrs 0.00

0.02

0.00

Autonomy X Subj
Wrkload
Control X Work hrs

-0.04

0.05

-0.03

0.00

0.02

0.00

Control X Subj
Wrkload
* p< .05, **p< .01

-0.05

0.05

-0.03

326

Table 102. Supplemental Analyses. Hypotheses 102. 24.15-22: Hierarchical
linear regression of interaction of detachment and workload predicting work life
interference (n=812)
B

SE

β

Step 1: controls

2

ΔR

.22**

Job level

0.00

0.00

0.02

Team size

-0.02

0.02

-0.03

Team task interdependence

0.01

0.03

0.01

Step 2:

.31**

Psych detachment

-0.25

0.04

-0.24**

Evening work

-0.03

0.04

-0.03

Weekend work

0.04

0.04

0.04

Vacation work

0.14

0.03

0.15**

Subjective workload

0.37

0.04

0.31**

Work hours

0.09

0.02

0.17**

Step 3:
.10**

Step 4:

.02**

Psych det X Work hrs

0.01

.02

0.02

Psych det X Subj Wrkload

-0.14

0.05

-0.11**

Wknd wrk X Work hrs

0.08

0.05

0.06

Wknd wrk X Subj Wrkload

-0.05

0.02

-0.08*

Eve wrk X Work hrs

0.05

0.03

0.09

Eve wrk X Subj Wrkload

-0.04

0.06

-0.03

Vaca wrk X Wrk hrs

-0.02

0.02

-0.04

Vaca wrk X Subj Wrkload

0.00

0.05

0.00

* p< .05, **p< .01

327

Table 103. Supplemental Analyses. Hypotheses 25-28: Hierarchical linear
regression of interaction of team relevant variables and workload predicting
work life interference (n=256)
B

SE

β

Step 1: controls

2

ΔR
.02

Job level

0.01

0.02

0.02

Team size

0.03

0.04

0.04

Team task interdependence

0.00

0.06

0.00

Step 2:

.10**

Manager WLI

-0.01

0.05

-0.01

Manager support

-0.14

0.05

-0.13**

Team Work flexibility ability
(aggregated)
Team emotional support (aggregated)

-0.22

0.08

-0.17**

0.24

0.09

0.17**

Step 3:

.29**

Subjective workload

0.45

0.07

0.37**

Work hours

0.17

0.03

0.33**

Step 4:

.03

Mgr WLI X Subj Wrkload

0.07

0.08

0.05

Mgr WLI X Wrk hrs

0.00

0.03

0.00

Mgr Sup X Subj Wrkload

0.11

0.09

0.07

Mgr Sup X Wrk hrs

-0.04

0.04

-0.06

Team WFA X Subj Wrkload

0.09

0.05

0.12

Team WFA X Wrk hrs

-0.22

0.13

-0.13

Team Emot Sup X Subj Wrkload

0.13

0.14

0.06**

Team Emot Sup X Wrk hrs

-0.15

0.05

-0.18

* p< .05, **p< .01

328

Table 104. Supplemental Analyses. MANCOVA of Function and Region with Bonferroni posthoc tests in relation to Emotional Exhaustion, Work-Life Interference and Retention
Variable

df

df
error

F

Retention

3

1135

4.99** .00

Emotional
Exhaustion

3

1135

7.55** .00

Work-Life 3
Interference

1135

.948

sig

.42

Region

Means

99% Confidence Interval

N. America
EMEA
Asia/PacIslands
Latin America
N. America
EMEA
Asia/PacIslands
Latin America
N. America
EMEA
Asia/PacIslands
Latin America

3.99
3.94
3.84
4.32
2.70
2.42
2.61
2.39
2.87
2.94
2.81
2.95

Low bound
3.93
3.84
3.72
4.09
2.63
2.32
2.49
2.15
2.80
2.83
2.68
2.70

Up bound
4.05
4.04
3.95
4.55
2.76
2.53
2.73
2.62
2.93
3.05
2.93
3.20

*Significant at the .05 level; **significant at the .01 level

Variable

df

df
error

F

sig

Function

Means

99% Confidence Interval

Retention

2

1135

3.91*

.02

Emotional
Exhaustion

2

1135

Eng
Sales
General
Eng
Sales
General
Eng
Sales
General

4.12
3.90
4.04
2.48
2.58
2.53
2.78
2.92
2.97

Low bound
4.00
3.80
3.90
2.35
2.45
2.38
2.64
2.82
2.82

Work-Life 2
Interference

.47
.75

1135

.12
2.12

*Significant at the .05 level; **significant at the .01 level

329

Up bound
4.24
4.00
4.18
2.61
2.69
2.68
2.91
3.02
3.12

REFERENCES

330

REFERENCES

Adams, G. A., & Jex, S. M. (1999). Relationships between time management, control, work–
family conflict, and strain. Journal of Occupational Health Psychology, 4, 72–77
Abdallah, T. (2009). Prevalence and predictors of burnout among Palestinian social
workers. International Social Work, 52(2), 223-233.
Allen TD (2001) Family supportive work environments: the role organizational perceptions.
Journal of Vocational Behavior, 58(4), 414-435.
Allen, Tammy D., Herst, D. E. L., Bruck, C. S., & Sutton, M. (2000). Consequences associated
with work-to-family conflict: A review and agenda for future research. Journal of
Occupational Health Psychology, 5(2), 278-308.
Aryee, S. (1993). Dual-earner couples in Singapore: An examination of work and nonwork
sources of their experienced burnout. Human Relations, 46(12), 1441-1468.
Ashforth, B. E., Kreiner, G. E., & Fugate, M. (2000) All in a day’s work: Boundaries and
micro role transitions. The Academy of Management Review, 25(3), 472-491.
Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization.Academy of
management review, 14(1), 20-39.
Baba, V. V., Jamal, M., & Tourigny, L. (1998). Work and mental health: A decade in Canadian
research. Canadian Psychology/Psychologie Canadienne, 39(1-2), 94-107.
Bailey, S. J. (2007). Family and work role-identities of divorced parents: The relationship of
role balance to well-being. Journal of Divorce and Remarriage, 46(3-4), 63-82.
Bakker, A. B., Demerouti, E., & Schaufeli, W. B. (2002). Validation of the Maslach burnout
inventory--general survey: An internet study. Anxiety, Stress & Coping: An
International Journal, 15(3), 245-260.
Baltes, B. B., & Heydens-Gahir, H. (2003). Reduction of work-family conflict through the use
of selection, optimization, and compensation behaviors. Journal of Applied
Psychology, 88(6), 1005-1018.
Banjeree D., & Perrucci, C. (2012). Employee Benefits and Policies: Do They Make a
Difference for Work/Family Conflict?. Journal Of Sociology & Social Welfare,39(3),
131-147.
331

Barnes, C. M., & Hollenbeck, J. R. (2009). Sleep deprivation and decision-making teams:
Burning the midnight oil or playing with fire?. Academy of Management Review, 34(1),
56-66.
Barnett, R. C., & Hyde, J. S. (2001) Women, men, work, and family. American Psychologist,
56(1), 781-796.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology,51(6), 1173-1182
Baron, R. M., & Kenney, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual strategic and statistical considerations. Journal of
Personality and Social Psychology, 51(6), 1173-1182.
Batt, R. & Valcour, P. M. (2003). Human resources practices as predictors of work- family
outcomes and employee turnover. Industrial Relations, 42(2), 189-220.
Beauregard, T. A., & Henry, L. C. (2009). Making the link between work-life balance practices
and organizational performance. Human resource management review, 19(1), 9-22.
Beauregard, T. A., Ozbilgin, M., & Bell, M. P. (2009). Revisiting the social construction of
family in the context of work. Journal of Managerial Psychology, 24(1), 46-46-65
Becker, W. J. & Cropanzano, R. (2011) Dynamic aspects of voluntary turnover: an integrated
approach to curvilinearity in the performance-turnover relationship. Journal of Applied
Psychology, 96(2), 233-246.
Behson, S. J. (2002) Coping with family-to-work conflict: the role of informal work
accommodations to family. Journal of Occupational Health Psychology, 7(4), 324-341.
Bertaux, D. (1981). From the life-history approach to the transformation of sociological
practice. In Daniel Bertaux (Ed.), Biography and society: The life history approach in
the social sciences (pp.29-45). London: Sage.
Binnewies, C., Sonnentag, S., & Mojza, E. J. (2009). Daily performance at work: Feeling
recovered in the morning as a predictor of day‐level job performance. Journal of
Organizational Behavior, 30(1), 67-93.
Blomme, R. J., van Rheede, A., & Tromp, D. M. (2010). The use of the psychological contract
to explain turnover intentions in the hospitality industry: A research study on the impact
of gender on the turnover intentions of highly educated employees. The International
Journal of Human Resource Management, 21(1), 144-162
332

Boles, J. S., Johnston, M. W., & Hair, J. F. (1997). Role stress, work-family conflict and
emotional exhaustion: Inter-relationships and effects on some work-related
consequences. Journal of Personal Selling & Sales Management, 17(1), 17-28.
Bolino, M. C., Turnley, W. H., Gilstrap, J. B., & Suazo, M. M. (2010). Citizenship under
pressure: What's a "good soldier" to do? Journal of Organizational Behavior, 31(6),
835-855
Bowen, G. L., & Orthner, D. K. (1991).Effects of organizational culture on fatherhood. New
York, NY, US: Springer Publishing Co.
Boyar, S. L., Maertz, C. P., Pearson, A. W., & Keough, S. (2003). Work-family conflict: A
model of linkages between work and family domain variables and turnover
intentions. Journal of Managerial Issues, 15(2), 175-190.
Brayfield, A. H., & Crockett, W. H. (1955). Employee attitudes and employee
performance. Psychological Bulletin, 52(5), 396-424
Brough , P. & O’Driscoll, M. P (2010) Organizational interventions for balancing work and
home demands: An overview. Work & Stress, 24(3), 280-297.
Bulger, C. A., Matthews, R. A., & Hoffman, M. E. (2007) Work and personal life boundary
management: Boundary strength, work/personal life balance, and the segmentationintegration continuum. Journal of Occupational Health Psychology, 12(4), 365-375.
Butler, A. B., Grzywacz, J. G., Bass, B. L., & Linney, K. D. (2005). Extending the demandscontrol model: A daily diary study of job characteristics, work-family conflict and workfamily facilitation. Journal of Occupational and Organizational Psychology, 78(2), 155169.
Butler, A. B., Grzywacz, J. G., Ettner, S. L., & Liu, B. (2009) Workplace flexibility, self
reported health and health care utilization. Work & Stress, 23(1), 45-59.
Byron, K. (2005). A meta-analytic review of work-family conflict and its antecedents. Journal
of Vocational Behavior, 67(2), 169-198
Campion, M. A. (1991). Meaning and measurement of turnover: Comparison of alternative
measures and recommendations for research. Journal of Applied Psychology,76(2), 199212
Caplan, R. D. (1987). Person-environment fit theory and organizations: Commensurate
dimensions, time perspectives, and mechanisms. Journal of Vocational behavior, 31(3),
248-267.
Carlson D. S. & Perrewé, P. L. (1999) The role of social support in the stressor–strain
333

relationship: An examination of work–family conflict, Journal of Management 25, 513–
540.
Casper, W. J., Eby, L. T, Bordeaux, C., Lockwood, A., & Lambert, D. (2007) A review of
research methods in IO/OB work-family research. Journal of Applied Psychology, 92(1),
28-43.
Chalofsky, N. (2008). Work-life programs and organizational culture: The essence of
workplace community. Organization Development Journal, 26(1), 11-18.
Civian, Janet T., Richman, A. L., Shannon, L. L., Shulkin, S., & Brennan, R. T. (2008). Using
a multi-organization database: Research methods, strengths, and limitations. Community,
Work & Family, 11(2), 139-148.
Clark, S. C. (2000) Work/family border theory: a new theory of work/family balance. Human
Relations, 53(6), 747-770.
Claessens, B. J., Van Eerde, W., Rutte, C. G., & Roe, R. A. (2004). Planning behavior and
perceived control of time at work. Journal of Organizational Behavior, 25(8), 937-950.
Costa, G., Sartori, S., & Åkerstedt, T. (2006). Influence of flexibility and variability of
working hours on health and well-being. US: Informa Healthcare.
Creswell, John (1998). Qualitative inquiry and research design: Choosing among five
traditions. Thousand Oaks, CA: Sage.
Dawley, D., Houghton, J. D., & Bucklew, N. S. (2010). Perceived organizational support and
turnover intention: The mediating effects of personal sacrifice and job fit. The Journal of
Social Psychology,150(3), 238-257
Deery, M. (2008). Talent management, work-life balance and retention strategies. International
Journal of Contemporary Hospitality Management,20(7), 792-806.
Desrochers, S., Andreassi, J., & Thompson, C. (2004). Identity Theory1.Organization
Management Journal, 1(1), 61-69.
Desrochers, S., Hilton, J. M., & Larwood, L. (2005). Preliminary validation of the work-family
integration-blurring scale. Journal of Family Issues,26(4), 442-466
Diefendorff, M. M., Brown, D. J., Kamin, A. M., & Lord, R. G. (2002). Examining the roles of
job involvement and work centrality in predicting organizational citizenship behaviors
and job performance. Journal of Organizational Behavior, 23(1), 93-108.
Doorewaard, H. & Benschop, Y. (2004) HRM and organizational change: an emotional
endeavor. Journal of Organizational Change Management, 16(3), 272-286.
334

Doorewaard, H. & van Bijsterveld, M. (2001). The osmosis of ideas: an analysis of the
integrated approach to IT management from a translation theory perspective.
Organization, 18(3), 55-76.
Dunham, R. B., Pierce, J. L. & Castañeda. M. B. (1987) Alternative Work Schedules: Two
Field Quasi-Experiments. Personnel Psychology, 40(2), 215-42.
Eaton, S. C. (2003), If You Can Use Them: Flexibility Policies, Organizational Commitment,
and Perceived Performance. Industrial Relations: A Journal of Economy and Society,
42: 145–167.
Eby, L. T., Casper, W. J., Lockwood, A., Bordeaux, C., & Brinley, A. (2005) Work and family
research in IO/OB: Content analysis and review of the literature (1980-2002). Journal of
Vocational Behavior, 66(1), 124-197.
Edwards, J. R. (1996). An examination of competing versions of the person–environment fit
approach to stress. Academy of Management Journal, 39(2), 292-339.
Fleetwood, S. (2007) Why Work-Life Balance Now? The International Journal of Human
Resource Management, 18(3), 387-400.
Ford, M. T., Heinen, B. T., & Langkamer, K. L. (2007) Work and Family Satisfaction and
Conflict: A Meta-Analysis of Cross-Domain Relations. Journal of Applied
Psychology, 92(1) 57-80.
Fortune website (2008). 100 Best Companies to Work For. Retrieved May 1, 2011, from
http://money.cnn.com/magazines/fortune/bestcompanies/2008/index.html
Fortune website (2009). 100 Best Companies to Work For. Retrieved May 1, 2011, from
http://money.cnn.com/magazines/fortune/bestcompanies/2009/index.html
Fortune website (2010). 100 Best Companies to Work For. Retrieved May 1, 2011, from
http://money.cnn.com/magazines/fortune/bestcompanies/2010/index.html
Fortune website (2011). 100 Best Companies to Work For. Retrieved May 1, 2011, from
http://money.cnn.com/magazines/fortune/bestcompanies/2011/index.html
Fritz, C., & Sonnentag, S. (2006). Recovery, well-being, and performance-related outcomes:
The role of workload and vacation experiences. Journal of Applied Psychology, 91(4),
936.
Frone, M. R., Russell, M. & Barnes G. M. (1996) Work–family Conflict, Gender, and HealthRelated Outcomes: A Study of Employed Parents in Two Community Samples. Journal
of occupational health psychology 1(1) 57-69.
335

Frone, M. R., Russell, M., & Cooper, M. L. (1992) Antecedents and outcomes of work-family
conflict: testing a model of the work-family interface. Journal of Applied Psychology,
77(1), 65-78.
Gan, Y., & Cheung, F. M. (2010). From proactive personality to organizational citizenship
behavior: Mediating role of harmony. Psychological Reports, 106(3), 395-404.
Gardner, D. G. (1986) Activation Theory and Task Design: An Empirical Test of several New
Predictions. Journal of Applied Psychology, 71(3), 411-418.
Gardner, D. G., Dunham, R. B., Cummings, L. L., & Pierce, J. L. (1989). Focus of attention at
work: Construct definition and empirical validation. Journal of Occupational
Psychology, 62(1), 61-77.
Garger, E. M. (1999). Holding on to high performers: A strategic approach to
retention. Compensation and benefits Management, 15, 10-17.
Geurts, S. A. E., Kompier, M. A. J., Roxburgh, S., & Houtman, I. L. D. (2003). Does workhome interference mediate the relationship between workload and well-being? Journal
of Vocational Behavior, 63(3), 532-559.
Giardini, A., & Kabst, R. (2008). Effects of work-family human resource practices: A
longitudinal perspective. The International Journal of Human Resource
Management, 19(11), 2079-2094.
Gilboa, S., Shirom, A., Fried, Y., & Cooper, C. (2008) A meta-analysis of work demand
stressors and job performance: examining main and moderating effects. Personnel
Psychology, 61(2), 227-271.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for
qualitative research. Chicago: Aldine.
Golden, T. D. Veiga, J. F. & Dino, R. N. (2008) The impact of professional isolation on
teleworker job performance and turnover intentions: Does time spent teleworking,
interacting face-to-face, or having access to communication-enhancing technology
matter? Journal of Applied Psychology, 93(6), 1412-1421.
Grandey, A. A., & Cropanzano, R. (1999). The conservation of resources model applied to
work–family conflict and strain. Journal of Vocational Behavior, 54(2), 350-370.
Grawitch, M. J. & Barber, L. K. (2010) Work flexibility or nonwork support? Theoretical and
empirical distinction for work-life initiatives. Consulting Psychology Journal: Practice
and Research, 62(3), 169-188.
Greenhaus, J. H., & Beutell, N. J. (1985). Sources of conflict between work and family
336

roles. Academy of management review, 10(1), 76-88.
Greenhaus, J. H., Bedeian, A. G., & Mossholder, K. W. (1987) Work experiences, job
performance and feelings of personal and family well-being. Journal of Vocational
Well-Being, 31(2), 200-215.
Greenhaus, J. H., Parasuraman, S. & Collins K. M. (2001) Career Involvement and Family
Involvement as Moderators of Relationships between work–family Conflict and
Withdrawal from a Profession. Journal of occupational health psychology, 6(2), 91100.
Greenhaus, J. H. & Powell, G. N. (2006) When work and life are allies: a theory of work
family enrichment. The Academy of Management Journal, 31(1), 72-92.
Greenhaus, J. H. & Singh, R. (2003) Work-family linkages. From Sloan Work and Family
Research Network. http://wfnetwork.bc.edu/encyclopedia_entry.php?id=263.
Gregory, A. & Milner, S. (2009) Editorial: Work-life balance: a matter of choice? Gender,
Work and Organization, 16(1), 1-14.
Grzywacz, J. G., Carlson, D. S., & Shulkin, S. (2008) Schedule flexibility and stress: Linking
formal flexible arrangements and perceived flexibility to employee health. Community,
Work & Family, 11(2), 199-214.
Grzywacz, J. G., & Fuqua, J. (2000). The social ecology of health: Leverage points and
linkages. Behavioral Medicine, 26(3), 101-115.
Guest, D. E. (1997). Human resource management and performance: a review and research
agenda. International Journal of Human Resource Management,8(3), 263-276.
Guzzo, R. A., & Dickson, M. W. (1996). Teams in organizations: Recent research on
performance and effectiveness. Annual review of psychology, 47(1), 307-338.
Halbesleben, J. R. B., Harvey, J., & Bolino, M. C. (2009) Too engaged? A conservation of
resources view of the relationship between work engagement and work interference with
family. Journal of Applied Psychology, 94(6), 1452-1465.
Halbesleben, Jonathon R. B. (2006). Sources of social support and burnout: A meta-analytic
test of the conservation of resources model. Journal of Applied Psychology,91(5), 11341145.
Hall, G. B., Dollard, M. F., Tuckey, M. R., Winefield, A. H., & Thompson, B. M. (2010). Job
demands, work–family conflict, and emotional exhaustion in police officers: A
longitudinal test of competing theories. Journal of Occupational and Organizational
Psychology, 83(1), 237-250.
337

Hamilton, E. (2010) Bringing work home: Advantages and challenges of telecommuting.
White paper for Center for Work and Family. Retrieved from website May 15, 2010,
http://www.bc.edu/content/dam/files/centers/cwf/research/publications/pdf/BCCWF_Tel
ecommuting_Paper.pdf.
Hammer, L. B., Kossek, E. E., Anger, W. K., Bodner, T., & Zimmerman, K. L. (2011)
Clarifying work-family intervention processes: the roles of work-family conflict and
family-supportive supervisor behaviors. Journal of Applied Psychology, 96(1), 134-150.
Hammer, L. B., Kossek, E. E., Yragui, N. L., Bodner, T. E., & Hanson, G. C. (2009).
Development and validation of a multidimensional measure of family supportive
supervisor behaviors (FSSB). Journal of Management, 35(4), 837-856.
Hammer, T. H., Saksvik, P. O., Nytro, K., Torvatn, H., & Bayazit, M. (2004) Expanding the
psychosocial work environment: Workplace norms and work-family conflict as
correlates of stress and health. Journal of Occupational Health Psychology, 9(1), 83-97.
Hausknecht, J. P., Rodda, J., & Howard, M. J. (2009) Targeted employee retention:
Performance-based and job-related differences in reported reasons for staying. Human
Resource Management, 48(2), 269-288.
Heaney, Catherine A., Price, Richard H., & Rafferty, J. (1995). Increasing coping resources at
work: A field experiment to increase social support, improve work team functioning,
and enhance employee mental health. Journal of Organizational Behavior, 16(4), 335353.
Hetland, H., Sandal, G. M., & Johnsen, T. B. (2007). Burnout in the information technology
sector: Does leadership matter? European Journal of Work and Organizational
Psychology, 16(1), 58-75.
Hill, E. J., Erickson, J. J., Holmes, E. K. & Ferris, M. (2010) Workplace flexibility, work hours
and work-life conflict: finding an extra day or two. Journal of Family Psychology, 24(3),
349-358.
Hill, E. J., Hawkins, A. J., Ferris, M., & Weitzman, M. (2001). Finding an Extra Day a Week:
The Positive Influence of Perceived Job Flexibility on Work and Family Life
Balance*. Family Relations, 50(1), 49-58.
Hinkin, T. R. (1998). A brief tutorial on the development of measures for use in survey
questionnaires. Organizational Research Methods, 1, 104-125.
Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing
stress. American Psychologist, 44(3), 513-524.
338

Hobson, C. J., Delunas, L., & Kesic, D. (2001). Compelling evidence of the need for corporate
work/life balance initiatives: results from a national survey of stressful life‐events.
Journal of Employment Counseling, 38(1), 38-44.
Hollenbeck, John R., Ilgen, D. R., Tuttle, D. B., & Sego, D. J. (1995). Team performance on
monitoring tasks: An examination of decision errors in contexts requiring sustained
attention. Journal of Applied Psychology, 80(6), 685-696.
Hofmann, D. A., Griffin, M. A., & Gavin, M. B. (2000). The application of hierarchical linear
modeling to organizational research.
Hornung, S., Rousseau, D. M. & Glaser, J. (2008) Creating flexible work arrangements
through idiosyncratic deals. Journal of Applied Psychology, 93(3), 655-664.
Hornung, S., Rousseau, D.M., & Glaser, J. (2009). Why supervisors make idiosyncratic deals:
antecedents and outcomes of i-deals from a managerial perspective. Journal of
Managerial Psychology, 24(8), 738-760.
Hornung, S., Rousseau, D.M., Weigl, M., Glaser, J., & Angerer, P. (2009). Idiosyncratic deals
of hospital physicians: Negotiating for flexibility and development. Presentation at the
14th European Congress on Work and Organisational Psychology. Santiago de
Compostela, Spain, 05/14.
Hsu, Y. R. (2011). Work-family conflict and job satisfaction in stressful working
environments. International Journal of Manpower, 32(2), 233-248.
Hunter, E. M., Perry, S. J., Carlson, D. S., & Smith, S. A. (2010) Linking team resources to
work-family enrichment and satisfaction. Journal of Vocational Behavior, 77(2), 304312.
Hunter, Emily M., Perry, S. J., Carlson, Dawn S., & Smith, S. A. (2010). Linking team
resources to work–family enrichment and satisfaction. Journal of Vocational
Behavior, 77(2), 304-312.
Iacovides, A., Fountoulakis, K. N., Kaprinis, S., & Kaprinis, G. (2003). The relationship
between job stress, burnout and clinical depression. Journal of Affective
Disorders, 75(3), 209-221.
Ilies, R., Dimotakis, N., & De Pater, I. E. (2010). Psychological and physiological reactions to
high workloads: Implications for well-being. Personnel Psychology, 63(2), 407-436.
Ilies, R., Schwind, K. M., Wagner, D. T., Johnson, M. D., DeRue, D. S., & Ilgen, D. (2007)
Can employees have a family life? The effects of daily workload and affect on workfamily conflict and social behaviors at home. Journal of Applied Psychology, 92(5),
339

1368-1379.
Innstrand, S. T., Langballe, E. M., Espnes, G. A., Falkum, E., & Aasland, O. G. (2008)
Positive and negative work—family interaction and burnout: A longitudinal study of
reciprocal relations. Work & Stress, 22(1), 1-15.
Janssen, O. (2001) Fairness perceptions as a moderator in the curvilinear relationships between
job demands, job performance and job satisfaction. Academy of Management Journal,
41(5), 1039-1050.
Jin, J. F., Ford, M. T., & Chen, C. C. (2013). Asymmetric Differences in Work–Family
Spillover in North America and China: Results from Two Heterogeneous
Samples. Journal of business ethics, 1-14.
Jones, Blake L., Scoville, D. Phillip, Hill, E. J., Childs, G., Leishman, J. M., & Nally, K. S.
(2008). Perceived versus used workplace flexibility in Singapore: Predicting workfamily fit. Journal of Family Psychology, 22(5), 774-774-783.
Judge, T. A., Ilies, R., & Scott, B. A. (2006). Work-family conflict and emotions: Effects at
work and at home. Personnel Psychology, 59(4), 779-779-814
Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D., & Rosenthal, R. A.
(1964). Organizational stress: Studies in role conflict and ambiguity. Oxford, England:
John Wiley.
Kahnweiler, William M., and Margaret A. Thompson. "Levels of Desired, Actual, and
Perceived Control of Employee Involvement in Decision Making: An Empirical
Investigation." Journal of Business and Psychology 14.3 (2000): 40727.ProQuest. Web. 14 Sep. 2013.
Karatepe, O. M. * Kilic, H. (2007) Relationships of supervisor support and conflict in the
work-family interface with the selected job outcomes of frontline employees. Tourism
Management, 28(1), 238-252.
Katerberg, R., Hom, Peter W., & Hulin, C. L. (1979). Effects of job complexity on the
reactions of part-time employees. Organizational Behavior & Human
Performance, 24(3), 317-332.
Kim, T., Hon, A. H. Y., & Lee, D. (2010). Proactive personality and employee creativity: The
effects of job creativity requirement and supervisor support for creativity. Creativity
Research Journal, 22(1), 37-45.
Kinnunen, U., & Mauno, S. (2008). Work-family conflict in individuals' lives: Prevalence,
antecedents, and outcomes. New York, NY, US: Cambridge University Press
Kitterød, R. H., & Lyngstad, T. H. (2005). Diary versus questionnaire information on time
340

spent on housework–The case of Norway. Electronic International Journal of Time Use
Research, 2(1), 13-32.
Knudsen, H. K., Ducharme, L. J., & Roman, P. M. (2009). Turnover intention and emotional
exhaustion "at the top": Adapting the job demands-resources model to leaders of
addiction treatment organizations. Journal of Occupational Health Psychology, 14(1),
84-95.
Konrad, A. M. & Mangel, R. (2000) The impact of work-life programs on firm productivity.
Strategic Management Journal, 21(12), 1225-1237.
Kopelman, R. W., Prottas, D. J., Thompson, C. A., and Jahn, E. W. (2006) A multilevel
examination or work-life practices: is more always better? Journal of Managerial Issues,
18(2), 232-253.
Kossek, E. E. & Nichol, V. (1992) The effects of on-site child care on employee attitudes and
performance. Personnel Psychology, 45(3), 485-509.
Kossek, E. E., & Lambert, S. J. (Eds.). (2012). Work and life integration: Organizational,
cultural, and individual perspectives. Psychology Press.
Kossek, E. E., Lewis, S. & Hammer., L. B., (2010) Work-life initiatives and organizational
change: overcoming mixed messages to move from the margin to the mainstream.
Human Relations, 63(3), 3-19.
Kossek, E., & Ozeki, C. (1998). Work–family conflict, policies, and the job–life satisfaction
relationship: A review and directions for organizational behavior–human resources
research. Journal of Applied Psychology,83(2), 139-149.
Kossek, E. E., & Ozeki, C. (1999). Bridging the work-family policy and productivity gap: A
literature review. Community, Work & Family, 2(1), 7-32.
Kozlowski, S. W. J. & Klein, K J. (2000). A multilevel approach to theory and research in
organizations. Contextual, temporal and emergent processes. In K. J. Klein and S. W. J.
Kozlowski (Eds.) Multilevel theory, research and methods in organizations:
Foundations, extensions and new directions (pp.2-90). San Francisco, CA: Jossey-Bass.
Kreiner, G. E. (2006). Consequences of work‐home segmentation or integration: a person‐
environment fit perspective. Journal of Organizational Behavior, 27(4), 485-507.
Kreiner, Glen E., Hollensbe, Elaine C., & Sheep, M. L. (2009). Balancing borders and bridges:
Negotiating the work-home interface via boundary work tactics. Academy of
Management Journal, 52(4), 704-730.
Kristof, A. L. (1996). Person‐organization fit: An integrative review of its conceptualizations,
341

measurement, and implications. Personnel psychology,49(1), 1-49.
Krueger, R. A. (1994). Focus groups: A practical guide for applied research. Thousand Oaks:
Sage Publications, Inc.
Krueger, R. A., & Casey, M. A. (2000). Focus groups. A practical guide for applied research
(3rd ed.). Thousand Oaks, CA: Sage Publications.
Lai, L., Rousseau, D. M., & Chang, K. T. T. (2009) Idiosyncratic deals: coworkers as
interested third parties. Journal of Applied Psychology, 94(2), 574-556.
Langfred, C. W. (2005). Autonomy and performance in teams: The multilevel moderating
effect of task interdependence. Journal of management, 31(4), 513-529.

Lapierre, L. M., Spector, P. E., Allen, T. D., Poelmans, S., Cooper, C. L., O'Driscoll, M. P.
(2008). Family-supportive organization perceptions, multiple dimensions of work-family conflict, and employee satisfaction: A test of model across five samples. Journal
of Vocational Behavior, 73(1), 92-106.
Lapierre, Laurent M., & Allen, T. D. (2006). Work-supportive family, family-supportive
supervision, use of organizational benefits, and problem-focused coping: Implications
for work-family conflict and employee well-being. Journal of Occupational Health
Psychology, 11(2), 169-181.
LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 questions about interrater reliability and
interrater agreement. Organizational Research Methods,11(4), 815-852.
Lee, G., Magnini, V. P., & Kim, B. C. (2011) Employee satisfaction with schedule flexibility:
Psychological antecedents and consequences within the workplace. International
Journal of Hospitality Management, 30(1), 22-30.
Lee, Y. S., & Waite, L. J. (2005). Husbands’ and wives’ time spent on housework: A
comparison of measures. Journal of Marriage and Family, 67(2), 328-336.
Leiter, M. P., Gascon, S., & Martinez-Jarreta, B. (2010) Making sense of work-life: a structural
model of burnout. Journal of Applied Social Psychology, 40(1), 57-75.
Lewin, J. E., & Sager, J. K. (2007). A process model of burnout among salespeople: Some new
thoughts. Journal of Business Research, 60(12), 1216-1224.
Lewis, S. (1997), ‘Family Friendly’ Employment Policies: A Route to Changing
Organizational Culture or Playing About at the Margins?. Gender, Work &
Organization, 4: 13–23.

342

Lingard, H., Brown, K., Bradley, L., Bailey, C., & Townsend, K. (2007). Improving
employees’ work-life balance in the construction industry: Project alliance case study.
Journal of Construction Engineering and Management, 133(10), 807-815.
Livingston, B. A., & Judge, T. A. (2008). Emotional responses to work-family conflict: An
examination of gender role orientation among working men and women. Journal of
Applied Psychology, 93(1), 207-207-216
Locke, K. (2002). The grounded theory approach to qualitative research. San Francisco, CA,
US: Jossey-Bass.
Lu, L., Kao, S., Chang, T., Wu, H., & Cooper, Cary L. (2008). Work/family demands, work
flexibility, work/family conflict, and their consequences at work: A national probability
sample in Taiwan. International Journal of Stress Management, 15(1), 1-1-21.
Marecek, J., & Ballou, D. J. (1981). Family roles and women's mental health. Professional
Psychology, 12(1), 39-46.
Martínez-Iñigo, D., Totterdell, P., Alcover, C. M., & Holman, D. (2007). Emotional labour and
emotional exhaustion: Interpersonal and intrapersonal mechanisms. Work &
Stress, 21(1), 30-30-47.
Maslach, C., & Jackson, S. E. (1984). Burnout in organizational settings. Applied Social
Psychology Annual, 5, 133-153.
Maslach, C., & Leiter, M. P. (2008). The truth about burnout: How organizations cause
personal stress and what to do about it. John Wiley & Sons.
Maslach, C., & Jackson, S. E. (1984). Burnout in organizational settings. Applied Social
Psychology Annual, 5, 133-153.
Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of
Psychology, 52, 397-422
Matthews, R. A., Barnes-Farrell, J. L., & Bulger, C. A. (2010). Advancing measurement of
work and family domain boundary characteristics. Journal of Vocational
Behavior, 77(3), 447-460.
Mathieu, J., Maynard, M. T., Rapp, T., & Gilson, L. (2008). Team effectiveness 1997-2007: A
review of recent advancements and a glimpse into the future. Journal of
Management, 34(3), 410-476.

343

Matthews, R., Allen, T., Barling, J., Eby, L., Greenhaus, J., Kossek, E., & Poelman, S. (2011)
Work-family research: the crossroads. Society for Industrial Organizational Psychology:
Proceedings from Annual Conference, Chicago, IL.
Matthews, R. A., Swody, C. A., & Barnes‐Farrell, J. L. (2012). Work Hours and Work–Family
Conflict: The Double‐edged Sword of Involvement in Work and Family. Stress and
Health, 28(3), 234-247.
McCarthy, A., Darcy, C. & Grady, G. (2010) Work-life balance policy and practice:
understanding line manager attitudes and behaviors. Human resource Management
Review, 20, 158-167.
McEvoy, G. M., & Cascio, W. F. (1987). Do good or poor performers leave? A meta-analysis
of the relationship between performance and turnover. Academy of Management
Journal, 30(4), 744-762.
McMillan, H. S., Morris, M. L., & Atchley, E. K. (2011). Constructs of the work/ life interface:
A synthesis of the literature and introduction of the concept of work/life
harmony. Human Resource Development Review, 10(1), 6-6-25.
Mescher, S., Benschop, Y., & Dorrewaard, H. (2009) Representations of work-life balance
support. Human Relationship, 63(1), 21-39.
Mesmer-Magnus, J., Murase, T., DeChurch, L. A., & Jimenez, M. (2010) Coworker informal
work accommodations to family: scale development and validation. Educational and
Psychological Measurement, 70(3), 511-531.
Michel, J. S., Mitchelson, J. K., Kotrba, L. M., LeBreton, J. M., & Baltes, B. B. (2009). A
comparative test of work-family conflict models and critical examination of work-family
linkages. Journal of Vocational Behavior, 74(2), 199-218.
Michel, J. S., Mitchelson, J. K., Pichler, S., & Cullen, K. L. (2010). Clarifying relationships
among work and family social support, stressors, and work–family conflict. Journal of
Vocational Behavior, 76(1), 91-104.
Mikkelsen, A., & Burke, R. J. (2004). Work-family concerns of Norwegian police officers:
Antecedents and consequences. International Journal of Stress Management, 11(4),
429-444.
Montgomery, A. J., Pangopolou, E. & Benos, A. (2006) Work-family interference as a
mediator between job demands and job burnout among doctors. Stress and Health,
22(3), 203-212.
Morgan, D. L. (1998). Practical strategies for combining qualitative and quantitative methods:
Applications to health research. Qualitative Health Research, 8(3), 362-376
344

Morganson, V. J., Major, D. A., Oborn, K. L., Verive, J. M., & Heelan, M. P. (2010).
Comparing telework locations and traditional work arrangements: Differences in worklife balance support, job satisfaction, and inclusion. Journal of Managerial
Psychology, 25(6), 578-595.
Morse, Janice M. (1994). Designing funded qualitative research. In Norman K. Denzin &
Yvonna S. Lincoln (Eds.), Handbook of qualitative research (2nd ed., pp.220-35).
Thousand Oaks, CA: Sage.
Netemeyer, R. G., Boles, J. S., & McMurrian, R. (1996). Development and validation of work–
family conflict and family–work conflict scales. Journal of Applied Psychology, 81(4),
400-410.
Nippert-Eng, C. (1996). Calendars and keys: The classification of "home" and "work".
Sociological Forum, 11(3), 563-582
Nyberg, A. (2010) Retaining your high performers: moderators of the performancesatisfaction-voluntary turnover relationship. Journal of Applied Psychology, 95(3), 440453.
O'Driscoll, M. P., Poelmans, S., Spector, P. E., Kalliath, T., Allen, T. D., Cooper, C. L., &
Sanchez, J. I. (2003). Family-responsive interventions, perceived organizational and
supervisor support, work-family conflict, and psychological strain. International Journal
of Stress Management, 10(4), 326.
O'Driscoll, Michael P., Ilgen, D. R., & Hildreth, K. (1992). Time devoted to job and off-job
activities, interrole conflict, and affective experiences. Journal of Applied Psychology,
77(3), 272-279.
O’Neill, J. W., Harrison, M. M., Cleveland, J., Almeida, D., Stawski, R., & Crouter, A. C.
(2009). Work-family climate, organizational commitment, and turnover: Multievel
contagion effects of leaders. Journal of Vocational Behavior, 74(1), 18-29.
Olson-Buchanan, J. B. & Boswell, W. R. (2006) Blurring boundaries: correlates of integration
and segmentation between work and nonwork. Journal of Vocational Behavior, 68, 432445.
Paine, W. S. (1984). Professional burnout: Some major costs. Family & Community Health:
The Journal of Health Promotion & Maintenance, 6(4), 1-11.
Parker, P. A., & Kulik, J. A. (1995). Burnout, self- and supervisor-related job performance, and
absenteeism among nurses. Journal of Behavioral Medicine, 18(6), 581-599
Peeters, M. C. W., de Jonge, J., Janssen, P. P. M., & van, d. L. (2004). Work-home
345

interference, job stressors, and employee health in a longitudinal
perspective. International Journal of Stress Management, 11(4), 305-322.
Peeters, M. C., Montgomery, A. J., Bakker, A. B., & Schaufeli, W. B. (2005) Balancing work
and home: how job and home demands are related to burnout. International Journal of
Stress Management, 12(1), 43-61.
Pisarksi, A., Lawrence, S. A., Bohle, P. & Brook, C. (2008) Organizational influences on the
work-life conflict and health of shiftworkers. Applied Ergonomics, 39(5), 580-588.
Pocock, B. (2005). Work/Care regimes: Institutions, culture and behaviour and the australian
case. Gender, Work and Organization, 12(1), 32-49.
Podsakoff, Nathan P., LePine, Jeffery A., & LePine, Marcie A. (2007). Differential challenge
stressor-hindrance stressor relationships with job attitudes, turnover intentions, turnover,
and withdrawal behavior: A meta-analysis. Journal of Applied Psychology, 92(2), 438454.
Porath, C. L., & Bateman, T. S. (2006). Self-regulation: from goal orientation to job
performance. Journal of Applied Psychology, 91(1), 185.
Powell, G. N., & Greenhaus, J. H. (2006). Managing incidents of work-family conflict: A
decision-making perspective. Human Relations, 59(9), 1179-1212.
Ransford, C. R., Crouter, A. C., & McHale, S. M. (2008). Implications of work pressure and
supervisor support for fathers', mothers' and adolescents' relationships and well-being in
dual-earner families.Community, Work & Family, 11(1), 37-60.
Rasmussen, T H. & Jeppesen, H. J. (2006) Teamwork and associated psychological factors: a
review. Work & Stress, 20(2), 105-128.
Raudenbush, S. W. & Bryk, A. S. (2002). Applications and Data Analysis Methods, Second
Edition, Thousand Oaks,CA: Sage.
Richman, A. L., Civian, J. T., Shannon, L. L., Hill, E. J., & Brennan, R. T. (2008). The
relationship of perceived flexibility, supportive work-life policies, and use of formal
flexible arrangements and occasional flexibility to employee engagement and expected
retention. Community, Work & Family, 11(2), 183-197.
Rode, J. C., Rehg, M. T., Near, J. P. & Underhill, J. R. (2007) The effect of work/family
conflict on intention to quit: the mediating roles of job and life satisfaction. Applied
Research in Quality of Life, 2(2), 65-82.
Rosin, H. & Korabik, K. (1995) Organizational experiences and propensity to leave: a
multivariate investigation of men and women managers. Journal of Vocational
346

Behavior, 46, 1-16.
Rothbard, N. P., Phillips, K. W., & Dumas, T. L. (2005) Managing multiple roles: Workfamily policies and individuals’ desires for segmentation. Organizational Science, 16(3),
243-258.
Rousseau, D. M. (2001). The idiosyncratic deal: flexibility vs. fairness. Organizational
Dynamics, 29(4), 260-273.
Rousseau, D. M. (2005). I-deals: Idiosyncratic deals employees bargain for themselves. ME
Sharpe.
Rupprecht, E. & Grawitch, M. J. (2010) Managers and subordinate stress. Good Company
Newsletter, 4(9), retrieved from
http://www.phwa.org/resources/goodcompany/newsletter/article/226 June 10, 2011.
Rychlak, J. F. (1981). Logical learning theory: Propositions, corollaries, and research
evidence. Journal of Personality and Social Psychology, 40(4), 731-749.
Sackett, Paul R., & Larson, James R. (1990).Research strategies and tactics in industrial and
organizational psychology. Palo Alto, CA, US: Consulting Psychologists Press.
Sanz-Vergel, A., Demerouti, E., Moreno-Jiménez, B., & Mayo, M. (2010). Work-family
balance and energy: A day-level study on recovery conditions. Journal of Vocational
Behavior, 76(1), 118-130.
Schaffer, J. A. & Postlewaite, B. E. (2012) A matter of context: A meta-analytic investigation
of the relative validity of contexitualized and noncontexualized personality measures.
Personnel Psychology, 65, 445-494.
Schaufeli, W. B. & Bakker, A. B. (2004) Job demands, job resources, and their relationship
with burnout and engagement: a multi-sample study. Journal of Organizational
Behavior, 25(3), 293-315.
Schaufeli, W. B., Bakker, A. B., Van der Heijden., & Prins, J. T. (2009). Workaholism,
burnout and well-being among junior doctors: The mediating role of role conflict. Work
& Stress, 23(2), 155-172
Scott, W. E. (1966). Activation theory and task design. Organizational Behavior & Human
Performance, 1(1), 3-30.
Shirom, A. (1989). In Cooper C. L., Robertson I. T.(Eds.), Burnout in work organizations.
Oxford, England: John Wiley & Sons.
Shirom, A., & Melamed, S. (2005). In Antoniou A. G., Cooper C. L.(Eds.), Does burnout
347

affect physical health? A review of the evidence. Northampton, MA, US: Edward Elgar
Publishing.
Shirom, A., Nirel. N., & Vinokur, A. D. (2009) Work hours and caseload as predictors of
physician burnout: the mediating effects of perceived workload and by autonomy.
Applied Psychology, 59(4), 539-565.
Seidman, I. (2012) Interviewing as qualitative research A guide for researching in education
and the social sciences. Teachers college press.
Seibert, S. E., Crant, J. M., & Kraimer, M. L. (1999). Proactive personality and career success.
Journal of Applied Psychology,84(3), 416-427.
Singh, P., Suar, D., & Leiter, M. P. (2011) Antecedents, work-related consequences, and
buffers of job burnout among Indian software developers. Journal of Leadership and
Organizational Studies, 19(1). 83-104.
Singleton, R. A., & Straits, B. C. (1999). Approaches to social research (3rd ed.). New York,
NY, US: Oxford University Press.
Small, S. A., & Riley, D. (1990). Toward a multidimensional assessment of work spillover into
family life. Journal of Marriage & the Family, 52(1), 51-61
Sonnentag, S., & Bayer, U. (2005). Switching off mentally: Predictors and consequences of
psychological detachment from work during off-job time. Journal of Occupational
Health Psychology, 10(4), 393-414.
Sonnentag, S., & Fritz, C. (2007). The Recovery Experience Questionnaire: development and
validation of a measure for assessing recuperation and unwinding from work. Journal of
occupational health psychology, 12(3), 204.
Sonnentag, S., & Kruel, U. (2006). Psychological detachment from work during off-job time:
The role of job stressors, job involvement, and recovery-related self-efficacy. European
Journal of Work and Organizational Psychology, 15(2), 197-217.
Sonnentag, S., & Zijlstra, F. R. H. (2006). Job characteristics and off-job activities as
predictors of need for recovery, well-being, and fatigue. Journal of Applied
Psychology, 91(2), 330-350.
Spector, P. E., Allen, T. D., Poelmans, S., LaPierre, L. M., Cooper, C. L., et al (2007) Crossnational differences in relationships of work demands, job satisfaction, and turnover
intentions with work-family conflict. Personnel Psychology, 60(4), 805-835.
Spreitzer, G. M. (1995). Psychological empowerment in the workplace: Dimensions,
measurement, and validation. Academy of management Journal,38(5), 1442-1465.
348

Sprung, J. M., & Jex, S. M. (2012). Work locus of control as a moderator of the relationship
between work stressors and counterproductive work behavior. International Journal of
Stress Management, 19(4), 272-291.
Standen, P., Daniels, K., & Lamond, D. (1999). The home as a workplace: Work–family
interaction and psychological well-being in telework. Journal of Occupational Health
Psychology, 4(4), 368-381.
Stavrou, E., & Kilaniotis, C. (2010). Flexible work and turnover: An empirical investigation
across cultures. British Journal of Management, 21(2), 541-554.
Stewart, G. L. (2006). A meta-analytic review of relationships between team design features
and team performance. Journal of Management, 32(1), 29-55.
Strine, T. W., & Chapman, D. P. (2005). Associations of frequent sleep insufficiency with
health-related quality of life and health behaviors. Sleep medicine, 6(1), 23-27.
Tekleab, Amanuel G., Quigley, Narda R., & Tesluk, P. E. (2009). A longitudinal study of team
conflict, conflict management, cohesion, and team effectiveness. Group & Organization
Management, 34(2), 170-205
ten Brummelhuis, L. L. & van der Lippe, J. (2010) Effective work-life balance support for
various household structures. Human Resource Management, 49 (2), 173-193.
Thomas, L. T., & Ganster, D. C. (1995). Impact of family-supportive work variables on workfamily conflict and strain: A control perspective. Journal of Applied Psychology, 80(1),
6-15
Thompson, Briony M., Brough, Paula A., & Schmidt, H. (2006). Supervisor and subordinate
work-family values: Does similarity make a difference? International Journal of Stress
Management, 13(1), 45-63.
Thompson, C. A., Beauvais, L. L., & Lyness, K. S. (1999). When work–family benefits are not
enough: The influence of work–family culture on benefit utilization, organizational
attachment, and work–family conflict. Journal of Vocational Behavior, 54(3), 392-415
Thompson, C. A., & Prottas, D. J. (2006). Relationships among organizational family support,
job autonomy, perceived control, and employee well-being.Journal of occupational
health psychology, 11(1), 100.
Trevor, Charlie O., Gerhart, B., & Boudreau, J. W. (1997). Voluntary turnover and job
performance: Curvilinearity and the moderating influences of salary growth and
promotions. Journal of Applied Psychology, 82(1), 44-61
Um, M., & Harrison, D. F. (1998). Role stressors, burnout, mediators, and job satisfaction: A
349

stress–strain–outcome model and an empirical test. Social Work Research, 22(2), 100115.
van Daalen, G., Willemsen, T. M., & Sanders, K. (2006). Reducing work-family conflict
through different sources of social support. Journal of Vocational Behavior, 69(3), 462476.
van der Doef, H. & Maes, S. (1999). The job demand-control (-support) model and
psychological well-being: A review of 20 years of empirical research. Work &
Stress, 13(2), 87-114.
van Steenbergen, E. F., Ellemers, N., & Mooijaart, A. (2007). How work and family can
facilitate each other: Distinct types of work-family facilitation and outcomes for women
and men. Journal of Occupational Health Psychology, 12(3), 279-300.
van Yperen, N. W. & Hagedoorn, M. (2003) Do high job demands increase intrinsic
motivation or fatigue or both? The role of job control and job social support. The
Academy of Management Journal, 46(3), 339-348.
Vallerand, R. J., Paquet, Y., Philippe, F. L., & Charest, J. (2010). On the role of passion for
work in burnout: A process model. Journal of Personality, 78(1), 289-312.
Verquer, M. L., Beehr, T. A., & Wagner, S. H. (2003). A meta-analysis of relations between
person-organization fit and work attitudes. Journal of Vocational Behavior, 63(3), 473489.
Voydanoff, P. (2005). Consequences of boundary-spanning demands and resources for workto-family conflict and perceived stress. Journal of Occupational Health
Psychology, 10(4), 491-503.
Wanous, John P., & Hudy, Michael J. (2001). Single-item reliability: A replication and
extension. Organizational Research Methods, 4(4), 361-375
Wayne, J. H., Randel, A. E., & Stevens, J. (2006). The role of identity and work-family
support in work-family enrichment and its work-related consequences. Journal of
Vocational Behavior, 69(3), 445-461.
Winkel, D. E., & Clayton, R. W. (2010). Transitioning between work and family roles as a
function of boundary flexibility and role salience. Journal of Vocational
Behavior, 76(2), 336-343.
Witt, L. A., & Carlson, D. S. (2006). The work-family interface and job performance:
Moderating effects of conscientiousness and perceived organizational support. Journal
of Occupational Health Psychology,11(4), 343-357.
350

Wittmer, Jenell L. S., & Martin, James E. (2010). Emotional exhaustion among employees
without social or client contact: The key role of nonstandard work schedules. Journal of
Business and Psychology, 25(4), 607-623.
Xanthopoulou, D., Bakker, A. B., Dollard, M. F., Demerouti, E., Schaufeli, W. B., Taris, T.
W., & Schreurs, P. J. G. (2007). When do job demands particularly predict burnout?
Journal of Managerial Psychology, 22(8), 766-786.
Zabkiewicz, D. (2010). The mental health benefits of work: Do they apply to poor single
mothers? Social Psychiatry and Psychiatric Epidemiology, 45(1), 77-87.

Zijlstra, F. R., & Sonnentag, S. (2006). After work is done: Psychological perspectives on
recovery from work. European Journal of Work and Organizational Psychology, 15(2),
129-138.
Zimmerman, R. D. & Darnold, T. C. (2009) The impact of job performance on employee
turnover intentions and the voluntary turnover process: A meta-analysis and path model.
Personnel Review, 38(2), 142-158.

351