PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KzlProleccaPresICIRCIDateDue.indd

 

 

NEGATIVE AFFECTIVE RESPONSES TO WORK-FAMILY CONFLICT
By

Megan Leigh Huth

A THESIS

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS
Psychology

2009

ABSTRACT
NEGATIVE AFFECTIVE RESPONSES TO WORK-FAMILY CONFLICT
By

Megan Leigh Huth

The prevalent literature on work-family conﬂict suggests that working adult, and
women in particular, experience guilt when work and family roles conﬂict with one
another. Although guilt is oﬁen mentioned in the literature, it is rarely directly assessed.
In addition, other negative emotions besides guilt may play a crucial role in working
adult emotional responses to work-family conﬂict. Drawing on SDT Theory (Higgins,
1987), this thesis sought to more fully understand the role a variety of negative emotions
(guilt, worry, anger and frustration) play in work to family and family to work conﬂict.
Findings suggest that all four emotions are directly related to experiences of work-family
conﬂict and that women are more prone to experiencing all these emotions. Links are
also made to life satisfaction and job satisfaction. Future directions of the study of

emotions and work-family conﬂict are proposed.

Copyﬁghtby
MEGAN LEIGH HUTH

2009

ACKNOWLEDGEMENTS

This thesis was undertaken with much guidance and support from my thesis committee
chair, Dr. Ann Marie Ryan. Her inputs greatly improved both the design and analysis
portions Of this thesis. My other committee members, Dr. J. Kevin Ford and Dr. Rick
DeShon, provided invaluable insight into the proposal and design process. I would like

to thank them all for their valuable support and input on this thesis.

iv

TABLE OF CONTENTS

LIST OF TABLES ______________________________________________________________________________________________ vii

LIST OF FIGURES ____________________________________________________________________________________________ 1x

INTRODUCTION ______________________________________________________________________________________________ 1

CHAPTER 1

LITERATURE REVIEW ___________________________________________________________________________________ 3
Work-family conﬂict ______________________________________________________________________________ 3
Outcomes of WF C and Negative Mood ________________________________________________ 8
Self-Discrepancy Theory _______________________________________________________________________ 10
Guilt _________________________________________________________________________________________________________ 15
Worry (Anxiety) _____________________________________________________________________________________ 21
Anger _______________________________________________________________________________________________________ 2 4
Frustration ............................................................................................... 2 8
Work and Family Importance _______________________________________________________________ 32
Gender Differences ................................................................................ 34
Job and Life Satisfaction ________________________________________________________________________ 38

CHAPTER 2

METHODS __________________________________________________________________________________________________________ 4 2
SamPle ..................................................................................................... 4 2
Participants _____________________________________________________________________________________________ 4 3
Measures _________________________________________________________________________________________________ 44

CHAPTER 3

RESULTS ____________________________________________________________________________________________________________ 4 9
Sample Description ________________________________________________________________________________ 4 9
Measure Development: The WFC Emotions Scale _____________________________ 4 9
Modiﬁed Hypotheses ____________________________________________________________________________ 60
Descriptive Statistics ________________ .. _____________________________________________________________ 68
Hypothesis Testing _________________________________________________________________________________ 75

CHAPTER 4

DISCUSSION _____________________________________________________________________________________________________ 96
Contributions 96

Limitations and Future directions

Research and Practical Implications _____________________________________________________

APPINDICES _____________________________________________________________________________________________________
Appendix A: Online Consent Form ______________________________________________________
Appendix B: Study Advertisement _______________________________________________________
Appendix C: Second Wave Recruitment Email ___________________________________
Appendix D: Third Wave Recruitment Email ______________________________________
Appendix E: Work-Family Interference Negative Affect Scale _________
Appendix F: Work-Family Conﬂict Scale ............................................
Appendix G: PANAS Scale ___________________________________________________________________
Appendix H: Work Importance _____________________________________________________________
Appendix I: Family Importance ____________________________________________________________
Appendix J: Job Satisfaction .................................................................
Appendix K: Life Satisfaction _______________________________________________________________
Appendix L: Demographics ___________________________________________________________________

REFERENCES

...................................................................................................

vi

107
111

115
115
117
118
119
121
127
129
131
132
133
134
135

139

TABLE 1
TABLE 2
TABLE 3

TABLE 4

TABLE 5

TABLE 6

TABLE 7

TABLE 8
TABLE 9

TABLE 10

TABLE 1 1

TABLE 12

TABLE 13

TABLE 14

TABLE 15

LIST OF TABLES

The Six Types Of Work-Family Conﬂict ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

Deﬁnitions of Guilt

Descriptive Information for Demographic Variables _____________________

Varimax Rotated Maximum Likelihood Factor Structure

for All 49 WFC Negative Affect Items ___________________________________________

Varimax Rotated Maximum Likelihood Factor Structure

for 24 Other-References WF C Negative Affect Items ...................

Confirmatory Factor Analyses _________________________________________________________

Factor Structure for 24 Self-Reference WFC Negative
Affect Items

Original and Modiﬁed Hypotheses __________________________________________________

Means, Standard Deviations, and Intercorrelations ________________________

Time-Based WIF and F [W Conﬂict Regressed on WF C

Time-Based Extemalizing and Intemalizing Emotions ,,,,,,,,,,,,,,,,,

Strain-Based WIF Conﬂict Regressed on Strain-Based WIF
Emotions with Interactions of Strain Based WIF Conﬂict

and Family Importance _____________________________________________________________________

Strain-Based FIW Conﬂict Regressed on Strain-Based F IW
Emotions with Interactions of Strain Based F [W Conﬂict

and Job Importance ...........................................................................

Behavior-Based WIF Conﬂict Regressed on Behavior-Based
WIF Emotions with Interactions of Behavior Based WIF

Conﬂict and Family Importance ......................................................

Behavior-Based F IW Conﬂict Regressed on Behavior-Based
FIW Emotions with Interactions of Behavior-Based FIW

Conﬂict and Job Importance ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

Time-based FIW Conﬂict Regressed on Time Based WFC
Extemalizing and Intemalizing Emotions with Interactions

vii

_____ 5

18

42

51

53

55

59

61

69

77

77

78

78

8O

TABLE 16

TABLE 17

TABLE 18

TABLE 19

TABLE 20

TABLE 21

TABLE 22

TABLE 23

TABLE 24

TABLE 25

TABLE 26

of Time-Based FIW Conﬂict and Job Importance ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

Time-Based WIF Conﬂict Regressed on Time-Based
WFC Extemalizing Emotions with Interactions of
Time-Based WIF Conﬂict and Family Importance ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

Time-Based WIF Conﬂict Regressed on Time-Based WF C
Intemalizing Emotions with Interactions of Time-Based WIF
Conﬂict and Family Importance ___________________________________________________________

Time-Based FIW Conﬂict Regressed on Time-Based WFC
Intemalizing Emotions with Interactions of Time-Based F IW
Conﬂict and Gender _______________________________________________________________________________
Time-Based WIF Conﬂict Regressed on Time-Based WFC
Intemalizing Emotions with Interactions of Time-Based WIF
Conﬂict and Gender _______________________________________________________________________________
Time-Based WIF Conﬂict Regressed on Time-Based WFC
Intemalizing Emotions with Interactions of Time-Based WIF
Conﬂict and Gender _______________________________________________________________________________
Time-Based WFC Extemalizing, Time-Based WFC

Intemalizing, Strain-Based WIF and Behavior Based WIF
Emotions Regressed on Job Satisfaction ______________________________________________

Time-Based WF C Extemalizing, Time-Based WFC

Intemalizing, Strain-Based WIF, Strain-Based FIW,
Behavior-Based WIF and Behavior-Based FIW Emotions
Regressed on Job Satisfaction _______________________________________________________________

Summary of New Hypotheses and Whether They Were
Supported Through Analysis _________________________________________________________________

Varimax Rotated Factor Structure for Other Referenced
Items, With Factor Number Constrained to Three ______________________________

Time-Based WIF Conﬂict, Time-Based FIW Conﬂict,
Strain-Based WIF Conﬂict, Strain-Based F IW Conﬂict,
Behavior-Based WIF Conﬂict and Behavior-Based FIW

Conﬂict Regressed on Other WFC Intemalizing Emotion,

Other FIW Extemalizing Emotions and Other WIF

Extemalizing Emotions _________________________________________________________________________

Other Referenced WFC Emotion Scales Regressed on
Job Satisfaction and Life Satisfaction __________________________________________________

viii

8O

82

82

84

84

86

86

88

9O

93

94

95

FIGURE 1

FIGURE 2

FIGURE 3

FIGURE 4

FIGURE 5

LIST OF FIGURES

Proposed Model of the relationship Between WFC
and Emotions, with Work Importance, Family Importance
and Gender as Moderators 16

Proposed Moderation Relationship of Job and Family Importance___33

Scree Plot of Rotated Factor Matrix for the 24 Self-Referenced

Emotion Items ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 57
Proposed Model of Modiﬁed Hypotheses ____________________________________________ 65
Scree plot for other referenced emotions _____________________________________________ 88

ix

“We 're moms. We work all day, bring home the bacon, and ﬂy it up in a pan. Oh, and while
we 're at it, we 're raising young children, along with our spouses/partners. As any working
mom knows, we often battle the big "G. ” Guilt creeps up on us when we least expect it. Join
us in our ongoing struggle to resist the guilt and embrace the journey. ”

—Working Moms Against Guilt weblog motto (2008)

Parental guilt associated with work-family conﬂict has become an assumed part of
our popular lexicon. As the above quote from the popular weblog “Working Moms Against
Guilt” illustrates, the idea that parents, and moms in particular, experience guilt about
balancing work and family is stated as so obvious as to be unambiguous. Yet surprisingly
little empirical research has been dedicated to examining the ways adults emotionally
experience work-family conﬂict. First, although guilt is often assumed to results from
experiences of conﬂict, how prevalent is it really? If guilt does indeed occur, is it the only
relevant emotion when studying conﬂict? Are women more susceptible to negative emotional
effects of work-family conﬂict than men? Are there other variables that might contribute to
individual emotional experiences of work-family conﬂict? All these questions have yet to be
adequately answered.

The emotional component of work-family conﬂict is a glaring hole in the current
literature. More fully understanding how work-family conﬂict links to emotions could prove
invaluable in further research and the development of practical interventions. While research
has revealed much about the antecedents and consequences of work-family conﬂict (Eby,
Casper, Lockwood, Bordeaux, & Brinley, 2005), it has been less successful in identifying
workplace or personal methods and tactics that help alleviate the negative effects. The study
of the emotional reactions of individuals to work-family conﬂict could identify a targeted
intervention point that researchers and practitioners could focus on to help mitigate the

consequences of work-family conﬂict. The coping literature suggests that different ways of

coping may be more effective to counteract negative effects of speciﬁc emotions (Bennett,
Lowe & Honey, 2003); therefore, it behooves us as researchers to identify what emotions,
precisely, individuals experience in conjunction with work-family conﬂict.

Although several authors in the ﬁeld assume an emotional component to work-family
conﬂict (e.g. Casper, Martin, Buffardi & Erdwins, 2002; Wharton & Erickson, 1993;
Williams & Alliger, 1994), few have set out to systematically understand what emotions play
a part in this conﬂict. Much of the extant research has linked work-family conﬂict to
personality-like trait levels of emotions, such as negative affectivity, but does not take the
next step to identify speciﬁc emotional responses (e.g. Heller & Watson, 2005; Song, Foo, &
Uy, 2008). In their model describing health outcomes of work-family conﬂict, Greenhaus and
Spector (2006) noted the lack of research on this topic and pointed out that emotions are
likely more proximally related to experiences of conﬂict than health outcomes. They suggest
that the experience of emotions may have very real effects on later development of negative
health outcomes. Along with other researchers in the ﬁeld (Judge, Ilies & Scott, 2006;
Livingston & Judge, 2008), Greenhaus and Spector (2006) call for further research linking
Speciﬁc state emotions to work-family conﬂict.

A few researchers have started on the path to understanding the discrete emotional
responses to work-family conﬂict (Hochwarter, Perrewe, Meurs, & Kacmar, 2007; Judge,
Ilies & Scott, 2006; Livingston & Judge, 2008) but they treat what emotions they examine
rather idiosyncratically. In general, they focus on one speciﬁc emotion (guilt) but fail to
consider other types of negative emotions that may be related to experiences of conﬂict.
Before further research is conducted on emotional responses to work-family conﬂict, it is
necessary to fully explicate what sorts of emotions are likely to be experienced so that

researchers can fully capture the range of negative emotional reactions.

Using Higgins (1987) self-discrepancy theory as a framework for understanding
emotional responses to work-family conﬂict, I conducted study that examines the prevalence
and relationship strength of four speciﬁc emotions in response to work-family conﬂict—
namely guilt, worry, anger and frustration. The results of this study can have important
implications for further research into emotions and work-family conﬂict. At a basic level,
this study will help illuminate negative emotions other than guilt that might be related to
work-family conﬂict, namely worry, anger and frustration. Different emotions may have
different relationships to other outcome variables such as satisfaction and health outcomes,
which could lead to more effective interventions on ways to mitigate the negative effects of
work family conﬂict.

In order to understand these relationships, it is ﬁrst necessary to fully deﬁne the
constructs of interest. To do this, I will review the literature on work-family conﬂict,
introduce self-discrepancy theory as a framework to understand emotional reactions to
conﬂict, and explore deﬁnitions of each emotion in turn. Following that I will address the
proposed moderators of the work-family conﬂict/emotion relationship
Work-Family Conﬂict

Social scientists ﬁrst began turning an eye to work-family conﬂict about forty years
ago. Kahn, Wolfe, Quinn, Snoek, and Rosenthal (1964) introduced work-family conﬂict as a
function of role identity theory where the demands of one role are fundamentally
incompatible with the demands of another role. This in turn produces stress when the
demands of one role cannot be met. Kahn et al (1964) introduced this idea in the context of
the many roles an individual assumes, including multiple roles at work and multiple roles at
home. While they focused mainly on roles in the work domain, it is easy to extrapolate the
idea that any role an individual assumes might at some point come into direct conﬂict with

another role.

For the next twenty years or so, the area of work-family role conﬂict research presented
by Kahn et al (1964) was largely left unexplored. The late 1960’s and 1970’s saw a few
articles about this role conﬂict, but they were mainly exploratory in nature and did not focus
on speciﬁc theories or constructs (e.g. Matthews & Tiedeman, 1964) or only referenced
work-family conﬂict indirectly (e.g. Kessler, 1979).

The late 19703 and 19805 saw an increasingly dramatic demographic shift in the
nation’s workforce. More and more women were entering the workforce and more and more
marriages included dual—earner couples where both adults worked (Bond, Thompson,
Galinsky & Prottas 2003). This increased participation in the workforce served to increase
the work demands on a system where the family demands remained and needed to be met.
Americans began to have more demands on their time and the roles of work and family life
began to conﬂict with each other more and more.

The strain this conﬂict placed on working individuals and families was becoming
increasingly obvious in the 19805. This is when the work-family research really took ﬂight.
What had been a meager trickle of empirical articles grew to a stream, and then a raging
river. That is where we stand today. A simple search on psycinfo for articles related to
work-family conﬂict and balance returns roughly 700 articles, most of which have been
published in the last ten to ﬁfteen years. The next step in understanding work-family conﬂict
is to look at the conceptual foundation much of this research relies upon.

Recognizing the lack of a strong theoretical base for an increasingly relevant concept,
Greenhaus and Beutell (1985) wrote an article that outlined the basic theoretical structure of
work-family conﬂict. This article has become the seminal reference to understanding the
nature of work-family conﬂict. The basic idea of Greenhaus and Beutell’s conceptualization
of work-family conﬂict is that it has two directions, work-to-family and family-to-work, and

three types, time-based, strain-based, and behavior-based, and that they are fully crossed.

Please see Table 1 for a concise description of the directions and types of WFC.

First, work-family conﬂict is bi-dimensional. Experience in the work realm can impact

experiences in the family realm (work interference with family, hence referred to WIF) and

experiences in the family realm can impact experiences in the work realm (family

interference with work, hence referred to as F 1W). This distinction has been held up repeated

in the literature, indicating that individuals consistently differentiate between these two

directions (Byron, 2005).

Work-family conﬂict also consists of three types. Time-based conﬂict is perhaps the

easiest to understand and conceptualize. Time spent in one role can physically not be spent

in another role. For example, an individual would experience time-based conﬂict if he/she

needed to attend a child’s soccer game that was scheduled at the same time as an

Table l
The Six Types of Work-Family Conﬂict

 

WFC Direction

 

WF C Type

Work-Interfering-With-
Family
(WI F)

Family-Interfering-With-
Work
(FIW)

 

Time-Based

Strain-Based

Behavior-Based

WIF-time

Responsibilities from work
role temporally conﬂict with
responsibilities from family
role

WIF-strain

Stress and strain from work
role makes it difﬁcult to
adequately perform in family
role

WIF-behavior

Behaviors that are adaptive
in the work role are
maladaptive in the family
role

F IW-time

Responsibilities from family
role temporally conﬂict with
responsibilities from work
role

FIW-strain

Stress and strain from family
role makes it difﬁcult to
adequately perform in work
role

F IW-behavior

Behaviors that are adaptive in
the family role are
maladaptive in the work role

important work meeting. Strain-based conﬂict is when emotional factors such as stress and
strain experienced in one role can impact experiences in another role. For example, an
individual would experience strain-based conﬂict if pressures from work deadlines were so
emotionally draining that he/she is too tired or worn out to emotionally contribute to their
family at the end of the day. Behavior-based conﬂict is perhaps the most difﬁcult to
conceptualize and occurs when behaviors that are adaptive and appropriate in one role are
incompatible with the other. For example, an individual would experience behavior-based
conﬂict if his/her work required him/her to be inﬂexible and strict with coworkers but this
same behavior would be detrimental if enacted within family relationships.

The dimensions (FIW and WIF) and types of conﬂict are fully crossed, meaning that all
three types of conﬂict can occur in either direction of conﬂict. In fact, Greenhaus and Beutell
would argue that a single experience could have the potential to go in either direction. For
example, an individual has competing role responsibilities such that his work role dictates
that he needs to stay late to ﬁnish a project but his family role dictates that he needs to leave
early to watch his child’s soccer game. At the point of indecision, this time-based conﬂict
has the potential to go either way. Once the individual decides on a course of action, it can
either be family interfering with work (if he decides to attend his child’s soccer game) or
work interfering with family (if he decides to stay late at work).

The self-deﬁned nature of directionality in WFC is fairly straightforward in terms of
time-based WFC where a clear decision must be made regarding which event the individual
will actually attend. Strain-based and behavior-based conﬂict ﬁt less neatly into this
framework, although both have aspects of directionality self-deﬁnition. For strain-based
conﬂict, people have to recognize that they are worn out from activities in certain domains,
and this recognition and attribution could function differently between people or times. For

example, if an individual is at home and feeling stressed, that individual can attribute the

stress to either something that happened at work or at home. While the individual has some
control over interpreting the origination of the strain, the basic direction remains factually
ﬁxed, such that it is the stress experienced from one domain effects performance in the other.
Behavior-based conﬂict is even less likely to have a strong self-deﬁned directionality
aspect to it. Like strain-based conﬂict, the ambiguity lies in the recognition and
understanding of what behaviors stem from what domain. For example, if an individual
behaves in a cajoling way with a co-worker that is ineffective in that situation but is effective
with her children, then she has to recognize where that behavior originates before she can
ascribe conﬂict directionality. Namely, she has to correctly ascribe the fact that it is a family-
based behavior that is conﬂicting with work-based role. The recognition and labeling of the
conﬂict is the self-deﬁned aspect of directionality. But like strain-based conﬂict, behavior
based conﬂict does have a ﬁxed element to it, such that the above mentioned conﬂict is
clearly in the direction of family-to-work and the ambiguity only lies in the attribution.
Greenhaus and Beutell’s (1985) general framework of work-family conﬂict has been
well accepted by the research community and most current research is based, at least to some
extent, on this conceptualization of conﬂict. In practice, much of the current research tends
to look at direction (F IW or WIF) but treats the dimensions in idiosyncratic ways, often not
measuring them at all (e.g. Adams, King & King, 1996; Batt & Valcour, 2003; Burley, 1995).
When all three types are measured, however, they all often have unique relationships with
outcomes and antecedents (e.g. Carlson, Derr & Wadsworth, 2003; Lambert, Hogan, Camp &
Ventura, 2006). Reviews of the literature have reiterated the importance of measuring all
directions and dimensions of conﬂict in order to fully capture the construct (Eby et al 2005;
Ford, Heinen & Langkamer, 2007). Due to these types of ﬁndings, this paper will adopt the
full model of Greenhaus and Beutell’s two by three fully crossed construct. The model of

WFC used in this paper will examine the two directions (WIF and F IW) and three

dimensions (strain, time and behavior).
Outcomes of WF C and Negative Mood

In the literature, experiences of WFC have been linked to a variety of negatively
valenced outcomes. The most widely researched correlates of WFC often have to do with
satisfaction, such that individuals who experience higher levels of WFC also experience
lower satisfaction with both job and life (Eby et al, 2004; Ford et al, 2007). The satisfaction
relationships generally occur in predicted ways, such that individuals who experience more
work-interfering—with—family experience lower job satisfaction and those that experience
more family-interfering—with-work experience lower family satisfaction (Eby et al, 2004). A
meta analysis by Byron (2005) also supported this distinction, ﬁnding that by and large work
related issues such as job stress were more closely related to WIF while family related issues
such as family stress were more closely related to FIW. A meta-analysis by Ford et al
(2007), however, illustrates the WFC-satisfaction relationship to be more complicated. They
found that there are considerable cross-domain inﬂuences, such that WIF is also related to
family satisfaction and F [W is also related tojob satisfaction. Although often theorized as a
direct result of increased WFC, there have been few studies that have examined the
satisfaction-WFC correlation in a manner that would allow causal attribution conclusions (for
exceptions, see Grandey, Cordeiro and Crouter, 2005 and Huang, Hammer, Neal and Perrin,
2004 who utilized longitudinal designs).

In addition to satisfaction, other sorts of outcomes have also been examined. At a
personal health level, many researchers have looked at how work-family conﬂict affects
experiences of certain mental and physical health outcomes such as depression, hypertension
and alcoholism. Generally, research has shown that increased levels of work-family conﬂict,
in either work to family or family to work directions, correlate with higher levels of

depression (Frone, Russell & Barnes, 1996; Thomas & Ganster, 1995). Some studies,

however, have found a relation between FIW conﬂict and depression but not WlF conﬂict
(Frone, Russell & Cooper, 1992a; Frone, Russell & Cooper, 1997).

A physical health outcome that has been examined in relation to work-family conﬂict is
hypertension. Thus far, only higher levels of FIW conﬂict have been found to be related to
higher levels of hypertension (Prone, Russell, & Cooper, 1997). Higher levels of cholesterol
are also related to higher levels of work-family conﬂict in general, but this study did not
parse out family to work and work to family directionality (Thomas & Ganster, 1995).

A ﬁnal health outcome that has received attention in the research literature is that of
alcohol use and its relationship to work-family conﬂict. Both F 1W and WIF have been linked
to increased levels of alcohol consumption (Frone, Russell, Cooper, 1992b; Frone, Russell, &
Barnes, 1996). Although the early research showed a clear positive relationship between
alcohol use and work-family conﬂict, it was impossible to infer causality from these studies
because they were done at a single point in time. Therefore the directionality could be either
way, such that alcohol use leads to more work-family conﬂict, or more work-family conﬂict
leads to more alcohol use. Frone, Russell and Cooper (1997) recognized this shortcoming of
the previous studies and conducted a longitudinal study to more clearly comprehend causal
factors in the alcohol/work-family conﬂict relationship. They found that higher levels of
work-to-family conﬂict at time one led to heavier alcohol consumption at time two. F amily-
to-work conﬂict did not appear to lead to higher alcohol consumption.

Emotional outcomes of WFC have been less well studied. The research on affect and
WFC that does exist tends to focus on general mood and not on speciﬁc emotions. While
related, moods and emotions are distinct constructs. In his 2004 book, Eysenck sums up the
differences in deﬁnitions as follows. Moods are typically deﬁned as lasting a relatively
longer duration, being less intense and not being related to a speciﬁc object. Emotions, on

the other hand, are typically shorter, more intense and are related to a speciﬁc object. For

 

 

{.

.th

I] '5 b:

“fibril

instance, one usually feels a discrete emotion, like anger, toward something in particular (e.g.
“I am so angry at my boss todayl”) whereas moods are more diffuse and objectless (e.g. “I’m
in an angry mood today.”) There have been some semantic inconsistencies in the literature
where researchers label their construct “emotions” even though they do not assess them as
occurring as the result of a discrete event (e.g. Matjasko & Feldman, 2006). I will discuss
ﬁndings in the context of the deﬁnition above rather than the inconsistent labeling authors
might use within a paper.

The WFC and mood research has shown that negative mood does tend to Spill over
from one life domain to another (Williams and Alliger, 1994). This negative mood can also
be contagious to others in close relationships. A recent study by Matjasko and Feldman
(2006) found that negative moods experienced at work not only spilled over to individual’s
experiences at home, but also affected the spouse’s moods at home.

Although the extant literature seems to suggest that negative mood has a role in WF C,
little research has focused on speciﬁc negative emotional outcomes of WF C. The few studies
that examine WFC in relation to speciﬁc emotions will be discussed in detail below. In order
to understand the relationship between WFC and emotions, it is important to examine WF C
in an emotion relevant theoretical framework. Thus, the next portion of this paper will focus
on a speciﬁc self-regulation theory and its relationship with four speciﬁc negative emotions:
guilt, worry, anger and frustration.

Self—Discrepancy Theory

When examining emotions and work-family conﬂict, one must consider the
theoretical ways in which such conﬂict might occur. One way to understand individuals’
experiences of conﬂict is through the lens of Self-Discrepancy Theory (SDT, Higgins, 1987).
Although few researchers have explicitly linked SDT with work-family conﬂict (see

Hochwarter et al, 2007 for an exception), it provides an illuminating way to think about the

10

exact nature of conﬂict and how it might relate to emotional responses. Like other theories
focusing on internal dissonance or self-inconsistency (e.g. Aronson, I992; Festinger, 1957),
SDT draws links to internal feelings of incongruity and subjective feelings of negative
emotion. Unlike these other theories, however, SDT make speciﬁc predictions about distinct
types of negative affect that result from different discrepancies. This speciﬁcity makes it
ideal for studying distinct negative emotions rather than a general, broad negative affectivity
that is discussed in the other theories.

Self-Discrepancy Theory was ﬁrst proposed by Higgins (1987) to describe the ways
individuals evaluate their actual selves in relation to two other possible selves. She proposes
that people have internal representations of both an “ideal self’ and an “ought self” with
which the “actual self” is compared. The actual self is deﬁned as “your representation of the
attributes that someone (yourself or another) believes you actually possess.” p. 320. In
contrast, ideal and ought selves are comprised of qualities that the individual does not
necessarily actually possess. The ideal self is deﬁned as “your representation of the attributes
that someone (yourself or another) would like you, ideally, to possess” (p. 320). Higgins
goes on to elaborate that the ideal self is comprised of characteristics that are hopes,
aspirations or wishes. The ought self, on the other hand, is deﬁned as “your representation of
the attributes that someone (yourself or another) believes you should or ought to possess” (p.
321). This is elaborated to include characteristics of the self that are related to a sense of
duty, Obligations or responsibilities.

Higgins proposes that actual/ought self-discrepancies are related to a different set of
speciﬁc negative emotions than actual/ideal self discrepancies. Past research into emotional
responses has tied dejection related emotions, such as depression and sadness, to the failure
to obtain positive outcomes (e.g. Lazarus, 1991). In the framework of SDT, dejection related

emotions should therefore be related to actual/ideal self-discrepancies because this

11

discrepancy represents a failure to obtain one‘s wishes and aspirations. On the other hand,
past research has tied agitation related emotions, such as guilt and threat, to the presence of
negative outcomes (Watson & Pennebaker, 1989). Because actual/ought self-discrepancies
are essentially a person’s failure to achieve a minimum of what they believe it is his/her
“duty” to achieve, this situation represents the presence of negative outcomes. Thus,
actual/ought discrepancies would be more strongly related to agitation type negative
emotions.

Since its proposal in 1987, Higgin’s SDT theory has been examined numerous times
and general support has been found for the actual/ideal discrepancy relationship to dejection
type negative emotions and the actual/ought discrepancy relationship to agitation type
negative emotions (e.g. Bizman, Yinon & Krotman, 2001; Higgins, Shah & Friedman, I997;
Strauman & Higgins, 1987). Although a few researchers have found more equivocal results
(Bruch, Rivet & Laurenti, 2000; Carver, Lawrence & Scheier, 1999), it seems reasonable to
adopt the SDT framework when considering speciﬁc emotional responses to internal self-
conﬂict.

In relation to work-family conﬂict, it is important to consider what type of self-
discrepancy might be activated when an individual experiences WFC. With the changing
nature of the modern workforce and an increasingly large number of dual earner families or
single parent households, many WFC researchers have noted the increased role demands
working adults are faced with (e.g. Grandey, Cordeiro & Crouter, 2005; Higgins, Duxbury &
Lee, 1994). The traditional family structure in which one parent (usually the man) was
responsible for supporting the family through external work and the other parent (usually the
woman) was responsible for managing the family within the home has now become more of
the exception than the rule. Both parents now ﬁnd themselves in situations in which they

mustjuggle competing demands from both work and family roles. Although the number of

12

roles working adults inhabit has increased, internalized societal norms of adequate
performance in these roles have been slower to change (Livingston & Judge, 1998). Thus,
performance norms that could reasonably be obtained when individuals focused on one role
predominantly are less easy to obtain when individuals add roles that are equally demanding.
As described above, work-family conﬂict has been deﬁned as instances in which meeting
goals in one domain make it impossible to meet goals in the other domain, whether it be
through time, strain or behavior incompatibilities (Greenhaus & Beutell, 1985).

Thus, work-family conﬂict can be understood in terms of failure to meet obligations
an individual feels are necessary for acceptable performance in either the work or the family
role. In this way, conﬂict more closely resembles the actual/ought self-discrepancy than the
actual/ideal self-discrepancy. The failure to live up to what an individual sees as his/her duty
or obligation to family or work represents the presence of negative outcomes. Following
SDT, the negative emotions that should be most closely aligned with WFC are the agitation-
related types.

In support of conceptualizing WFC as an actual/ought rather than an actual/ideal self-
discrepancy, one can look to how WFC is traditionally measured. In a widely used scale
developed but Carlson, Kacmar, and Williams (2000), the items ask responders to indicate
agreement with a variety of statements. For example, in measuring time-based work-
interfering-with-family, one item reads: “The time I must devote to my job keeps me from
participating equally in household responsibilities and activities.” The use of speciﬁc words
such as “responsibilities” falls directly in line with Higgins’s (I987) conceptualization of
actual/ideal discrepancies such that an individual notices the failure of living up to
obligations or responsibilities. The term “responsibility” occurs throughout the scale of WF C

and supports the conceptualization of WFC as an actual/ought discrepancy.

13

For these reasons, the actual/ought discrepancy seems to describe work-family
conﬂict well. In short, the conﬂict exists because individuals are feeling tension between two
obligations. This is not to say, however, that actual/ideal discrepancies do not play a role in
work and family satisfaction. For example, it is easy to imagine an individual who wishes
that she is able to be home when children get off from school (the ideal role) and feels some
dissatisfaction that her work role prevents it. This, however, is not a discrete example of
conﬂict; it isjust an illustration of how the actual role is different than the ideal role. The
fact that her work schedule always prevents her from being home at the end of the school day
means that neither she nor her family can realistically expect that she be home at the time.
Therefore, her presence at home is not an obligation but rather a wish.

Conﬂict occurs when an individual is expected by himself or others to perform in one
role despite interference from another role—in other words, it is a clash of obligations. That
is the reason I am focusing on actual/ought instead of actual/ideal discrepancies as a means to
understand the emotional responses likely to occur in WF C. I am not stating that actual/ideal
discrepancies cannot occur at the intersection of work and family lives, just that such
discrepancies do not qualify as conﬂict. The speciﬁc emotions I will focus on, therefore, are
those of the actual/ought discrepancies, namely agitation based emotions.

Due to the sharp focus on the self, SDT emphasizes the more traditional self-blame
related agitation-type negative emotions like guilt and worry. I propose, however, to expand
the conceptualization of agitation-type negative emotions to include two more negative
emotions, anger (hostility) and frustration. Typically anger and frustration have been deﬁned
as outwardly focused emotions consisting mainly of other-blame for failures or unwanted
circumstances (Vansteelandt & Van Mechelen, 2006; Kuppens & Van Mechelen, 2007).
This conceptualization has been challenged in the literature, however, and some research

suggests that anger and frustration can and do have very real relationships to situations of

14

self-blame, such as those that may arise from WFC (Kulik & Brown, 1979; Parkinson, 1999).
In order to more fully understand the unique ways in which guilt, worry, anger and
frustration can be related to WFC, it is necessary to take a closer look at each emotion in turn.

Because discrete emotions have been infrequently examined in relation to WFC, a
portion of this study focuses on discovering whether individuals do in fact experience each of
these four emotions with regard to WFC. Figure I lays out the general assertion that people
who experience higher levels of the two directions of WFC, WIF and FIW, will also report
higher instances of feeling guilt, worry, anger and frustration in the context of that conﬂict.
Work and family importance act as moderators in the model such that individuals who place
greater emphasis on these domains will experience negative emotions in relation to WFC
differently. Speciﬁcally, individuals who place greater importance on family will experience
more WIF negative emotions than individuals who place less importance on family in
response to similar levels of WIF. A similar moderation relationship is proposed for work
importance such that individuals who place greater importance on work will experience more
FI W negative emotions than individuals who place less importance on work in response to
similar levels of FIW. In addition, gender will moderate the experience of worry in both
directions of WFC such that women will be more likely to experience this emotion than men.
For the direction of FIW, gender will moderate the relationship between FIW and frustration
such that women will experience it more often than men. The rationale for this model will be
more fully explained below as each emotion is examined in turn.
Guilt

Of the four emotions targeted in this study, guilt has had the most direct mention in
the WFC literature. Before examining those ﬁndings, however, it is necessary to deﬁne the

term guilt for clarity in further discussions of the construct.

IS

Figure 1: Proposed Model of the relationship between WFC and emotions, with Work

importance, Family Importance and Gender as moderators.

 

FIW
Time

 

 

 

 

FIW
Strain

 

 

 

 

 

F IW
Behavior

 

H5

 

 

 

 

Work
Importance

 

 

 

 

WIF
Time

 

Family
Importance

 

 

 

 

WIF
Strain

H6

 

‘4)

H7a

 

H8

 

 

Gender

 

 

H7b

 

 

 

WIF
Behavior

 

\/ \/

 

 

l6

 

Hla

H23

H33

H4a

Hlb

H2b

H3b

 

FIW
Guilt

 

 

F IW
Worry

 

 

FIW
Anger

 

 

 

F 1W
Frustration

 

 

WIF
Guilt

 

 

WIF
Worry

 

 

WIF
Anger

 

 

 

WIF
Frustration

 

 

 

The concept of guilt has been the focus of much research, spanning psychology
disciplines from cognitive to clinical to social. Deﬁnitions of guilt vary accordingly, such
that social psychologists place more emphasis on relational variables of guilt (e.g. Baumeister
et al, 1994) while cognitive psychologists place more emphasis on internal appraisals (e.g.
Bemdsed et al, 2004). In order to make sense of the diverse literature on the subject, please
see Table 2 for a review of guilt deﬁnitions.

Although the deﬁnitions differ in speciﬁcity and focus, they basically agree on some
general characteristics of the emotion guilt. For the purposes of this paper, guilt will be
deﬁned as a negative, local self-evaluation that one has failed to meet one ’s own or someone
else ’s moral standards. This emotion can be felt in response to things done or not done. It is
clear in this deﬁnition how guilt is theoretically closely tied to actual/ought self-discrepancies
in SDT.

Although research on the direct relationship between WFC and emotions is sparse, guilt is
one particular emotion that is mentioned often in the literature, although it is rarely directly
measured. From the early conceptualizations of WFC, researchers have often mentioned
guilt as a possible response to feelings of failure at effectively balancing work and family life
(e.g. Casper, Martin, Buffardi, & Erdwins, 2002; Judge, Ilies & Scott, 2005; Livingston &
Judge, 2008). Some researchers have speciﬁcally tied this to “maternal guilt”, suggesting
that women are particularly vulnerable to feelings of guilt when they experience WFC
(Marecek and Ballou, 1981). Although guilt related WF C is often considered as gender
speciﬁc, there is no empirical basis to suggest that men cannot and do not experience guilt in
relation to WFC as well. Indeed, with both sexes adopting more roles in the work and family
domains, it is likely that men would be likely to experience role conﬂict as well as women.
In addition, a recent literature review and a meta-analysis suggest that men and women

experience similar amounts of conﬂict (Eby et al 2007; Ford et al, 2007).

I7

Table 2
Definitions of Guilt

 

Source

Deﬁnition

 

Baumeister, Stillwell &
Heatherton, 1994

Lewis, 1971

Hochwarter, Perrewe, Meurs,
& Kacmar, 2007

Higgins, 1987

Judge, Ilies & Scott, 2006

Bemdsed, van der Pligt,
Doosje, & Manstead, 2004.

“By guilt we refer to an individual’s unpleasant emotional
state associated with possible objections to his or her
actions, inaction, circumstances or intentions. Guilt is an
aroused form of emotional distress that is distinct from fear
and anger and based on the possibility that one may be in
the wrong or that others may have such a perception.” P.
245 (emphasis in the original)

“The prototypical cause of guilt would be the inﬂiction of
harm, loss, or distress on a relationship partner.” P. 245

“In guilt, the self is not the central object of negative
evaluation, but rather the thing done or undone is the focus.
In guilt, the self is negatively evaluated in connection with
something but is not itself the focus of the experience.” P.
30 (emphasis in the original)

“Work induced guilt is conceptualized as an emotional
reaction arising from the interference of one’s job activities
on the family domain” p. 126

“Guilt involves feeling that one has broken one’s own rules
concerning how one ought to conduct one’s life.” P. 323

“Guilt is an inward-focused emotion in that it is directed
toward oneself—in other words, a self-conscious emotion.”
P. 782

“guilt... involves interpersonal appraisals, for example, the
feeling of letting one’s parents down.” P. 57 “Guilt... seems
to be an emotion in which peOple think about their
transgressions, for example, by worrying about being
rejected by others.” P. 57 “Guilt is an emotional state in
which one focuses on other people... Guilt might be a
product of interpersonal harm.” P. 57

 

18

Despite its occasional presence in the literature, the concept of guilt as an outcome of
WFC is generally not well articulated. For instance, articles are more likely to discuss guilt
in the context of a speciﬁc directionality of WFC, namely work interference with family (e.g.
Hochwarter, Perrewe, Meurs & Kacmar, 2007). Additionally, guilt is also mostly referred to
in reference to two types of WFC, the time-based and the strain-based types or as an
undifferentiated aggregate of all types (e.g. Judge, Ilies & Scott, 2006; Marecek & Ballou,
1981).

There are three notable studies that explicitly address emotional responses to work-
family conﬂict. Judge, Ilies and Scott (2006) discuss guilt as a response to stressors that
occur at work or at home. Using an experience sampling methodology, they showed that both
family to work conﬂict and work to family conﬂict were related to higher levels of guilt in
the targeted domain. Although this study has much to recommend it in terms of its
longitudinal, within person design, the methodology employed to measure emotions leaves
something to be desired. Guilt was measured using the momentary version of the PANAS,
which is an untargeted measure assessing current mood. The researchers then used the
setting of each of these assessments as a proxy for directionality of the feeling. Respondents,
however, just reported their State affect, without reference to what they felt “guilty” about. It
is possible that their reference for these emotions were events that occurred beyond the
setting in which they reported them. Indeed, it is not hard to imagine an individual mulling
over a work interfering with family event while at work. In this case, a state measure of guilt
would attribute that emotion to family interfering with work—the opposite direction than the
experienced emotion. This non-referenced report of affect fails to clearly indicate the exact
trigger for the experienced affect.

The second study was also undertaken by Judge. Livingston and Judge (2008)

examined FIW and WIF in relation to guilt and gender role orientation. They found that guilt

l9

was positively related to both FIW and WIF and that gender role orientation moderated this
relationship such that those who adhered to more traditional gender role orientations
experienced more guilt in relation to both FIW and WIF. They measured guilt in much the
same ways as Judge, [lies and Scott (2006) in that the measure was not referenced on any
speciﬁc event.

The third notable study addressing guilt and WF C was conducted by Hochwarter et al
(2007). They studied the relationship between work-induced guilt and the ability to manage
resources and the resultant impact on job and life satisfaction. Although their measure was
targeted such that it speciﬁcally referenced WF C, it only measured one direction of guilt
experienced (work interfering with family). In addition, their ﬁndings proved a little
startling, suggesting that individuals who experience more guilt actually had higher life
satisfaction. These ﬁndings deserve ﬁthher scrutiny in order to adequately address the
seemingly counterintuitive results and to ensure such conclusions are not sample speciﬁc
artifacts.

These studies suggest that guilt can be experienced in relation to both FIW and WIF.
Although Hochwarter et al (2007) assumed that individuals would be more likely to
experience guilt in the WIF direction (indeed, that is the only direction they measured),
Livingston and Judge (2008) found that guilt was signiﬁcantly related to both directions of
WFC. None of the studies divided WFC into component parts of time, strain or behavior
based conﬂict, so any unique relationships between those facets are unknown. It is likely that
all facets are related to guilt, although there may be stronger relationships for some than

others.

Hypothesis 1 a: Individuals who experience more FIW (stress, time and behavior-

based) will experience more FIW guilt.

20

Hypothesis 1 b. Individual who experience more WIF (stress, time and behavior-

based) will experience more WIF guilt.

Worry (Anxiety)

Like guilt, the study of worry and anxiety is broad and varied. Most research on
anxiety is couched in clinical terms, such that too much worry is a disordered and mentally
unhealthy state. Such clinically high levels of worry are termed Generalized Anxiety
Disorder (GAD) and are considered a documented mental health disorder in the DSM-IV.
GAD has been linked to a variety of negative outcomes such as social withdrawal,
depression, alcoholism, and suicide (Yates, 2007). GAD is, in fact, considered a negative
outcome in and of itself due to its negative impact on an individual’s ability to function in a
normal manner (Yates, 2007). Although important in the study of mental health, this extreme
type of worry and anxiety will not be the focus of this paper. Instead, it is necessary to deﬁne
won'y as a non-disordered, discrete emotion similar to guilt, anger and frustration.

While the majority of the literature focuses on clinical levels of anxiety and tends to
loosely deﬁne the “worries” that contribute to GAD, there is a subset of the literature that
focuses on discrete worry as a normal, non-disordered response to situations. In its basic
form, worry can be understood as a cognitive reaction of a series of negative thoughts in
response to an event or situation (Borkovec, Robinson, Pruzinsky, & DePree, 1983; Sugiura,
2007). This series of thoughts is generally conceptualized as uncontrollable, although ways
of coping with excessive worry have been developed in clinical research (e.g. Hunt, Wisocki
& Yanko, 2003; Myers, 1998). Santanello and Gardner (2007) reﬁne this idea of worry
further, stating that worry is “triggered by the perceptions of a threat, is an effort to problem

solve or discern how to cope with negative events” (p. 321). Worry also contains an element

21

of heightened arousal such that the individual feels “antsy” or “keyed up” (Larson, Nitschke,
& Davidson, 2007).

While several deﬁnitions incorporate a negative event as a trigger for won'y, it is
important to note that worry can occur in the absence of a speciﬁc event. Indeed, individuals
can worry about feared possibilities. Borkovec and Roemer (1995) conducted a study that
examined sorts of things individuals generally worry about. They discovered that won-y was
used as a means to think about how to avoid future problems or feared events. While this
future orientation is interesting, it does not appear to be necessary for worry, with individuals
being capable of worrying about actual past events as well as hypothetical future ones.
Because this paper examines worry in connection with discrete events of work-family
conﬂict, we will focus on worry as a reaction to a speciﬁc event.

For the purposes of this paper, I propose to deﬁne worry as a series of negative
thoughts triggered in reaction to an event or situation that the individual deﬁnes as
threatening and results in increased arousal. The event or situation that triggers the negative
thoughts can be internal or external.

Like guilt, Higgins’ self-discrepancy theory explicitly identiﬁes anxiety as an
“agitation-type” emotion that results from actual/ought discrepancies. The research on self-
discrepancy theory shows the trend that anxiety is uniquely related to actual/ought rather than
actual/ideal discrepancies (Higgins, Shah & Friedman, I997; Roney & Sorrentino, 1995;
Scott & O’Hara, 1993). Following the conceptualization of WFC as an actual/ought
discrepancy, it stands to reason that anxiety in the form of targeted worries would occur in
response to WFC.

Although non-clinical levels of worry have not been explicitly studied in the WP C
literature, GAD has been examined. Frone, Russell and Cooper (l992b) conducted a study

that examined the longitudinal relationship between higher levels of WFC and clinical levels

22

Of worry and discovered that those who experienced more WFC later exhibited more cases of
GAD. Although the researchers did not speciﬁcally measure discrete worries about WFC,
such ﬁndings seem to indicate that increased worry is a consequence of conﬂict.

Research on self-perceptions of in-role performance would back up this contention.
Although worry has been infrequently addressed in the WFC literature, it is often the topic of
family research. Numerous studies have examined parental worries about child emotional,
social and academic progress and also over the parent’s own perceived success or failure as a
“good parent” (Stickler, Salter, Broughton & Alario, 1991; Measelle, Ablow, Cowan and
Cowan, 1998). Because all three types of work-interfering-with-family conﬂict (i.e. time-,
strain-, and behavior-based) could easily call into doubt the individual’s perceptions of their
“success” within the family role, it stands to reason that worry would result.

Although worry is not generally examined in the work context, the basic logic still
stands. Just as an individual can hold concerns about his/her success as a “good parent”,
he/she can hold concerns about his/her success as a “good employee”. In this case, it is
possible that family-interference-with-work conﬂict could be one source of these worries
about competency and concerns about possible failure within the role.

These research ﬁndings suggest that it is reasonable to suppose that worry is
experienced in response to both WIF and FIW forms of conﬂict. Because no prior research
has been conducted on worry as having a speciﬁc relationship with WFC, there is no way to
tell if the speciﬁc facets of behavior, strain or time based conﬂict would have differential
relationships with worry.

Conceptually, it is easy to see how all three types and each direction of WF C could
be related to worry. For example, if an individual has to miss her child’s soccer game to
attend a work meeting, she might feel worried about her competencies as a mother and how

her family will react to her absence. Likewise, the worry can occur in the opposite direction,

23

where if the individual chose to attend the soccer game instead of the meeting, she may feel
worried about being an inadequate employee or being perceived as such by coworkers and
supervisors. Similar arguments can be made for strain based WFC. An individual who is
unable to contribute to his family in the evening after a particularly stressful day at work may
worry about his inability to fulﬁll his responsibilities as partner and father. In the other
direction, a worker who is feeling stress from family problems may worry about his
competencies at work. Finally, behavior based conﬂict can also ﬁt in the worry framework.
When an individual engages in a behavior at home that works on the job, such as being
overly stem with a young child, she may feel worried about her impact on her family or her
competencies as a mother. At work, if the individual is overly cajoling with a coworker, a
behavior that is functional at home, she may worry about her inability to behave
appropriately or what her coworkers may think of her. Therefore, it is likely that all three

facets of WFC will be positively related to worry.

Hypothesis 2a: Individuals who experience more F IW (stress, time and behavior-
based) will experience more FIW worry.
Hypothesis 2b: Individuals who experience more WIF (stress, time and behavior--

based) will experience more WIF worry.

Anger

The emotion literature has often deﬁned anger as a bookend to guilt, such that guilt
results from feelings of self-blame whereas anger results from feelings of other-blame (e.g.
Smith & Lazarus, 1993). In line with the majority of research on anger, Smith, Haynes,
Lazarus and Pope (I993) posited a model of anger in which an individual appraises an

unpleasant situation as having an external locus of control such that the negative experience

24

is someone else’s “fault”. Their model of guilt is similar, excepting that the locus of control
is internal, such that negative experiences are one’s own “fault”. If we accept this deﬁnition,
there is little theoretical basis to include a measure of anger in relation to work-family
conﬂict. In SDT, the negative emotional responses to actual/ought self-discrepancies arise
due to negative self-appraisals. Thus emotions stemming from other-appraisals would seem
irrelevant.

Although the majority of theoretical research on anger continues to make the “other-
blame” distinction, empirical work on anger reveals that the ways individuals experience the
emotion of anger are not so cut and dried. In a study of emotional reactions to social
criticism, Gilbert and Miles (2000) found that contrary to theoretical deﬁnitions of anger,
“blaming self” was related to increased anger proneness and hostile attitudes but “blaming
others” was not. This context may be especially relevant when considering actual/ought self-
discrepancies in WFC due to the potential for feelings of self-criticism in failing to fulﬁll
obligations.

Other studies have also found a similar trend in anger experiences in response to self-
blame. Kulik and Brown (1979) found that when individuals appraised situations as other
thwarting such that other people were responsible for negative outcomes, they responded in
the predicted manner with other directed anger. When individuals appraised situations as
self-caused thwarting such that their own actions were responsible for negative outcomes,
they experienced anger as well, although this was self-directed anger. Parkinson (1999) also
found a similar response, showing that people did indeed respond with anger to situations
they deﬁned as self-blaming.

These studies suggest that an external locus of control is not necessary for an
experience of anger. Indeed, it would appear that individuals can experience anger in much

the same Situations that may inspire guilt in other people. Although not traditionally

25

considered in SDT research, it would appear that anger could also be considered an agitation-
type negative emotion and thus be related to actual/ought discrepancies.

When attempting to understand the exact deﬁnition of anger, the differences between
theoretical development and empirical ﬁndings muddy the waters. For the purposes of this
paper, anger will be considered in its self-referenced form. In order to further understand the
deﬁnition of anger and how it is distinct from other negative emotions, I will now brieﬂy turn
to the appraisal literature on emotions.

As mentioned above, anger is usually conceptualized as resulting from an appraisal of
“other-blame” in a negative situation, but in order to fully understand this concept, it is
necessary to look at other appraisals that are relevant to the experience of anger. In a study
that attempted to ﬁnd speciﬁc and necessary appraisal conditions for the experience of anger,
Kuppens, Van Mechelen, Smits and De Boeck (2003) looked at four situation appraisal types
that contributed to feelings of anger. These appraisal types were 1) goal obstacle, where an
individual feels his/her progress towards a goal is impeded, 2) other accountability, where an
individual blames another for a negative situation, 3) control, where the individual believes
that he/she or another person had control over a situation, and 4) unfairness, where the
individual believes that the negative situation was unfair in some way. All four of these
appraisal heuristics were found to be related to experiences of anger, but no single one was
necessary or sufﬁcient for experiencing anger. In addition to providing further support for
the idea that appraisals of other blame are not necessary to experience anger, this study
highlights other potential causes of anger that are more relevant to the study of work-family
conﬂict. In the context of this study, the goal obstacle, control and unfairness appraisals may
all be theoretically relevant when considering individual appraisal of work-family conﬂict

situations.

26

Taking into account the Kuppens et al (2003) appraisal work, I will deﬁne anger as a
negative evaluation of a situation that one believes to be unfair, controllable, or goal
impeding or a combination of the three. This evaluation can be externally or internally
based, although in this context it is most likely to be self-referenced. Like guilt, this emotion
can be felt in response to things done or not done.

As for the treatment of anger in the WFC literature, there has been only one study
that directly examines hostility (an attitudinal variation of anger) and it is described above in
the section on guilt. In addition to looking at guilt, Judge, Ilies and Scott (2006) also looked
at the construct of “hostility”, ﬁnding both family to work conﬂict and work to family
conﬂict were related to higher levels of hostility in the targeted domain. The methodological
limitations mentioned above also apply to the measures of hostility, so further work is needed
to clarify the precise relationship between hostility and WFC.

A few other studies exist that less explicitly examine anger or hostility in the context
of work-family conﬂict. Studies by Haines, Marchand and Harvey (2006) and Matjasko and
F eldman (2006) examine emotional crossover effects between marital partners. Both studies
refer to anger or hostility as emotions that can be experienced in the work domain and then
bleed over to family relationships. Although not explicitly linking anger emotions to work-
family conﬂict, these studies Show the rationale in considering anger in relation to work-
family conﬂict.

The single WF C study on anger by Judge, Ilies and Scott (2006) suggests that people
are likely to endorse feelings of anger in relation to both WIF and FIW. While they did not
divide their results speciﬁcally by facet, there is little basis to presuppose that time-based,
strain-based or behavior-based conﬂict would function differently with regard to anger. All
facets are likely to be related to experiences of anger. For example, an individual who

experiences a time conﬂict between work and family may feel angry at himself for being

27

 

unable to adequately balance work and family time responsibilities or arrange a functional
schedule. An individual who experiences strain-based conﬂict may be angry at herself for
her perceived inability to balance competing stressful demands, or she may be angry at the
domain causing the stress (her family or her coworkers). For behavior-based conﬂict, the
relationship to anger is less clear. It is conceivable that an individual could feel internalized
anger at being unable to behave appropriately in speciﬁc roles, but this relationship may be
less likely that the other facets. Therefore, the relationships between the types of conﬂict
might be of different strengths such that anger is experienced more often in response to one

than another type of conﬂict.

Hypothesis 3a: Individuals who experience more FIW (stress, time and behavior-
based) will experience more FIW anger.
Hypothesis 3b: Individuals who experience more WIF (stress, time and behavior-

based) will experience more WI F anger.

Frustration

Frustration, like anger, is not explicitly linked to SDT, although its theoretical
connection to “agitation” type emotions is obvious. The inclusion of frustration as a discrete
emotion in this study needs explanation because most theories of frustration deﬁne it as a
component of anger and not a speciﬁc emotion per se (e.g. Kuppens, Van Kenworthy,
Canales, Weaver & Miller, 2003; Mechelen, Smits, De Boeck, & Ceulemans, 2007).
Although often theorized as a condition of anger, research has shown that frustration is
neither a necessary or sufﬁcient condition for anger, and that feelings of frustration can occur

without feelings of anger (Clore & Centerbar, 2004; Roseman, Spindel & Jose, 1990). Using

28

these ﬁndings, 1 will treat frustration as a discrete emotion that can be related to but is
ultimately fundamentally different than anger.

Extant research on frustration is mostly tied to speciﬁc scenarios in which individuals
are “frustrated” in the pursuit of carefully prescribed goals such that they do not receive
expected or promised rewards for their behaviors (e.g. Anderson & Bushman, 2002; Kuppens
& Van Mechelen, 2007). This research base has led to a relatively simple and well-received
deﬁnition. As Anderson and Bushman (2002) put it “frustration can be deﬁned as the
blockage of goal attainment” (p. 37).

Similarly to anger, frustration is often deﬁned as having an external locus of control,
such that an individual is often frustrated by an external event or person (Fox & Spector,
1999). Unlike anger, frustration can conceptually more easily occur in conditions where
there is no clear indication of Situational control or blame. It is easy to imagine a situation in
which a person becomes frustrated while learning a new computer software because he/she is
failing to master the task as quickly and easily as hoped. This situation is not due to someone
else’s action or to a controllable circumstance. In cases such as these, it may be more
reasonable to view frustration as having an internal locus of control, such that an individual
can be frustrated with his/her own performance or failure to obtain a desired goal.

Due to these considerations, I propose the following deﬁnition of frustration.
Frustration is deﬁned as a negative emotional reaction to set of events or circumstances that
one believes to be blocking his/her attainment of a desired goal. The source of goal
blocking can have either an external or internal locus of control.

It is important to note that although the deﬁnition of anger includes a facet that
includes goal impediment, anger and frustration are considered distinct emotions. Goal
impediment and frustration does not always lead to anger and anger can exist without goal

impediment (Roseman, Spindel & Jose, 1990). Although similar, it is clear that these

29

emotions can be experienced separately, and thus warrant attention and examination in their
own rights.

In the case of SDT and work-family conﬂict, goals can be described as domain
speciﬁc such that they are either work or family referenced. When work and family goals
come into conﬂict, the individual experiences a “blocking” of fulﬁlling or reaching one of the
goals. This results in a perception of self-discrepancy due to the individual’s inability to
fulﬁll a perceived obligation in one of the roles.

Although much of the frustration literature uses very speciﬁc goals in describing
frustration, it is likely that in the context of work-family conﬂict that these goals might be
more broad. For instance, an individual might hold two simultaneous goals of “being a
dependable and responsible worker” and “being a caring and supportive parent”. A Situation
in which this individual must choose whether to attend an important work meeting or watch
his/her child perform in a soccer game would lead to the blocking of one of these goals. If
this individual chooses to attend the meeting, the goal of “being a caring and supportive
parent” might be perceived as blocked, resulting in frustration.

Most frustration research focuses on contexts of performance rather than social or
relational contexts and it has never been explicitly studied in relation to either SDT or WFC.
One study conducted by Kuppens and van Mechelen (2007) has looked explicitly at the
relationship between frustration, anger and affiliative and achievement goals. In addition to
providing further support that anger and frustration can have unique relationships to other
constructs, this study showed that for women in particular frustration occurred in both
afﬁliative and achievement contexts. This shows that people can and do experience
frustration in relation to social and relational contexts, which supports the study of frustration
in this paper. It also highlights a potentially important sex difference in how men and women

experience frustration. Because WFC can be conceptualized as a conﬂict between highly

30

afﬁliative roles, especially family, these differential ﬁndings have potential to inform how
and why different people may respond to WFC.

Because no research has yet looked at the role of frustration as an emotional reaction
to WFC, the proposed relationship is exploratory. Within the framework of SDT, WFC is
conceptualized as the failure to adequately meet felt obligations. These obligations can exist
towards both work and family and can be understood as goals of desired behavior in each
role. As frustration is deﬁned as a reaction to goal impediments, it stands to reason that an
individual could experience frustration in response to both WIF and F IW directions. In
addition, within the context of WF C, these goals could be deﬁned as either time based, strain
based or behavior based. For example, an individual might have a goal to attend all of his
child’s soccer games and also a goal of performing well at a meeting scheduled at the same
time as the soccer game. Regardless of which event the individual chooses, one goal will be
blocked, which could result in frustration. The same can be said for strain-based conﬂict.
An individual may have a goal to be emotionally present during evening family time, but the
demands of their job may leave them so stressed out that they fail to reach this goal. This
also stands for behavior-based conﬂict. An individual might have a goal of appropriate
behavior at work, and experiences goal blockage when they realize the behavior appropriate
at home is inappropriate at work. Therefore, it is likely that individuals can and would

experience frustration in relation to time-, strain- and behavior-based conﬂict.

Hypothesis 4a: Individuals who experience more FIW (stress, time and behavior-
based) will experience more FIW frustration.
Hypothesis 4b: Individuals who experience more WIF (stress, time and behavior-

based) will experience more WIF frustration.

3l

Work and Family Importance

Experiences of negative emotions may be moderated by certain individual differences
in the importance placed on work and family roles. The work-family conﬂict research
reveals that role importance affects the extremity with which individuals experience negative
outcomes. Carlson and Kacmar (2000) found that when high importance was placed on the
family, antecedents from the work domain were more important in determining conﬂict.
Likewise, they found that when high importance was placed on the work role, antecedents
from the family domain were more important in determining conﬂict. Greenhaus,
Parasuraman, Granrose, Rabinowitz and Beutell (1989) also found the importance of the
work role to impact experiences of conﬂict for both men and women.

It is important to note that high family and high work role importance are not
mutually exclusive constructs. Indeed, it is possible that both of these roles are highly valued
at the same time. This dual importance can have important implications. Carlson Kacmar
(2000) found that individuals that highly valued both roles had highly related job and life
satisfaction variables as compared to their more stratiﬁed counterparts.

Because guilt is deﬁned as a reaction to a perceived failure to meet moral standards,
people who place more importance on certain roles may be more likely to assign moral value
to those roles. This may impact their likelihood to experience guilt as a reaction to different
directionalities of conﬂict. Worry is deﬁned as negative thoughts triggered by an event
deﬁned as threatening. Individuals who place more importance on certain roles may feel
more events are threatening to their abilities to perform competently in that role. Anger is
deﬁned as a reaction to unfair, controllable or goal impeding events, so people who place
more importance in certain roles may be more likely to develop clear notions of fairness or
numerous goals in these roles. This may impact their likelihood to experience anger as a

reaction to different directionalities of conﬂict. Finally, frustration may also be moderated by

32

individual differences in the importance placed on work and family roles. Depending on the
importance of the role, an individual can have differing numbers of goals and these goals can
be differently salient. A person who places more importance on work roles, for example,
may be more likely to have a greater number of Speciﬁc goals for performance and behavior
at work. These goals may also be more salient for a person who places great importance on
work as Opposed to someone who places less importance on work. See Figure 2 for graphs
of these proposed relationships.

Hypothesis 5: The relationship between F IW and all F IW emotions will be moderated
by work importance, such that individuals who place greater importance on their work will
be more likely to experience the negative F IW emotions

Hypothesis 6: The relationship between WIF and WIF emotions will be moderated by
family importance, such that individuals who place greater importance on their family will be

more likely to experience the negative WIF emotions.

Figure 2: Proposed moderation relationship of job and family importance

 

 

 

 

 

 

 

 

3 _= 5 '“" High family
53’ .2” ,x’ importance
2:
r: 1”, —— Low family
'5 x", importance
0 ,v"
.2 /
r» a
Z ._1
Low High
WIF
g L, '“"' Highjob
O ,J .
‘g: .99 x" importance
s= ,x
; /,v’ — Low job
E "'5’ Importance
o ,o’
.2 f
‘5
or) 3
2 .3
Low High
FI W

33

Gender Diﬂerences

The ﬁeld of WFC has Often been implicitly and explicitly linked to gender
differences. Indeed, the entire concept of work-family conﬂict gained more prominence and
attention as more and more women were joining the workforce and adding on the role of
employee to that of family member (Bureau of Labor Statistics, 2008). Although this gender
distinction is often interpreted as indicating that women experience more WF C as compared
to their male colleagues, research on WFC has failed to indicate consistent and clear gender
differences.

One recent meta-analysis conducted by Ford, Heinen, and Langkamer (2007)
examined gender as a moderator of work-interfering-with-family and family-interfering-with-
work effects and found no Signiﬁcant differences. Another literature review of the WP C
literature conducted by Eby, Casper, Lockwood, Bordeaux and Brinley (2005) found mixed
results in the literature regarding frequency of WFC as experienced by men and women.
They reported that the majority of studies had found no signiﬁcant differences between men
and women in regard to the amount of WF C experienced. These ﬁndings indicate that that
although often hypothesized, current research provides little evidence to support that women
and men experience different levels of WFC.

Although men and women may not experience different amounts of WF C, one study
suggests men and women may experience the work and family domains differently. Duxbury
and Higgins (1991) found gender differences in work and family involvement. Speciﬁcally,
they found that work involvement was a stronger predictor of WFC in women than in men
and family involvement was a stronger predictor of WF C in men than in women. This
ﬁnding may have some important implications for this study. Although men and women

may both be involved in work and family experiences, the may be differently involved in

34

each domain, perhaps leading to differences in experiences of WFC. While this study
explored antecedents of work and family involvement, it highlights how experiences of
variables such as involvement may have differential effects on men and women and may in
turn relate to WFC. It stands to reason, therefore, that men and women might display
differences in how they emotionally react to different types of conﬂict.

The research on sex differences in emotions has a similar disconnect between what is
theorized and what is revealed through empirical research. Although researchers often
theorize relatively strong and enduring gender differences in different kinds of emotions, a
look at the data shows that these differences are often insigniﬁcant, minor or occur in
circumscribed situations (e.g. Livingston & Judge, 2008; Silvfer & Helkama, 2007). In order
to more closely examine proposed gender differences in emotions, I will look at each of the
emotions highlighted above in further detail.

Guilt. Theoretical gender differences arguments for emotions such as guilt and worry
generally center around the idea that women experience these emotions more often than men.
This standpoint is argued by asserting that women are more internalizing than men, and men
are more externalizing than women (Garside & Klimes-Dougan, 2004). In this case, the
emotions of guilt and anxiety can be conceptualized as internalized emotions whereas
emotions such as anger and frustration can be conceptualized as extemalized emotions.
These differences might suggest that men and women experience these emotions with
different frequencies. The empirical literature on the subject is far ﬁ'om conclusive, however.

Guilt is the only emotion explicitly studied in the context of WF C and gender
differences. Livingston and Judge (2008) examined gender differences in experience of guilt
and related it to work-interfering-with-family conﬂict and family-interfering-with-work
conﬂict. They found that while gender in general was unrelated to experiences of guilt in

relation to WFC, gender role was related. They deﬁned gender role orientation as the extent

35

 

to which both men and women ascribed to traditional gender roles in which men are seen as
the breadwinners and women are seen as the family caretakers. Individuals who scored
lower on traditional gender role orientation viewed the roles of men and women within a
family as more egalitarian. Livingston and Judge (2008) measured both egalitarian and
traditional gender role orientation for both men and women and discovered that individuals
who ascribed to more traditional gender role orientations experience more guilt from FIW
and individuals who ascribed to more egalitarian gender role orientations experience more
guilt from WIF. These ﬁndings were true of both sexes.

As for gender differences in guilt in general, the ﬁndings are muddied. Although
there is some evidence that women report higher levels of moral emotions such as empathy,
guilt and shame (e.g. Tangney & Dearing, 2002), Silfver and Helkama (2007) recently
pointed out that these gender differences seem to be unusually dependent on the measure
used to assess guilt. Speciﬁcally, the measures tend to use examples and wording that may
be more salient to women and thus the measures themselves may be artiﬁcially inﬂating the
gender differences in those emotions. Because the measure utilized in this study does not
contain gendered items, I do not expect differences between men and women in experiences
of guilt.

Worry. Research on worry is similarly muddied. A meta-analysis conducted by
Feingold (I 994) found that while women appear to score slightly higher on levels of trait
anxiety, there were no gender differences in levels of social anxiety. A literature review
conducted by Simonds and Whiffen (2003) revealed a more distinct difference in anxiety
levels between men and women. They showed that women experience more anxiety than do
men. Even considering the possible gender differences in these studies, it is important to
keep in mind that in the context of the present study, I am looking at discretely triggered

emotions rather than general states. These differences may not be applicable in this case.

36

Because there is a dearth of literature on differences in worry or anxiety in response
to speciﬁc events, it is difﬁcult to speculate what gender differences might exist. Following
past research which states that women are more likely to experience higher trait-like anxiety
then men, we might suppose that women would be more likely to endorse worry responses to
WFC. Therefore, I hypothesize that there will be gender differences in experiences of WFC

worry.

Hypothesis 7a: Gender will moderate the relationship between F IW and F IW worry
such that women will be more likely to experience F [W worry then men.
Hypothesis 7b: Gender will moderate the relationship between WIF and WIF worry

such that women will be more likely to experience WIF worry than men.

Anger: Anger particularly seems to be an emotion that is often associated with higher
prevalence among men. This supposition is most likely due to the gender differences that
appear in speciﬁc behaviors thought to be the result of anger—namely aggression. A recent
meta-analysis by Archer (2004) indicated that there are signiﬁcant and robust gender
differences in prevalence and type of aggressive behaviors individuals engage in. A measure
of anger, however, displayed no gender differences. This suggests that men and women
might experience similar levels and frequencies of anger but they display it in different ways.
These ﬁndings were corroborated by another meta-analysis (Campbell, 2006), which again
showed no gender differences in anger.

Because this paper focuses on the experience of anger as referenced by a speciﬁc
conﬂict event and not on resultant behaviors triggered by that anger, it would appear that
gender differences are unlikely to exist in frequency or likelihood of experienced anger as a

response to WFC.

37

   

Frustration: Frustration as a discrete emotion has been less frequently studied than
the other emotions discussed in this paper. Therefore, there is even less information on how
men and women might differ in their experiences of this emotion. The studies that exist
suggest that there might be some gender differences that seem to fall along the lines of the
externalizing literature (Kuppens & Van Mechelen, 2007). That is, it would appear that men
might experience more frustration in response to speciﬁc events.

The study mentioned above by Kuppens and van Mechelen (2007) is one study that
looks at frustration as a potentially gendered construct. They show that frustration is
experienced differently by women and men, such that women experience frustration in both
afﬁliative and achievement contexts, whereas men experience frustration primarily in
achievement contexts only. This may indicate a possible gender difference in experienced
frustration as a result of WFC. Although both work and family realms likely have elements
of afﬁliative goals, it would seem more likely that family roles lack clear achievement goals
whereas work likely encompasses more of these. This might suggest that women would
respond with equal amounts of frustration to both WIF and FIW situations whereas men may
experience more frustration in the FIW direction because the achievement goals of work

might be disrupted by family demands.

Hypothesis 8: Gender will moderate the relationship between FIW and F IW

frustration such that men will be more likely to experience FIW frustration than women.

Job and Life Satisfaction
Although the main goal of the research is to more fully understand the emotional

landscape of WFC, the next step in research is to link the WFC related negative emotions to

38

important outcomes. Two outcomes that have been frequently studied in the WFC literature
are those of job and life satisfaction.

Job Satisfaction:

There is quite a large literature on the importance of job satisfaction on many
organizationally relevant outcomes, such as decreased performance, higher turnover
intentions and less organizational commitment (Harrison, Newman, & Roth, 2006). Targeted
work family conﬂict research has looked at these more down stream outcomes as related to
conﬂict as well. The general ﬁnding is that those who experience more work interference
with family conﬂict experience less commitment to the organization as manifested in higher
turnover intentions, more absenteeism, and more leaving work early (Boyar, Maertz, Pearson,
& Keough, 2003; Boyar, Maertz, & Pearson, 2005).

The theoretical ﬁamework for the relationship of work-family conﬂict to job
satisfaction is as follows: if work interferes with family, this leads to a change in satisfaction
with work, such that the higher the WIF conﬂict, the lower the work satisfaction. Grandey,
Cordeiro and Crouter (2005) found that higher levels of WIF were related with lower job
satisfaction. They found that WIF conﬂict predicted a change in work satisfaction from time
one to time two in a longitudinal study. There are several other studies that also show that
job satisfaction and WIF conﬂict are intertwined and related to each other (e.g. Greenhaus,
Parasuraman, & Collins, 2001; Huang, Hammer, Neal & Perrin, 2004; Adams, King, & King,
1996)

Job satisfaction has also been linked to emotion variables. The emotional labor
literature often considers job satisfaction in its studies and ﬁnds that increased surface acting
is related to decreased job satisfaction (e.g. Liu, Prati, Perrewe & Ferris, 2008). Other
affective variables such as mood and affectivity have also been linked to job satisfaction.

Payne and Morrison (2002) examined the role of negative affectivity in job variables and

39

 

 

 

found that it is related to decreased job satisfaction and commitment. Judge and Ilies (2004)
examined the effects of mood states on job satisfaction and also found a relationship between
the two, although this relationship was bidirectional and decreased over time. Affective
events theory (Weiss & Cropanzano, 1996) also consistently links negative emotions
experienced in regard to work events to decreased job satisfaction (e.g. Grandey, Tam &
Brauburger, 2002). These types of studies show that greater negatively emotionally valenced
events impact an individual’s satisfaction with his/herjob.

Both the WP C literatures and the emotion literatures show links to job satisfaction.
Therefore, a measure that assesses emotional responses to WF C is likely to be related to job
satisfaction as well. WIF is considered especially important with regard to job satisfaction.
Although there is research linking general negative affect to job satisfaction, no studies have
yet been done to assess individual emotional links. Therefore, it is impossible to predict if
one type of emotion is more likely to be linked to job satisfaction than others. Therefore, the
prediction for this study is that all the negative WIF emotions will be negatively related to job
satisfaction.

Hypothesis 9: Individuals who experience more WIF related negative emotions will
experience decreased job satisfaction.

Life Satisfaction.

Life satisfaction has also been linked to a variety of organizationally and personally
important outcomes. Individuals with decreased life satisfaction are more likely to present
with physical and mental health problems, such as increased somatic complaints and
increased depression (Goldberg & Harrow, 2005; Riddick, 1985). Although the association ‘
is weaker than for job satisfaction, individuals with decreased life satisfaction are also more

likely to be absent or turnover from organizations (Schoenfelt & Battista, 2004).

40

 

 

 

In the WFC literature, higher levels of WFC have been linked to decreased life
satisfaction (Ford et al, 2007). Because life satisfaction measures generally assess
satisfaction with life in general, which includes both work and family domains, both WIF and
FIW have been found to relate to decreased life satisfaction. Therefore, both directions
should be considered with regard to life satisfaction.

Emotional variables have also been linked to decreased life satisfaction. Negative
emotional events from all realms of life, including work (e.g. Blachbum, Horowitz, Edington
& Klos, 1986) and personal (Lansrey-Meyer, Gerard, & Guzell, 2005), contribute to
decreased satisfaction with life. This relationship may not hold for all emotions, however. In
the single study that directly examines guilt as it relates to WFC, Livingston and Judge
(2008) found that individuals who experienced more guilt related to WFC also reported more
life satisfaction. Livingston and Judge explain this seemingly contradictory ﬁnding through
the potential regulatory power of guilt. Guilt may be felt as a means to redirect the attention
of the individual at the realm of life inspiring that guilt to focus more personal attention on
this realm. This may in turn regulate the individual’s behavior such that he/she attempts to
rectify the action that caused the guilt and thus repair the potential damage from that action.
This could lead, over time, to a regulatory system that uses guilt as a cue to the individual to
focus on and maintain good relationships in important areas of his/her life. Due to this
ﬁnding, WFC guilt is not hypothesized to relate to life satisfaction.

Because the other emotion variables being studied in the project have not been
explicitly studied with regard to WFC, the hypothesized relationships are such that they align
with the other negative emotion research in the area of life satisfaction.

Hypothesis 10: Individuals who experience more WF C related worry, frustration or

anger (regardless of direction) will experience decreased life satisfaction.

41

Methods
Sample

A sample of adult, non-student workers was solicited for their participation.
Following the lead of other researchers in the work-family literature (see Casper, Eby,
Bordeaux, Lockwood & Lambert, 2007 for a review of methods in work-family research),
only individuals who work at least part-time and have a spouse, a child, or both were
included in the study. This is due to the wording of many of the measurement items that may
be less relevant in measuring work-life conﬂict of working adults without spouses or
children. The survey was administered online and was conﬁdential. Participant emails and
names were obtained so that they could be compensated with a $10 gift certiﬁcate for their
time. See Appendix A for a copy of the online consent form.

Sample procurement occurred in three waves. For the ﬁrst wave, parents of children
who attended a local school were solicited. A small advertisement was placed in the school
newsletter that was distributed to all parents of the school. Interested parents were instructed
to go to the online survey website and ﬁll out the survey. See Appendix B for this
recruitment advertisement.

For the second wave, various email list serves that were likely to target working
adults with a Spouse, a child or both were employed. An initial email solicitation was sent
out the a list serve that is comprised by individuals at a large Midwestern university who
have stated prior interest in keeping abreast of news having to do with work and family
issues. A solicitation email was sent out with the survey website address and information
about compensation and study participation requirements. Some participants offered to cross
post this solicitation email on other list serves they belonged to, which resulted in

participation from individuals beyond the original list serve. This wave of participant

42

 

 

.,- it, ,a'. .

Ls.» \l.'\'r

:«s perm. part.

 

procurement was quite successful and resulted in 256 responses. See Appendix C for the
recruitment email.

Upon reviewing the surveys completed to this point, it became apparent that many
people either did not qualify to complete the survey (e.g. they did not work or did not have a
spouse and/or a child) or they did not ﬁnish it. In addition, in order to test the hypothesized
gender effects, it would be necessary to have a sample that was at least one-third men. Up to
this point, participants had been a majority female.

The third wave of participant solicitation speciﬁcally targeted male participants. An
additional email solicitation was sent to an alumni list serve with the speciﬁc request for male
participants. This targeted solicitation proved effective and garnered an additional 115
participants. See Appendix D for this recruitment email.

Participants

A total of 385 participants were qualiﬁed to ﬁll out the survey (e.g. they were
married, parents or both and also worked at least 20 hours per week). These individuals all
began the study, but due to drop out rate, usable data decreased as the survey progressed.

The ﬁrst complete measure assessed WFC negative emotions. Because the measure is
utilized extensively in factor development and subsequent analyses, individuals who did not
progress far enough to complete this measure were deleted from the data set. This resulted in
a total of 347 participants. Please see Table 3 for complete. demographics by survey

recruitment wave.

43

 

 

Table 3
Descriptive Information for Demographic Variables

 

 

Demographic Variables N Minimum Maximum Mean SD
Gender 333 0 I .4 I .49
Age 333 22 64 39.25 9.33
Education 334 I 6 4.96 l .16
Race 332 0 l .92 .27
Work hours 334 l 4 2.64 .87
Work schedule 329 0 I .14 .35
Marital Status 334 0 l .07 .26
Parental Status 328 0 l .85 .35
Children at home 333 0 5 1.45 1.09
Age of child at home 256 0 30 8.45 6.45

 

Notes: Gender is coded 0=women, l=men; Education is coded l= some high school, 2 =
high school diploma, 3 = some college, 4 = college diploma, 5: some graduate school,
6=graduate degree; Race is coded 0=Not-White, l= White; Work schedule is coded
0=day shift, l= other shifts; Marital status is coded 0=married, l= not married; Parental
status is coded 0= has children, I: does not have children; Age of children at home is
averaged across children for individuals with multiple children at home

 

Measures

The measures were presented in the order they appear below.

Work-family Conﬂict Negative Affect: A new scale was created for this study of WF C
related negative affect. Each directionality (work interference with family and family
interference with work) and each type (time-based, strain-based and behavior-based) are
crossed to result in six prompts. These prompts encouraged participants to think of times

when work-family conﬂict occurred in the speciﬁc context of these directionalities and types.

44

Then the participant is asked to respond to a series of 8 questions that assess four speciﬁc
types of negative affect: guilt, anger, frustration and worry. In order to assure a broad
construct coverage, this measure was initially designed to assess WFC emotions targeted at
both the self and others. Part of this was to ensure that the traditionally externally referenced
emotions (anger and frustration) had adequate construct coverage. The extant literature
suggested that these emotions could be and are felt in reference to the self as well as other
people, but in order to be thorough, I included both reference points. Each emotion was
described as either self-based (e.g. “I feel guilty for letting down myself, being unable to ‘do
it all’”) or other-based (e.g. “I feel guilty for letting down my coworkers, supervisor or
organization, being unable to ‘pull my weight’”). This resulted in a total of 48 targeted
questions. Responses to each item are anchored on a 5-point, Likert-type attitude scale
ranging from 1 Never to 5 Always or Almost Always. This scale is reproduced in Appendix E.
Speciﬁc scale development issues will be discussed below in the results section.
Work-Family Conﬂict: Work family conﬂict was measured with a 12 item scale
developed by Carlson, Kacmar and Williams (2000). The full scale was utilized. This scale
consists of three items measuring time-based work interference with family (e.g. “My work
keeps me from my family activities more than I would Iike.”), three items measuring time-
based family interference with work (e.g. “I have to miss work activities due to the amount of
time I must spend on family responsibilities”), three items measuring strain-based work
interference with family (e.g. “When I get home from work I am often too frazzled to
participate in family activities/responsibilities ”), three items measuring strain-based family
interference with work (e.g. “Due to stress at home I am often preoccupied with family
matters at work”), three items measuring behavior-based work-interference with family (e.g.
“The problem solving behaviors I use in myjob are not effective in resolving problems at

home”), and three items measuring behavior-based family interfering with work (e.g. “The

45

behaviors that work for me at home do not seem to be effective at work”). Each item is
measured on a 5-point Likert-type attitude scale ranging from I strongly disagree to 5
strongly agree. Please see Appendix F for a reproduction of this measure.

Positive and Negative Aﬂectivity: Positive and negative affectivity was measured as a
control variable using the PANAS developed by Watson, Clark and Tellegan (1988). These
items are measured on a 5-point attitude scale indicating the extent that a person generally
feels the way described by a single emotion word. The 5-point scale ranges from I very
slightly or not at all to 5 extremely. Example emotion words include “Enthusiastic”, “Guilty”
and “Inspired”. Positive and negative affectivity was measured to examine intercorrelations
between affectivity and the speciﬁc WFC negative emotions. Following advice from
Spector, Zapf, Chen & Frese (2000), affectivity will not be controlled for in hypothesized
regression equations. A more thorough explanation of this will be included in the hypothesis
testing section. Please see Appendix G for a reproduction of the measure.

Career Importance: Career importance was measured through two subscales of the
Life Roles Salience scale developed by Amatea, Cross, Clark and Bobby (1986). The two
subscales that were used were the Occupation Role Reward Value subscale (ORRV), and the
Occupational Role Commitment subscale (ORC). Example items include, from the
Occupational Role Reward Value Subscale, “Having work/a career that is interesting and
exciting to me is my most important life goal”, and from the Occupational Role Commitment
Subscale, “I expect to make as many sacriﬁces as necessary in order to advance my
work/career”. These items were assessed on a 5 point Likert-type scale ranging from I
strongly disagree to 5 strongly agree. Please see Appendix H for a complete reproduction of
this scale.

Family Importance: Family importance was measured through two subscales of the

Life Roles Salience scale developed by Amatea, Cross, Clark and Bobby (1986). The Marital

46

 

and Parental Role Reward Value Subscale (MRRV and PRRV) and the Marital and Parental
Role Commitment Subscale (MRC and PRC) were utilized. The items of the Marital and
Parental subscale were modiﬁed to indicate a general “family” referent as opposed to spouse
or child speciﬁc items. Examples of items include, from Family Role Reward Value (F RRV)
Subscale, “My life would be empty if I never had a family”; and from the Family Role
Commitment (FRC) Subscale, “I expect to commit whatever time is necessary to making my
family feel loved, supported and cared for”. These items were assessed on a 5 point Likert-
type scale ranging from I strongly disagree to 5 strongly agree. Please see Appendix I for a
complete reproduction of this scale.

Although each of the family importance scales originally consisted of 5 items each,
reliability analyses for the PRC scale revealed that one item was not functioning well with
the other items of that scale. With the inclusion of this item, the alpha was on the low end of
acceptable at .63. Removal of this item enhanced the alpha considerably, to .72. In order to
maintain acceptable internal scale reliability, this item was excluded from further scale
creation and analysis. Please see Appendix I for information on which item was removed.

Job Satisfaction: Job satisfaction was assessed through ﬁve questions developed by
Brayﬁeld and Rothe (1951). An example item is “I feel enthusiastic about my job.” These
items were assessed on a 5 point Likert-type scale ranging from I strongly disagree to I
strongly agree. Please see Appendix J for a reproduction of this scale.

Life Satisfaction: Life satisfaction was assessed through 5 items from a scale
developed by Diener, Emmons, Larsen & Griffen (1985). An example item is “In most
ways, my life is close to ideal”. These items were assessed using a 5 point Likert-type scale
ranging from I - strongly disagree to 5- strongly agree. Please see Appendix K for a

reproduction of this scale.

47

Demographics: The demographics assessed included individual characteristics, job
characteristics and family characteristics. See Appendix L for a reproduction of the
demographic items. The individual characteristics included gender, age, education, race and
income. Gender was coded such that women were represented by “0” and men were

66",

represented by . Age is a continuous variable and thus did not need recoding. The
education variable increased in numerical value as years of education increased, so higher
numbers correspond to more education. As for the race variable, the sample was a large
majority White/Non-Hispanic and there was low endorsement for particular ethnic subtypes.
Due to this fact, the ethnic subgroups were collapsed and analyses were conducted using a
dichotomous White/Non-white variable (White = 0, Minority/Multiethnic = 1). The income
variable increases in numerical value as income level increased such that a higher number
corresponds to greater income.

Thejob characteristics included work hours, work schedule and job type. Work hours
were categorized as falling into hours worked outside the home per week (20-30 hours/week
= I; 31-40 hours/week = 2; 40-50 hours/week = 3; 50+ hours/week = 4). Because only
individuals who worked at least part time were included in the survey, all participants worked
outside the home for at least 20 hours per week. Work schedule assessed whether individuals
worked a standard day shift, or had different work shift arrangements (e.g. night shift,
afternoon shift, variable shifts). Because the majority of the sample worked a day shift, this
variable was dichotomized into a day/non-day shift variable (Day shift = 0, Non-day shift =
1) for further analyses. Job type was reported qualitatively, with individuals entering their
job title.

The family characteristics included marital status, parental status, number of children

at home, and age of children. Marital status was assessed with four response possibilities.

Because the majority of the sample was married and the other response possibilities were

48

infrequently endorsed, the other three categories were collapsed to create a dichotomous
Married/Non-married variable (Married == 0, Non-married = I). The age of children at home
was individually reported, but these values were averaged. Number of children at home was

a free response continuous variable.

Results
Sample Description.

Please see Table 3 for a description of the sample. This sample was selected for
some work and family characteristics so it is not surprising that a high proportion of the
sample is married or a parent. Also, the sample was selected for work status, and the
majority of the sample worked over 30 hours a week, with a minimum of 20 hours of work
outside the home per week. The sample was also unusually well educated, had high yearly
income, and was mostly White. This suggests a sample of mostly professional, Caucasian,
middle class individuals or relatively high socioeconomic status. Although his type of
sample is not unusual for work-family conﬂict literature (see Eby et al, 2005), it limits
generalizability to individuals in lower socioeconomic conditions and to non-White minority
workers.

Measure Development- The WF C Emotions Scale

Self versus other reference: This measure assessed 6 types of WF C, four emotions
nested within those type, and two referents (self and other), which resulted in 48 speciﬁc
emotion-WFC relationships. The ﬁrst step in the data analysis process was to determine if
both the self and the other referents of the WFC emotion items should be retained for further
analysis.

An exploratory factor analysis of all 48 items together was conducted as a ﬁrst step

toward considering whether to retain both types of referents. Using a maximum likelihood

49

estimation and Varimax rotation, a nine factor structure emerged. Please see Table 4 for the
rotated factor matrix. Closer examination of the matrix revealed some trends in the data that
might suggest that the self and other referenced items differ on a dimensional level. The self-
referenced items’ are consistent in their wording such that individuals report that they
experience the emotion with regard to their own perceived failures or weaknesses. The other
referenced items, on the other hand, differ in how the other is referenced. For the
internalizing emotions, individuals responded to each emotion with regard to feelings of
“letting others down” while for the externalizing emotions individuals responded to each
emotion with regard to feelings of anger or frustration at another person. This distinction
could be important and is likely contributing to the factor structure. Because the self-
referenced items have a rhetorical consistency that is not mirrored by the other referenced
items, the will be retained for the bulk of the further analyses.

The distinction between self and other referenced emotions may still be important, however,
and warrants further investigation. An exploratory factor analysis again using maximum
likelihood estimation and Varimax rotation of just the other referenced items reveals an
interesting factor structure. See Table 5 for the rotated factor matrix. For the other
referenced items, it appears that all of the internalizing emotions factored on a single
dimension. This suggests that individuals experience these emotions in regard to “letting
others down” in similar ways. The factor structure for the externalizing emotions, however,
is quite different. Although not completely free of cross loadings, the factor structure
suggests that individuals tend to experience WIF externalizing emotions in similar ways and
FIW externalizing emotions in similar ways. This may suggest that the target of the
externalizing emotions (e.g. a coworker/boss or a family member) is important in the

experience of externalizing emotions.

50

 

Table 4

Varimax Rotated Maximum Likelihood Factor Structure for All 48 WF C Negative Affect

 

 

 

Items
Factor

I 2 3 4 5 6 7 8 9
FIW time Worry Other .69
FIW time Guilt Other .68
F IW stress Worry Other .60 .42
FIW time Worry Self .59 .31
WIF time Worry Other .58
FIW time Guilt Self .53 .33
WIF time Won'y Self .53
WIF stress Worry Self .52 .48 .32
WIF time Guilt Other .51 .32
WIF stress Worry Other .51 .35 .37
F IW stress Guilt Other .51 .31
WIF time Guilt Self .44 .31
WIF behavior Guilt Other .79
WIF behavior Frust Self .77
WIF behavior Anger Self .77
WIF behavior Guilt Self .74
WIF behavior Worry Self .32 .74
WIF behavior Worry Other .40 .63 .
FIW behavior Frustration .82
Self
FIW behavior Worry Self .36 .68
FIW behavior Guilt Self .66
F [W behavior Anger Self .66
FIW behavior Guilt Other .63
F1 W behavior Worry Other .43 .62
WIF time Anger Other .77
WIF stress Anger Other .75 .33
WIF time Frustration Other .73
WIF stress Frustration .68 .30

Other

 

51

 

Table 4 (continued)

 

WIF behavior Anger Other .32
WIF behavior Frustration .35
Other

FIW time Anger Self

FIW time Frust Self .36

WIF time Anger Self

WIF time Frust Self

WIF stress Guilt Self .38

WIF stress Anger Self

WIF stress Frust Self .30
WIF stress Guilt Other .48

FIW time Frustration Other

FIW time Anger Other

FIW behavior Anger Other

FIW beh Frustration Other

F IW stress Anger Self

F IW stress Frust Self

FIW stress Worry Self .42 .35
FIW stress Guilt Self .32

FIW stress Frust Other

F IW stress Anger Other

N: 347, loadings under .30 were suppressed

.31

52

.58
.54

Factor
5

.77
.71
.70
.65

.32

.30

.22
.67
.64
.61
.55

.87
.83
.52
.46

.40
.44

.63
.53
.53
.48

.32

.85
.66

 

Table 5

Varimax Rotated Maximum Likelihood Factor Structure for 24 Other-Referenced

WF C Negative Affect Items

 

 

Factor
I 2
Fl W stress Won'y Other .81
WIF stress Worry Other .73
F IW time Worry Other .70
WIF behavior Worry Other .69 .34
FIW behavior Worry Other .64
WIF time Worry Other .64
FIW stress Guilt Other .62
FIW time Guilt Other .61
WIF stress Guilt Other .60
W1 F behavior Guilt Other .51 .42
WIF time Guilt Other .49
F IW behavior Guilt Other .48
WIF time Anger Other .83
WIF time Frustration Other .81
WIF stress Anger Other .70
WIF stress Frustration Other .63
FIW time Frustration Other .85
FIW time Anger Other .80
FIW behavior Anger Other .50 .35
FIW behavior Frustration Other .44 .32
WIF behavior Anger Other .33 .82
WIF behavior Frustration Other .30 .78
FIW stress Anger Other .37 .80
ﬂW stress Frustration Other .37 .78

 

N: 347, Loadingﬂder .30 suppressed

 

53

Although the other referenced items are indeed interesting, the remainder of this
paper will focus on the 24 self referenced emotions where the target is all the same (the self).
Supplementary analyses will be conducted with the other-referenced scales, however, and
will be brieﬂy discussed in later sections.

Conﬁrmatory Factor Analyses: There is a need to see if the self-referenced items
conform to an a priori hypothesized factor structure. As presented by the hypothesized
model, an eight factor solution should emerge, with each direction of work family conﬂict
(WIF or FIW) and each emotion (guilt, worry, anger or frustration) loading as a distinct
factor. It is possible, however, that a variety of other factors structures will be revealed.

Due to the fact that a variety of factor structures are theoretically defensible, several
C FAs were conducted. These models are summarized below and ﬁt indices are reported in
Table 6. For all of the models, the factors were allowed to correlate and means were
estimated for missing data.

Single Factor Model: This model tested the data against a factor structure where all
of the WP C emotion items load on a single factor of “general negative WFC affect”. This
factor consists of all items without distinction for WFC type (time, strain, behavior) direction
(FIW and W1 F) or emotion type (guilt, worry, anger, frustration).

Two Factor WIF and F I W Model: This model tested the data against a factor
structure where there are two main factors “WIF negative emotions” and “F [W negative
emotions”. These factors consist of all items that fall within that particular directionality of
WF C, with no distinction for WFC dimensions or speciﬁc emotion.

Three Factor WF C Dimensions model: This model tested the data against a factor
structure where there are three main factors of WF C types, “WF C Stress Emotions”, “WF C

Time Emotions” and “WFC Behavior Emotions”. These factors consist of all items that fall

54

within each particular dimensionality, with no distinction for direction of WFC or speciﬁc

 

 

 

emotion.
Table 6
C onﬁrmatory Factor Analyses

Model df Chi-Square CF 1 RM SEA

Single Factor Model 253 2894.3 .57 .17
Two Factor WIF and FIW 251 2550.9 .63 .16
Model
Six Factor WFC Type and 237 1 108.0 .86 .10
Direction Model
Two Factor Intemalizing 251 2627.8 .62 .16
and Extemalizing Emotions
Four Factor Emotion 246 2605.8 .62 .16
Model
Three Factor WFC Type 252 1835.06 .74 .13
Model
Four Factor WFC Direction 246 3125.24 .64 .19
and

Intemalizing/Extemalizing
Emotions Model

Six Factor WFC Types and 237 1747.57 .79 .14
Intemalizing/ Extemalizing
Emotions Model

Eight Factor Model Two 224 2126.69 .68 .18
Directions of WFC with

Four Emotions

N=340

 

 

Two Factor Intemalizing and Extemalizing Emotions Model: This model tested the
data against a factor structure where the main factors are “WFC Intemalizing Emotions” and

“WFC Extemalizing Emotions”. These are deﬁned in line with the general emotion literature

55

where guilt and worry are generally considered internalizing emotions and anger and
frustration are generally considered externalizing emotions. These factors consist of all items
that fall within these two categories of emotions, with no distinctions for direction or type of
WF C.

Four Factor Emotion Model: This model tested the data against a factor structure
where there are four factors that correspond to speciﬁc emotions, “WFC Anger”, “WFC
Guilt”, “WFC Worry” and “WFC Frustration”. These factors consist of all items that fall
within a particular emotion referent, with no distinction for types or directionality of WF C .

Six Factor Direction and Type WF C Model: This model tested the data against a
factor structure where there are four factors that correspond to the speciﬁc directionality and
type of WFC, “WIF Time Emotions”, “WIF Stress Emotions”, “F [W Time Emotions”, “FIW
Stress Emotions”, “WIF Behavior Emotions”, and “FIW Behavior Emotions”. These factors
consist of the items that fall within that directionality and type of WFC, with no distinction
for speciﬁc emotions.

Four Factor WF C Direction and Internalizing/Externalizing Emotions: This model
tested the data against a factor structure where factors consist of directionality of WFC and
category of emotions, “WIF Intemalizing Emotions”, “WIF Extemalizing Emotions”, “F 1W
Intemalizing Emotions”, “FIW Extemalizing Emotions”. These factors consist of the items
that fall within the WF C directionality and within the Intemalizing/Extemalizing categories
of emotions, with no distinction for WFC types.

Six Factor WF C Types and Internalizing/Externalizing Emotions: This model tested
the data against a factor structure where factors consist of type of WF C and category of
emotions, “WFC Time Intemalizing Emotions”, “WFC Strain Extemalizing Emotions”,

“WFC Time Intemalizing Emotions”, “WFC Extemalizing Strain Emotions”. These factors

56

consist of the items that fall within the WF C type and within the Intemalizing/Extemalizing
categories of emotions, with no distinction for WFC directionality.

Eight Factor WIF and FIW Directions with Nested Four Discrete Emotions: This
model tested the data against a factor structure where there are eight factors that correspond
to speciﬁc emotions within each of the two direction of WF C, “WIF Anger”, “WIF Guilt”,
“WIF Worry”, “WIF Frustration”, “F [W Anger”, “FIW Guilt”, “FIW Worry” and “FIW
Frustration”. This factor structure corresponds to the hypothesized structure. These factors
consist of all items that fall within a particular emotion referent, with no distinction for types
of WFC.

Exploratory Factor Analysis

Referring to Table 6, it is apparent that there is poor ﬁt of the data to all of these
factor structures. Although the data does not align strongly with any of the theoretical factor
structures, this result is not altogether unsurprising due to the variety of factor possibilities,
including direction and type of WFC, and categories and types of emotions. In addition,
these items are likely correlated with each other across factors, thus creating some model ﬁt
problems. In order to more fully understand how the WP C emotion items function in relation
to each other, an exploratory factor analysis was conducted. This EF A used the maximum
likelihood extraction method with Varimax rotation. Missing data was excluded Iistwise.
Please see Table 7 for this factor matrix with values under .30 suppressed.

From this EFA, a model with six factors with eigenvalues over 1 emerged. This
structure explained 68.09% of the total variance. The selection of a six factor structure was
also supported by a scree plot (see Figure 3), where the inﬂection point of the graph occurs

after the sixth factor.

57

Figure 3: Scree Plot of Rotated Factor Matrix for the 24 Self-Referenced
Emotion Items

 

Scree Plot

Eigenvalue

 

 

1 2 3 4 5 6 7 8 91011121314151617181920

Factor Number

 

 

T ime-based WF C and Emotions. For the time type of WFC, it appears that the
emotions WFC items factor on an intemaI/extemal emotion distinction rather than on conﬂict
direction. When individuals experience time conﬂict, the direction of the conﬂict does not

seem to matter to emotional responses, with items from both WIF and FIW emotions loading

together. For the time type of WFC emotions, the type of emotion seems to be more

\“Iportant, with the emotions worry and guilt factoring together and the emotions ﬁ'ustration
and anger factoring together. This distinction aligns with other emotion research in which
e"10tions are organized along “intemalizing” and “extemalizing” dimensions (Garside &

Kll'l‘l'res-Dougan, 2004). For the time type of WFC emotions, it appears that the factors of

58

internalizing (guilt and worry) and externalizing (frustration and anger) tend to emerge. This
would suggest that individuals tend to experience either internalizing emotions or

externalizing emotions in response to time-based conﬂict.
The factors for strain- and behavior- based conﬂict items look different than that for

time-based conﬂict. For both strain- and behavior-based conﬂict, all the emotions tend to

 

Table 7
Factor Structure for 24 Self-Referenced WF C Negative Aﬁ’ect Items
Factor

1 2 3 4 5 6
WIF time Anger Self .796 .324
WIF time Frust Self _ .775
F IW time Frust Self .716 g .403
F IW time Anger Self .715 .333 .361
WIF beh Anger Self .796
WIF beh Frust Self .791 .302
WIF beh Worry Self .688 .303
WIF beh Guilt Self .678
WIF stress Guilt Self .714
WIF stress Worry Self .664 .305 .324
WIF stress Frust Self .637 .301
WIF stress Anger Self .328 .595 .380
WIF time Worry Self .432 .397
FIW beh Frust Self .862
FIW beh Guilt Self .679
FIW beh Worry Self .677
FIW beh Anger Self .673
F l W stress Anger Self .773
FIW stress Frust Self .638
F1 W stress Guilt Self .557
FIW stress Worry Self .358 .488 .311
PC W time Guilt Self .691
Y\ W time Worry Self .433
WI F time Guilt Self .419 .428

 

 

Variance Explained 12.5% 12.72% 12.42% 12.35% 10.33% 7.52%
“:35”

59

factor together with distinctions made along direction (F IW and WIF) and type (strain and
behavior). This would suggest that when individuals experience the strain or behavior based
type of conﬂict, they tend to experience a general negative affectivity, with roughly equal
tendencies to experience guilt, worry, anger or frustration in response to these types of
conﬂict.

Final Factor Structure: Although this data did not conform to a hypothesized factor
structure overall, each dimension resulted in a factor structure that is defensible and coherent.
For the emotion WFC items, the exploratory factor analysis resulted in 6 distinct WFC
emotion scales. For the time type, direction of WFC is unimportant while type of emotion is,
resulting in two scales deemed I.) WFC time based internalizing emotions and 2.) WFC time
based externalizing emotions. For the strain and behavior types, type of emotion is less
important while direction of conﬂict is paramount. This results in four additional scales
deemed 3) WIF strain all emotions, 4) WIF behavior based all emotions, 5) F [W strain based
all emotions, and 6) FIW behavior based all emotions. These scales will be retained and used
for the remainder of the analyses.

Modified Hypotheses

As mentioned above, the factor structure that emerged from the WFC emotion items

was different from the a priori assumed structure. In light of this fact, some of the original
hypotheses needed to be altered slightly to accommodate the current structure. The old and
new hypotheses are summarized in Table 8. The modiﬁed model is pictured in Figure 4.

The most substantive changes occurred in the hypotheses that referenced speciﬁc
emotions. Because the factor structure of the WFC emotions scale did not distinguish

between speciﬁc emotions, some conceptual work was necessary to adequately change the

hJ’POtheses to ﬁt the structure. It is now necessary to address these changes systematically.

60

 

Table 8

Original hypotheses and modified hypotheses

 

Original Hypotheses

Modiﬁed Hypotheses

 

Hypothesis Ia: Individuals
who experience more FIW
(stress, time and behavior-
based) will experience more
FIW guilt.

Hypothesis 1 b. Individuals
who experience more WIF
(stress, time and behavior-
based) will experience more
WIF guilt.

Hypothesis Za: Individuals
who experience more F [W
(stress, time and behavior-
based) will experience more
Fl W worry.

Hypothesis 2b: Individuals
Who experience more WIF
(Stress, time and behavior--
based) will experience more
WIF worry.

Hypothesis 3a: Individuals
Who experience more FIW
(stress, time and behavior-

based) will experience more
Fl W anger.

\

Hypothesis 2b: Individuals who experience more FIW
time will experience more WFC time internalizing
emotions

Hypothesis 3b: Individuals who experience more FIW
strain will experience more FIW strain based negative
emotions.

Hypothesis 4b: Individuals who experience more FIW
behavior will experience more F [W behavior based
negative emotions.

Hypothesis 1 b: Individuals who experience more WIF
time will experience more WFC time internalizing
emotions.

Hypothesis 3a: Individuals who experience more WIF
strain will experience more WIF strain based negative
emotions

Hypothesis 4a: Individuals who experience more WIF
behavior will experience more WIF behavior based
negative emotions

See H2b, H3b, H4b

See Hlb, H3a, H4a

Hypothesis Za: Individuals who experience more FIW
time will experience more WF C externalizing emotions
See H3b, H4b

61

 

Table 8 (continued)
Original Hypotheses

Modified Hypotheses

 

Hypothesis 3b: Individuals
who experience more WIF
(stress, time and behavior-
based) will experience more
WIF anger.

Hypothesis 4a: Individuals
who experience more F IW
(stress, time and behavior-

based) will experience more
FIW frustration.

Hypothesis 4b: Individual
who experience more WIF
(stress, time and behavior-

based) will experience more
WIF ﬁ'ustration.

Hypothesis 5: The
relationship between FIW
and all F IW emotions will be
moderated by work
importance, such that
individuals who place greater
importance on their work will
be more likely to experience
the negative FIW emotions

Hypothesis 1 a: Individuals who experience more WIF
time will experience more WF C time externalizing
emotions.

See H3a, H4a

See New Hypotheses 2a, 3b, 4b

See New Hypotheses 1a, 3a, 4a

Hypothesis 5a: The relationship between F IW time and
WF C time externalizing emotions will be moderated by
work importance, such that individuals who place
greater importance on work will be more likely to
experience WFC time externalizing emotions

Hypothesis 5b: The relationship between FIW time and
WFC time internalizing emotions will be moderated by
work importance, such that individuals who place
greater importance on work will be more likely to
experience WFC time internalizing emotions

Hypothesis 5c: The relationship between F IW strain and
FIW strain negative emotions will be moderated by work
importance, such that individuals who place greater
importance on their work will be more likely to
experience negative FIW strain based emotions.

Hypothesis 5d: The relationship between FIW behavior
and FIW behavior negative emotions will be moderated
by work importance, such that individuals who place
greater importance on their work will be more likely to
experience negative FIW behavior based emotions.

62

 

Table 8 (continued)
Original Hypotheses
Hypothesis 6: The
relationship between WIF
and WIF emotions will be
moderated by family
importance, such that
individuals who place greater
importance on their family
will be more likely to
experience the negative WIF
emotions.

Hypothesis 7a: Gender will
moderate the relationship

. between FIW and FIW worry
such that women will be
more likely to experience
FIW worry then men.

Hypothesis 7b: Gender will

moderate the relationship
between WIF and WIF worry
such that women will be
more likely to experience
W] F worry than men.

Hypothesis 8: Gender will
moderate the relationship
between FIW and FIW
frustration such that men will
be more likely to experience
FlW frustration than women.

\

 

Modiﬁed Hypotheses

Hypothesis 6a: The relationship between WIF time and
WP C time externalizing emotions will be moderated by
family importance, such that individuals who place
greater importance on family will be more likely to
experience WF C time externalizing emotions

Hypothesis 6b: The relationship between WIF time and
WP C time internalizing emotions will be moderated by
family importance, such that individuals who place
greater importance on family will be more likely to
experience WFC time internalizing emotions

Hypothesis 6c: The relationship between WIF strain and
WIF strain negative emotions will be moderated by
family importance, such that individuals who place
greater importance on their family will be more likely to
experience negative WIF strain based emotions.

Hypothesis 6d: The relationship between WIF behavior
and WIF behavior negative emotions will be moderated
by family importance, such that individuals who place
greater importance on their family will be more likely to
experience negative WIF behavior based emotions.

Hypothesis 7a: Gender will moderate the relationship
between FIW time and WFC time internalizing emotions
such that women will be more likely to experience WFC
time internalizing emotions in relation to FIW time than
men.

Hypothesis 7b: Gender will moderate the relationship
between WIF time and WFC time internalizing emotions
such that women will be more likely to experience WFC
time internalizing emotions in relation to WIF time than
men.

Hypothesis 8: Gender will moderate the relationship
between F [W time and WFC externalizing emotions
such that men will be more likely to experience WF C
externalizing emotions in relation to FIW time than
women.

63

 

Table 8 (continued

 

Original Hypotheses

Modiﬁed Hypotheses

 

Hypothesis 9: Individuals
who experience more WIF
related negative emotions
will experience decreased job
satisfaction.

Hypothesis [0: Individuals
who experience more WFC
related worry, frustration or
anger (regardless of
direction) will experience
decreased life satisfaction.

Hypothesis 9a: Individuals who experience more WFC
time externalizing emotions will experience decreased
job satisfaction

Hypothesis 9b: Individuals who experience more WFC
time internalizing emotions will experience decreased
job satisfaction

Hypothesis 9c: Individuals who experience more WIF
strain negative emotions will experience decreased job
satisfaction

Hypothesis 9d: Individuals who experience more WIF
behavior negative emotions will experience decreased
job satisfaction

Hypothesis 10a: Individuals who experience more WF C
time externalizing emotions will experience decreased
life satisfaction

Hypothesis 10b: Individuals who experience more WF C
time internalizing emotions will experience decreased
life satisfaction

Hypothesis 10c: Individuals who experience more WIF
strain negative emotions will experienced decreased life
satisfaction

Hypothesis 10d: Individuals who experience more FIW
strain negative emotions will experienced decreased life
satisfaction

Hypothesis 10e: Individuals who experience more WIF
behavior negative emotions will experienced decreased
life satisfaction

Hypothesis 10f: Individuals who experience more F 1W

behavior negative emotions will experienced decreased
life satisfaction

64

Figure 4: Proposed Model of Modiﬁed Hypotheses

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Work
Importance Gender
H5a H5bH5cH5d ”73 H7b H8
WFC Time
H I a Extemal izing
WIF Emotions
Time 1B\ / 2a
/><\l
“W l ,/ “lb WFC Time
Time I ’/ Intemalizing
H2b Emotions
H3a WIF strain All
| S‘fleilin i r Emotions
l 81:11“: ! H3b _ FIW strain All
a n r Emotions
H
w": “43; WIF Behavior
Behavior All Emotions
FIW- i H4: FIW Behavior
Behavror ' All Emotions
I6all6b H6c H6d

 

 

 

 

 

 

Family
Importance

 

 

65

Job
Satisfaction

 
 

 
 

Life
Satisfaction

Emotion specific hypotheses and their relation to WF C: Originally, hypotheses 1-4
addressed predictions about speciﬁc emotions and their relation to WFC as distinguished by
particular directions of WFC. These hypotheses were all parallel, suggesting that individuals
who experience more WIF or FIW conﬂict will also experience more WIF or FIW negative
emotions. The previously stated theoretical rationalization for these relationships at the
individual emotion level can still hold when the emotions are aggregated to a general
negative affect scale as is the case with WIF and FIW behavior- and strain-based conﬂict
emotions. Because the WF C emotions scale emerged along WFC direction and type, the
hypotheses have been altered to more directly assess this. For FIW and WIF strain based
emotions, they are hypothesized to relate to F1 W and WIF strain based conﬂict, respectively.
For F [W and WIF behavior based emotions, they are hypothesized to related to FIW and WIF
behavior based conﬂict, respectively.

This rationalization also holds for the WFC time emotions scales that distinguish
between internalizing and externalizing emotions. As far as speciﬁc emotions are concerned,
the hypotheses relevant to the speciﬁc internalizing emotions (guilt and worry) and
externalizing emotions (anger and frustration) are parallel so collapsing these hypotheses
across speciﬁc emotions isjustiﬁed. The slight disconnect comes due to the fact that WFC
time emotions are not distinguishable by direction. For this case, it becomes necessary to
predict how WFC time internalizing and externalizing emotions relate to the speciﬁc
directions of conﬂict, FIW time based and WIF time based. Because WFC time internalizing
e"'IOtions and WFC time externalizing emotions include references that are both F IW and
WIF, it is hypothesized that individuals who experience WIF time based conﬂict or FIW time
based conﬂict will also experience more WFC time based internalizing and externalizing

emotions.

66

Work and Family Importance: The moderation hypotheses of work and family
importance originally did not distinguish between speciﬁc emotions. Therefore, there is no
need to modify them for the reason of accounting for the general emotion factors revealed by
the WFC emotion scale factor structure. There is a need, however, to modify them Slightly to
ﬁt with the differential directionality of the time-, strain- and behavior- based types. The
hypotheses can stand as is for the strain- and behavior-based constructs with the distinction of
direction of WFC added.

The time type of the WFC emotions scale functions differently such that direction of
WF C is collapsed within internalizing and externalizing subscales. The original hypotheses
regarding work and family importance predicted differential relationships based on direction.
This was that work importance would only affect emotion/conﬂict relationships in the FIW
direction while family importance would only affect emotion/conﬂict relationships in the
WIF direction. Because the WFC time internalizing and externalizing subscales do not
distinguish direction, this creates problems in modifying this hypothesis. The internalizing
and externalizing dimensions include reference to both FIW and WIF. Because of this, it is

likely that the negative emotions elicited by WFC time conﬂict could relate to FIW and WIF
conﬂict similarly (as is argued above). Therefore, the modiﬁed hypotheses would retain the
differential hypotheses for type of conﬂict, but not for the emotional responses to that
conﬂict.

Gender Moderation Hypotheses: Modifying the gender relation hypotheses pose
SOme potential theoretical problems. The original hypothesis 7a and 7b propose that women

would more likely to experience WIF and FIW related worry than men. Because worry is in
“0 way distinguished in the strain- and behavior-based types of the WP C emotions scale, this
hypothesis cannot be tested for those directions. For the WP C time emotions subscales, the

emOtions are divided into internalizing and externalizing. Worry falls within the

67

lfl'

~1

4

Jrl

internalizing scale along with guilt. The inclusion of guilt with the worry items in the WFC
time internalizing scale may alter any moderation hypotheses. Although guilt was originally
not thought be a gender related moderating variable, some literature does suggest that guilt
might be experienced more frequently by women than men (Marecek and Ballou, 1981) in a
similar manner as worry. Because of this literature, it may be reasonable to assume that the
WFC time internalizing emotions may be experienced more frequently by women than men
in response to WIF time and F IW time conﬂict.

The second gender moderation hypothesis was the original hypothesis 8. This
hypothesis made the differential hypothesis that men would experience more F I W frustration
in response to FIW conﬂict than women. For the strain- and behavior-based, it is impossible
to assess individual emotions in any way, so this moderation hypothesis cannot be tested.
For the WFC time emotions subscales, the emotions are divided into internalizing and
externalizing emotions with frustration falling in the externalizing subgroup along with
anger. Originally, there was no differential gender hypothesis for FIW anger experiences.
Some literature suggests, however, that men may experience all the externalizing type
emotions more frequently than women (Garside & Klimes-Dougan, 2004). For this reason,
the modiﬁed hypothesis will reﬂect that WFC externalizing emotions may be experienced
more Oﬂen by men than women in response to FIW conﬂict.

Job and Life Satisfaction: The job and life satisfaction original hypotheses did not
differentiate between speciﬁc emotions so they can stay largely the same. The only
mOdiﬁcation accommodates the factor structure such that the types of WFC are speciﬁed.
Desoriptive Statistics

Correlations: The correlations between all the variables in this study are presented in
Table 9. Some of the variables are highly intercorrelated which may cause some difﬁculties

l" Subsequent regression analyses. In particular, all of the WFC negative emotion scales are

68

signiﬁcantly correlated with each other, some as high as .66. In addition, all the WP C
negative emotion scales are correlated signiﬁcantly with the traditional WFC scales, though
not at such high levels as to indicate a lack of meaningful differences. Negative affect is also
highly correlated with the WFC emotion scales, and this issue will be addressed more
thoroughly in the section below. These intercorrelations may result in multicolinearity
among the variables.

In addition, the means of certain variables reveal that there may be a ceiling effect for
certain scales. The means for the job commitment and family commitment variables are
quite high (over 4 on a scale of I to 5) and indicate that individuals tend to report that they
are highly committed to both job and family. The mean of the job satisfaction scale is also
high (4.01 on a scale of 1 to 5). This ceiling effect may cause a restriction of range that may
make the proposed hypotheses including these variables difficult to assess.

Control Variables: All of the hypotheses were tested using hierarchical linear
regressions. For each test, the person speciﬁc characteristics of gender and age were

controlled. Both were controlled for due to the fact that they tended to correlate with the
main variables of this study, the WFC emotion scales (see Table 9 for these correlations). In
addition, gender has frequently been demonstrated to relate to both emotion variables (e.g.
Garside & KIimes-Dougan, 2004; Kuppens & Van Mechelen, 2007) and WFC variables
(Ford et al, 2007). Although not as frequently or speciﬁcally studied, age can also be an
important variable to control. Past research in WF C has shown that older workers and
Younger workers may experience WFC differently (e.g. F inegold, Mohrrnan, & Spreitzer,
2002). The emotion literature also addresses age, with some researchers ﬁnding that older

adults are less emotionally reactive than younger adults (e.g. Kliegel, Jager, & Phillips, 2007)

69

Table 9
Means, standard deviation and intercorrelations of the variable assessed in this stuay

 

 

Variable N Mean SD 1 2 3 4 5 6

1. Gender 333 .4 I .49 n/a

2. Age 333 39.25 9.33 .07 n/a

3. Education 334 4.96 1.16 -.02 .01 n/a

4. Race 332 .92 .27 -.OO .02 .01 n/a

5. Marital Stat. 334 .07 .26 .04 .08 -.08 -.09 n/a

6. Parental Stat 328 .85 .35 -.10 .36" .03 -.12" .12* n/a
7. Children hm 333 1.45 1.09 -.09 .01 -.02 -.1 1* -.01 .53"
8. Age of child 256 8.45 6.45 -.05 .74" -. I 7" .I l -.08 -.08
9. Work hr 334 2.64 .87 .39“ .07 .13* .04 -.05 -.11"'
10. Work sch 329 .14 .35 .06 -.02 -.03 .03 -.01 -.04

1 1. WFC time 345 2.16 .92 -.I 1* -.18‘” -.05 .02 .04 -.08
ext. em

12. WFC time 345 2.35 .80 -.24** -.22“”" -.01 .05 .06 -.02
int. em.

13. WIF stress 345 2.41 .89 -. 14* -. l 7'” -.05 .09 .00 -.l 1*
all em.

14. FIW stress 345 2.03 .81 -.I8** -.10 -.05 -.01 .14“ -.03
all em.

15. WIF beh. 345 2.05 .91 -.10 -.05 -.04 -.02 .12“ .02
all em.

16. FIW beh. 343 1.76 .75 -.08 -.07 -.O3 -.01 .I 1" -.07
all em.

17. WIF time 340 2.98 .99 .02 .01 .06 .1 1* -.16"”" -.O6
18. FIW time 340 2.76 .97 -.O4 -.18** .16“ -.01 -.04 .04
I9. WIF stress 340 2.89 1.06 -.10 -.03 -.03 .05 -.06 -.20**
20. FIW stress 340 2.17 .93 -.O6 .00 -.04 -.07 -.04 -.00
21. WIF beh 340 2.52 .78 .l 1 -.04 -.07 -.04 -.08 -.08
22. FIW beh 340 2.49 .81 .10 -.03 -.1 1* -.05 -.04 -. 10
23. Positive 338 3.53 .62 .03 .10 .20” -.O3 -.01 .09
Affect

24. Negative 338 1.92 .62 -.08 -.21*"' -.07 .01 -.Ol -.18**
Affect

25. Work Role 336 4.18 .60 -.05 -.l l .21" .04 -.04 -.06
Value

26. Work Role 336 3.07 .67 .19" .01 .15* .05 -.04 -.l l
Comm.

27. Family 335 4.60 .54 -.18** .05 .06 .05 -.l3" .19"
Role Value

28' ”my 335 4-23 58 -.i9“ -.or .10 .02 -.07 .21“
Role Comm.

29. Job Sat. 335 4.01 .71 -.05 -.13 .20" -.Ol .02 .17"
30. Life Sat. 335 3.72 .77 .01 -.06 .22" .10 -.15"”" .06

J< .05, **p<.01; . Where applicable, scale reliabilities are reported on the diagonal
Notes: Gender is coded 0=women, I=men; Education is coded 1= some high school, 2 = high school
diploma, 3 = some college, 4 = college diploma, 5= some graduate school, 6=graduate degree; Race
is coded 0=Not-White, l= White; Work schedule is coded 0=day shift, I= other shifts; Marital status
is coded 0=married, 1= not married; Parental status is coded 0= has children, l= does not have
Children; Age of children at home is averaged across children for individuals with multiple children
Jimmie

70

lable 9 (continued)

 

 

Variable N Mean SD 7 8 9 10 1 I 12

1. Gender 333 .41 .49

2. Age 333 39.25 9.33

3. Education 334 4.96 1.16

4. Race 332 .92 .27

5. Marital Stat. 334 .07 .26

6. Parental Stat 328 .85 .35

7. Children hm 333 1.45 1.09 n/a

8. Age of child 256 8.45 6.45 .1 1 n/a

9. Work hr 334 2.64 .87 -.12 .09 n/a

10. Work sch 329 .14 .35 .02 -.03 .04 n/a

11. WFC time 345 2.16 .92 -.O8 -.15* -.04 .I4"' (.90)

ext. em

12. WFC time 345 2.35 .80 -.00 -.10 -.07 .09 .64" (.79)
int. em.

13. WIF stress 345 2.41 .89 .02 -.06 -.07 .06 .55" .66"
all em.

14. FIW stress 345 2.03 .81 -.02 -.06 -.13* .08 .52" .60"
all em.

15. WIF beh. 345 2.05 .91 -.OO -.12 -.O3 .07 .43" .51"
all em.

16. FIW beh. 343 1.76 .75 -.07 -.05 -.07 .08 .39" .45"
all em.

17. WIF time 340 2.98 .99 -.10 .05 .26" .05 .24" .34”
18. FIW time 340 2.76 .97 .03 -.25" -.08 .15" .32" .33"
I9. WIF stress 340 2.89 1.06 -. l 7" .04 .01 -.Ol .31" .42"
20. F [W stress 340 2.17 .93 -.01 -.07 .12* .10 .26" .30"
21. WIF beh 340 2.52 .78 -.10 -. 12 .02 -.02 .16" .15”
22. FIW beh 340 2.49 .81 -.12* -.11 .05 .00 .17" .14“
23. Positive 338 3.53 .62 .04 .11 .14* .05 -.15" -.15*
Affect

24. Negative 338 1.92 .62 -.l 1 -.22** -.05 .08 .46" .48"
Affect

25. Work Role 336 4.18 .60 -.05 -.OS .I 1* .01 .08 .1 1*
Value

26. Work Role 336 3.07 .67 -. I 8" -.01 .25" .04 .03 -.03
Comm.

27. Family 335 4.60 .54 .14" .11 .05 -.08 .00 .07
Role Value

28- Family 335 4-23 53 .23“ -.05 -.11' -.01 -.04 .03
Role Comm.

29. Job Sat. 335 4.01 .71 .08 .05 .12“ -.07 -.I3" -.13*
30. Life Sat. 335 3.72 .77 .07 .02 .07 .00 -.l8** -.21**

ép < .05, "”"p<.01; . Where applicable, scale reliabilities are reported on the diagonal
Notes: Gender is coded 0=women, I=men; Education is coded 1= some high school, 2 = high school
diploma, 3 = some college, 4 = college diploma, 5= some graduate school, 6=graduate degree; Race
is coded 0=Not-White, 1= White; Work schedule is coded 0=day shift, l= other shifts; Marital status
is coded 0=married, 1= not married; Parental status is coded 0= has children, l= does not have
children; Age of children at home is averaged across children for individuals with multiple children

cu.

at home

K

71

Table 9 (continued)

 

 

Variable N Mean SD l3 I4 15 I6 17 18
1 . Gender 333 .41 .49

2. Age 333 39.25 9.33

3. Education 334 4.96 1.16

4. Race 332 .92 .27

5. Marital Stat. 334 .07 .26

6. Parental Stat 328 .85 .35

7. Children hm 333 1.45 1.09

8. Age of child 256 8.45 6.45

9. Work hr 334 2.64 .87

10. Work sch 329 .14 .35

ll. WFC time 345 2.16 .92

ext. em

12. WFC time 345 2.35 .80

int. em.

13. WIF stress 345 2.41 .89 (.88)

all em.

14. FIW stress 345 2.03 .81 .67" (.87)

all em.

15. WIF beh. 345 2.05 .91 .56" .59" (.91)

all em.

16. FIW beh. 343 1.76 .75 .48" .56" .61" (.88)

all em.

17. WIF time 340 2.98 .99 .30" .17" .18" .16" (.78)

I8. FIW time 340 2.76 .97 .21" .25" .18" .23" .22" (.80)
I9. WIF stress 340 2.89 1.06 .54" .40" .36" .33" .40“ .19"
20. FIW stress 340 2.17 .93 .29" .46" .25" .30" .12“ .42"
2|. WIF beh 340 2.52 .78 .16" .17" .26" .25" .21" .15"
22.F1W beh 340 2.49 .81 .19" .18" .23" .31" .19" .18"

23. Positive 338 3.53 .62 -.20** -.16" -.12* - -.10 -.I 1

Affect .18"

24. Negative 338 1.92 .62 .57" .47" .41” .36" .15 .21"

Affect

25. Work Role 336 4.18 .60 .07 .08 .04 .03 .07 .08

Value

26. Work Role 336 3.07 .67 .01 .00 -.03 .02 .10 0.09

Comm.

27. Family 335 4.60 .54 -.OO -.04 .01 -.02 .15" -.07

Role Value

28‘ Family 335 4°23 53 -.02 -.03 -.or -. 10 -.07 .03

Role Comm.

29. Job Sat. 335 4.01 .71 -.18** -.12* -.12"' - -.15** -06
.16“

30. Life Sat. 335 3.72 .77 -.22*"' -.28‘”‘ -.I6** - -.12* -.11"'
.18"

;L< .05, **p<.01; . Where applicable, scale reliabilities are reported on the diagonal

 

Notes: Gender is coded 0=women, I=men; Education is coded l= some high school, 2 = high school

diploma, 3 = some college, 4 = college diploma, 5= some graduate school, 6=graduate degree; Race

is coded 0=Not-White, l= White; Work schedule is coded 0=day shift, l= other shifts; Marital status

is coded 0=married, 1= not married; Parental status is coded 0= has children, 1= does not have

Children; Age of children at home is averaged across children for individuals with multiple children
Jillome

 

72

 

Table 9 (continued)

 

 

Variable N Mean SD 19 20 21 22 23 24

1. Gender 333 .41 .49

2. Age 333 39.25 9.33

3. Education 334 4.96 1.16

4. Race 332 .92 .27

5. Marital Stat. 334 .07 .26

6. Parental Stat 328 .85 .35

7. Children hm 333 1.45 1.09

8. Age of child 256 8.45 6.45

9. Work hr 334 2.64 .87

10. Work sch 329 .14 .35

I 1. WFC time 345 2.16 .92

ext. em

12. WFC time 345 2.35 .80

int. em.

13. WIF stress 345 2.41 .89

all em.

14. FIW stress 345 2.03 .81

all em.

15. WIF beh. 345 2.05 .91

all em.

16. FIW beh. 343 1.76 .75

all em.

17. WIF time 340 2.98 .99

I8. FIW time 340 2.76 .97

I9. WIF stress 340 2.89 1.06 (.88)

20. FIW stress 340 2.17 .93 .32" (.91)

21. WIF beh 340 2.52 .78 .26“ .22" (.74)

22. FIW beh 340 2.49 .81 .26" .24” .76" (.87)

23. Positive 338 3.53 .62 -.32" -.25“ -.31** -.37" (.87)

Affect

24. Negative 338 1.92 .62 .46“ .37" .27" .26" -.41" (.87)
Affect

25. Work Role 336 4.18 .60 .01 .02 -.15** -.15** -.15** -.01
Value

26. Work Role 336 3.07 .67 .00 -.02 -.04 -.01 .33" -.01
Comm.

27. Family 335 4.60 .54 -.1 1* -.19** -.03 -.08 .22" -.l4*"'
Role Value

23- Fami'y 335 4'23 '53 -. 14“ -.i7" -.10 -.18“ .22“ -.12’
Role Comm.

29. Job Sat. 335 4.01 .71 -.27"‘* -.2I** -.30** -.27" .49" -.40”
30. Life Sat. 335 3.72 .77 -.29" -.4l“'* -.18* -.22" .47" -.42“

"‘ g < .05, **p<.01; . Where applicable, scale reliabilities are reported on the diagonal

 

Notes: Gender is coded 0=women, I=men; Education is coded l= some high school, 2 = high school
diploma, 3 = some college, 4 = college diploma, 5= some graduate school, 6=graduate degree; Race
is coded 0=Not-White, 1= White; Work schedule is coded 0=day shift, 1= other shifts; Marital status
is coded 0=married, 1= not married; Parental status is coded 0= has children, 1= does not have
Children; Age of children at home is averaged across children for individuals with multiple children at

home

 

 

73

 

Table 9 (continued)

 

 

Variable N Mean SD 25 26 27 28 29 30
1. Gender 333 .41 .49

2. Age 333 39.25 9.33

3. Education 334 4.96 1.16

4. Race 332 .92 .27

5. Marital Stat. 334 .07 .26

6. Parental Stat 328 .85 .35

7. Children hm 333 1.45 1.09

8. Age of child 256 8.45 6.45

9. Work hr 334 2.64 .87

10. Work sch 329 .14 .35
I 1. WFC time 345 2.16 .92

ext. em
12. WFC time 345 2.35 .80
int. em.
13. WIF stress 345 2.41 .89
all em.
14. FIW stress 345 2.03 .81
all em.
15. WIF beh. 345 2.05 .91
all em.
16. FIW beh. 343 1.76 .75
all em.

17. WIF time 340 2.98 .99
18. FIW time 340 2.76 .97
I9. WIF stress 340 2.89 1.06
20. FIW stress 340 2.17 .93
21. WIF beh 340 2.52 .78
22. FIW beh 340 2.49 .81
23. Positive 338 3.53 .62

Affect

24. Negative 338 1.92 .62

Affect

25. Work Role 336 4.18 .60 (.76)

Value

26. Work Role 336 3.07 .67 .52" (.76)

Comm.

27. Family 335 4.60 .54 .20“ -.07 (.80)

Role Value

28. Family 335 4.23 .58 .10 -.17" .49" (_72)

Role Comm.

29. Job Sat. 335 4.01 .71 .27“ .24" .18" .12“ (.87)
30. Life Sat. 335 3.72 .77 .17“ .12" .26" .25" .50" (.88)

_“p < .05, **p<.01; . Where applicable, scale reliabilities are reported on the diagonal
Notes: Gender is coded 0=women, I=men; Education is coded l= some high school, 2 = high school
diploma, 3 = some college, 4 = college diploma, 5= some graduate school, 6=graduate degree; Race
is coded 0=Not-White, I= White; Work schedule is coded 0=day shift, 1= other shifts; Marital status
is coded 0=man'ied, 1= not married; Parental status is coded O= has children, 1= does not have
children; Age of children at home is averaged across children for individuals with multiple children
at home

 

74

and that regulating emotions is less difficult for older adults (e.g. Scheibe & Blanchard-
Fields, 2009). Controlling for gender and age allows interpretation of variance explained
above and beyond these demographic variables.

Many researchers control for positive and negative affectivity in their studies of
strain, satisfaction and other affective related variables. Spector et al (2000), however, argue
that this method is unsound. Because negative affectivity may play a substantive role in the
variables of interest. In this study, where speciﬁc negative emotions are assessed, an
individual’s tendency towards negative affectivity is likely to be substantively related to how
they respond to these items. I would argue, along with Spector et al (2000), that controlling
for negative affectivity would remove meaningful variability in the variables I am interested
in and thus result in incorrect statistical tests. For that reason, neither positive nor negative
affectivity will be considered as a control variable in subsequent analyses.

Hypothesis Testing

Main Eﬂect Hypotheses: Hypotheses la, 1b, 2a, 2b, 3a, 3b, 4a and 4b were all
parallel hypotheses and were tested in a series of hierarchical, linear regressions. Each type
of WFC emotion (WFC time externalizing, WFC time internalizing, WIF strain all emotions,
FIW strain all emotions, WIF behavior all emotions, and F IW behavior all emotions) was
tested as the dependent variable. In order to control for potential confounding person level
characteristics, the ﬁrst step of these interactions includes gender and age. Step two consists
of the work-family conﬂict variables by direction and type.

The regression for Hla and Hlb is presented in Table 10. These hypotheses
addressed the question of whether individuals who experience more WIF and FIW time-
based conﬂict also experience higher frequencies of WF C time externalizing emotions. Step
2 of the regression was signiﬁcant (AR2 = .13, p< .01) and both H 1a and Hlb were supported.

The regression for H2a and H2b addressed the hypothesis that individuals who experience

75

more FIW or WIF time-based conﬂict will also experience more WF C time-based
internalizing emotions. The regression for H2a and H2b is presented in Table 10. Step 2 of
the regression was signiﬁcant (AR2 = .17, p< .01), and FIW time and WIF time-based conﬂict
were both signiﬁcantly, positively related to WFC time-based internalizing. Thus, Hla, Hlb,
H2a and H2b were supported.

H3a and H3b addressed the hypothesis that individuals who experience more strain-
based work-family conﬂict will also experience more negative emotions related to strain-
based conﬂict. H3a is presented in Table II. Step 2 of this regression was signiﬁcant (AR2 =
.27, p< .01) such that the WIF strain negative emotions scale is positively predicted by WIF
strain conﬂict. H3b is presented in Table 12. Step 2 of this regression was signiﬁcant as well
(AR2 = .20, p< .01), such that F 1W strain negative emotions is positively predicted by FIW
strain conﬂict. Thus H3a and H3b were supported.

H4a and H4b addressed the hypothesis that individuals who experience more
behavior-based work-family conﬂict will also experience more negative emotions related to
behavior-based conﬂict. The regression for H4a is presented in Table 13. Step 2 of this
regression was signiﬁcant (AR2 = .08, p< .01), such that WIF behavior negative emotions is
positively predicted by WIF behavior conﬂict. The regression for H4b is presented in Table
14. Step 2 of this regression is signiﬁcant as well, such that FIW behavior negative emotions
is positively predicted by FIW behavior conﬂict. Thus H4a and H4b were supported.

In order to test the moderated hypotheses, another series of hierarchical linear
regressions were conducted. Step one is identical and step two adds in the relevant work-
family conﬂict items. A third and fourth step were added to test the interaction variables.

Work Importance Moderations: Hypotheses 5a and 5b predicted that the relationship

between FIW time and WFC time externalizing or internalizing emotions will be moderated

76

 

Table 10

Time-Based WIF and F I W Conﬂict Regressed on WF C Time-Based Extemalizing and Intemalizing

 

 

 

 

 

 

Emotions
Variables Time Based WFC Time Based WFC
Extemalizing Emotions Intemalizing Emotions
SE B a AR2 SE B B AR2
Step I .04" .10"
Gender -.17 .10 -.09 -.36 .08 -.22**
Age -.0| .0: -.13"' -.or .00 -.17**
StepZ .13" .17"
WIF lime based 18 .05 .19M .23 .04 .29M
conﬂict
PM I‘m based .25 .05 .26“ .19 .04 .23"
conﬂict
* p < .05, **p<.or,
N= 332
Table 11

Strain-Based WIF Conﬂict Regressed on Strain-Based WIF Emotions
with Interactions of Strain Based WIF Conﬂict and Family Importance

 

 

 

Variables Strain Based WIF All Emotions
SE B [3 AR2

Step I .04”

Gender -.099 .087 -.06

Age -.02 .00 -.15**

Step 2 .27"

WIF strain based conﬂict .45 .04 .53*

Step 3 .00

Family Role Value .08 .09 .05

Family Role Commitment .02 .09 .01

Step 4 .00

F RV X WIF strain based .00 .07 .00

conﬂict

FRC X WIF strain based -.02 .08 -.01

conﬂict

 

* p < .05, **p<.01, N=332

 

77

 

Table 12

Strain-Based F I W C on/lict Regressed on Strain-Based F I W Emotions
with Interactions of Strain Based F I W Conﬂict and Job Importance

 

 

 

Variables Strain Based FIW All Emotions
SE B B AR2

Step I .04”

Gender -.25 .08 -.15* *

Age -.01 .00 -.09

Step 2 .20”

FIW strain based conﬂict .40 .04 .45"

Step 3 .00

Occupational Role Value .07 .08 .05

Occupational Role .03 .07 .02

Commitment

Step 4 .01

ORRV X F 1W strain based -.03 .09 -.02

conﬂict

ORC X FIW strain based .12 .07 .10

conﬂict

 

*p < .05, **p<.01, N=332

 

 

Table 13

Behavior—Based WIF Conflict Regressed on Behavior-Based WIF
Emotions with Interactions of Behavior Based WIF Conﬂict and

Family Importance

 

 

 

Variables Behavior Based WIF All Emotions
' SE B p AR2

Step 1 .01
Gender -.24 .10 -. I 3*
Age -.00 .01 -.03
Step 2 .07"
WIF behavior based .31 .06 .27"
conﬂict
Step 3 .00
Family Role Value -.00 .1 I .00
Family Role Commitment -.07 .10 -.05
Step 4 .00
FRV X WIF behavior -.10 .14 -.05
based conﬂict
FRC X WIF behavior -.01 .14 -.01

based conﬂict

* p < .05, **p<.01,N=332

 

78

by work importance, such that individuals who place greater importance on work will be
more likely to experience WF C time externalizing or internalizing emotions. The interaction
terms of FIW time conﬂict with the two types of work importance (Work Role Value and
Work Role Commitment) were computed, resulting in 2 variables. In order to test the
interaction, these variables were entered in Step 3 of the regression, with the interaction terms
entered in Step 4. See Table 15 for the regression results for H5a and H5b. The interactions
in Step 4 were non signiﬁcant for both H53 and H5b (AR2 = .01, p> .05 and AR2 = .01,
p>.05, respectively) Thus H5a and H5b were not supported.

Hypotheses 5c predicted that the relationship between FIW strain and FIW strain
negative emotions would be moderated by work importance, such that individuals who place
greater importance on work will be more likely to experience F [W strain negative emotions.
Again, ORV and ORRC were entered in Step 3 and the interaction terms were computed
between F 1W strain and work importance and these variables were entered in Step 4. See
Table 12 for this regression. The fourth step of this regression was non signiﬁcant (AR2 =
.01, p> .05), so H5c was not supported.

Hypotheses 5d predicted that the relationship between F IW behavior and F [W
behavior negative emotions would be moderated by work importance, such that individuals
who place greater importance on work will be more likely to experience FIW behavior
negative emotions. Again, ORV and ORRC were entered in Step 3 and the interaction terms
were computed between FIW behavior and work importance and these variables were entered
in Step 4. See Table 14 for this regression. The fourth step of this regression was non
signiﬁcant (AR2 = .01, p> .05) so H5d was not supported.

Family Importance Moderations: Hypotheses 6a and 6b predicted that the

relationship between WIF time and WFC time externalizing or internalizing emotions

79

 

Table I4

Behavior-Based F I W Conﬂict Regressed on Behavior-Based F I W
Emotions with Interactions of Behavior-Based F I W Conﬂict and
Job Importance

 

 

 

Variables Behavior Based F [W All Emotions
SE B p AR2

Step 1 .01
Gender -. I 7 .08 -.l 1*
Age -.00 .00 -.04
Step 2 .10"
FIW behavior based .31 .05 .33"
conﬂict
Step 3 .01
Occupational Role .06 .08 .05
Value
Occupational Role .03 .07 .02
Commitment
Step 4 .01
ORRV X FIW .03 .09 .02
behavior based
conﬂict
ORC X FIW -.11 .08 -.08
behavior based
conﬂict

 

* p < .05, **p<.01,N=332

 

 

Table 15
Time-based F I W Conﬂict Regressed on Time Based WF C Extemalizing and Intemalizing Emotions with
Interactions of Time-Based F I W Conﬂict and Job Importance

 

 

 

Variables Time Based WFC Time Based WFC
Extemalizing Emotions Intemalizing Emotions

SE B B AR2 SE B [3 AR2
Step I .04" .10"
Gender -.17 .10 -.09 -.34 .08 -.21**
Age -.01 .01 -.I 1" -.01 .00 -.15**
Step 2 .09" .08"
FIW time based conﬂict .31 .05 .32" .25 .04 .30"
Step 3 .01 .00
Occupational Role Value .01 .10 .01 .08 .08 .06
Occupational Role .09 .09 .06 .01 .07 .01
Commitment
Step 4 .01 .01
ORRV X FIW time based -. I4 .10 -.09 -.08 .08 -.06
conﬂict
ORC X FIW time based .14 .09 .10 .12 .07 .10
conﬂict

 

* p < .05, "p<.01, N=332

 

80

will be moderated by family importance, such that individuals who place greater importance
on work will be more likely to experience WFC time externalizing or internalizing emotions.
The same two ﬁrst steps as above were utilized, with Step 3 including the two types of the
two types of family importance (Family Role Value and Family Role Commitment), and Step
4 including the interaction terms of these variables with WIF time. See Tables 16 and 17 for
these regressions. The resulting R-square values were non-signiﬁcant for step three in both
regressions (AR2 = .01, p> .05 and AR2 = .00, p>.05, respectively). Thus H6a and H6b were
not supported.

Hypotheses 6c predicted that the relationship between WIF strain and WIF strain
negative emotions would be moderated by family importance, such that individuals who
place greater importance on family will be more likely to experience WIF strain negative
emotions. The same two ﬁrst steps as above were utilized, with Step 3 including the two
types of the two types of family importance (Family Role Value and Family Role
Commitment), and Step 4 including the interaction terms of these variables with WIF strain.
See Table 11 for this regression. Step 4 was not signiﬁcant (AR2 = .01, p> .05 ) so H6c was
not supported.

Hypotheses 6d predicted that the relationship between WIF behavior and WIF
behavior negative emotions would be moderated by family importance, such that individuals
who place greater importance on family will be more likely to experience WIF behavior
negative emotions. The same two ﬁrst steps as above were utilized, with Step 3 including the
two types of the two types of family importance (Family Role Value and Family Role
Commitment), and Step 4 including the interaction terms of these variables with WIF
behavior. See Table 13 for this regression. Step 4 was not signiﬁcant (AR2 = .01, p> .05 ), so

H6d was not supported.

81

 

Table 16

Time-Based WIF Conﬂict Regressed on Time-Based WF C Extemalizing
Emotions with Interactions of Time-Based WIF Conﬂict and Family

 

 

Importance
Variables Time Based WFC Extemalizing
Emotions
SE B B AR2
Step I .04"
Gender -.20 .10 -.1 1*
Age -.02 .01 -.l7**
Step 2 .06”
WIF time based conﬂict .23 .05 .25"
Step 3 .00
Family Role Value -.1 l .11 -.06
Family Role Commitment -.O4 .10 -.02
Step 4 .01
FRV X WIF time -.06 .10 -.04
FRC X WIF time based -.08 .10 -.05
conﬂict

 

* p < .05, "p<.01,N=332

 

 

Table 17

Time-Based WIF Conﬂict Regressed on Time-Based WF C
Intemalizing Emotions with Interactions of Time-Based WIF Conﬂict

and Family Importance

 

 

 

Variables Time Based WFC Intemalizing
Emotions
SE B [3 AR2
Step 1 .10“
Gender -.37 .08 -.23**
Age -.02 .00 -.2 I **
Step 2 .12"
WIF time based conﬂict .27 .04 .34"
Step 3 .00
Family Role Value -.06 .09 -.04
Family Role Commitment .04 .08 .03
Step 4 .00
FRV X WIF time based -.00 .08 -.00
conﬂict
FRC X WIF time based -.07 .08 -.05
conﬂict

 

’p < .05, **p<.01

 

82

Gender Moderations: Hypotheses H7a, H7b and H8 addressed whether WF C
negative emotions were experienced differently by each gender. H7a predicted that gender
would moderate the relationship between FIW time and WFC time internalizing emotions
such that women would be more likely to experience WFC time internalizing emotions in
connection to F 1W time than men. This was tested by examining the interaction between
gender and F 1W time. See Table 18 for this regression. Step I and two were the same as
above and the interaction term of FIW time and gender was entered as step 3. Although the
main effect of gender was signiﬁcant (13 = -.23, p<.01) and suggested that women experience
more WFC time internalizing emotions than men, the interaction term was not signiﬁcant
(AR2 = .00, p>.05), suggesting that the genders do not experience these emotions in relation
to FIW time in structurally different ways. Thus Hypothesis 7a was not supported.

H7b predicted that gender would moderate the relationship between WIF time and
WP C time internalizing emotions such that women would be more likely to experience WFC
time internalizing emotions in connection to WIF time than men. This was tested by
examining the interaction between gender and W1 F time. See Table 19 for this regression.
Step 1 and two were the same as above and the interaction term of WIF time and gender was
entered as step 3. The interaction term was not signiﬁcant (AR2 = .00, p>.05), suggesting
that the genders do not experience these emotions in relation to WIF time in structurally

different ways. Thus Hypothesis 7b was not supported.

83

 

Table I8
Time-Based F I W Conﬂict Regressed on Time-Based WF C

lnternalizing Emotions with Interactions of Time-Based F I W
Conﬂict and Gender

 

 

 

Variables Time Based WFC Intemalizing
Emotions
SE B B AR2

Step I .10“
Gender -.35 .08 -.22**
Age -.01 .00 -.15**
Step 2 .08"
FIW time based conﬂict .24 .04 .30"
Step 3 .00

F 1W time based conﬂict X .07 .09 .04
Gender

 

* p < .05, **p<.or, N=332

 

 

Table 19
Time-Based WIF Conﬂict Regressed on Time-Based

WF C Intemalizing Emotions with Interactions of Time-
Based WIF Conﬂict and Gender

 

 

 

Variables Time Based WF C
Intemalizing Emotions
SE B B AR2

Step I .10"

Gender -.37 .07 -.23**

Age -.02 .00 -.21**

Step 2 .12"

WIF time based .27 .04 .34"

conﬂict

Step 3 .00

WIF time X Gender .00 .08 .00
"' p < .05, "‘*p<.01, N=332

 

 

84

H8 predicted that gender will moderate the relationship between FIW time and WP C
externalizing emotions such that men will be more likely to experience WFC externalizing
emotions than women. This was tested by examining the interaction between gender and
FIW time. See Table 20 for this regression. Step I and two were the same as above and the
interaction term of FIW time and gender was entered as step 3. The main effect of gender
was signiﬁcant (B = -.23, p<.01) and suggested that women experience more WFC time
externalizing emotions than men, which is contrary to the proposed hypothesized direction.
The interaction term was not signiﬁcant (AR2 = .00, p>.05), suggesting that the genders do
not experience these emotions in relation to FIW time in structurally different ways. Thus
Hypothesis 8 was not supported.

Job Satisfaction. A separate regression was conducted to determine how experiencing
negative WIF emotions relates to job satisfaction. See Table 21 for this regression. The ﬁrst
step contained the same person level characteristics as above, gender and age. The dependent
variable was job satisfaction. The second step consisted of four WFC emotion scales; WF C
time internalizing emotions, WFC time externalizing emotions, WIF strain based emotions,
and WIF behavior based emotions. All four types were expected to relate to less job
satisfaction (H9a-d). The second step of this regression was signiﬁcant (AR2 = .04, p<.05),
but only one type of negative emotions were signiﬁcantly, negatively related tojob
satisfaction. WIF strain based negative emotions were negatively related to job satisfaction ([3
= -.I 5, p<.05), supporting Hypothesis 9c. H9a, H9b and H9d were not supported.

Life Satisfaction. A separate regression was conducted to determine the relation of

negative WFC emotions to life satisfaction. The ﬁrst step contained the same person level

85

 

Table 20

Time-Based WIF Conﬂict Regressed on Time-Based WF C
Intemalizing Emotions with Interactions of Time-Based

 

 

 

WIF Conﬂict and Gender
Variables Time Based WFC
Extemalizing Emotions
SE B B AR2
Step I .04“
Gender -.16 .10 -.09
Age -.01 .01 -.I 1*
Step 2 .09"
WIF time based conﬂict .30 .05 .31"
Step 3 .00
WIF time X Gender -.01 .10 .05

 

* p < .05, "p<.01, N=332

 

 

Table 21

Time-Based WF C Extemalizing, Time-Based WF C
Intemalizing, Strain-Based WIF and Behavior Based WIF
Emotions Regressed on Job Satisfaction

 

 

 

Variables Job Satisfaction

SE B [3 Am
Step I _02:r:
Gender -.|2 .08 -.08
Age .01 .oo .10
Step 2 _044
Time Based WFC
Extemalizing Emotions '-03 -06 --04
Time Based WFC
Intemalizing Emotions -02 -08 -02
Strain Based WIF AII
Emotions -.12 .06 -.15*
Behavior Based WIF All
Emotions -03 .05 .03

 

* p < .05, "p<.01, N=332

 

86

characteristics as above, gender and age. The dependent variable was life satisfaction. The
second step consisted of all six WFC emotion scales, the four tested for job satisfaction plus
FIW strain and FIW behavior based negative emotions. Hypotheses 10a, 10b, 10c, 10d, 10e,
and 10f addressed whether these WFC emotion scales predicted life satisfaction above the
other person characteristics. See Table 22 for this regression. The second step was
signiﬁcant (AR2 = .09, p<.01), but only one of the emotion scales was Signiﬁcantly related to
life satisfaction. FIW strain based negative emotions was signiﬁcantly, negatively related to
life satisfaction (B = -.22, p<.01), thus supporting H10d. HIOa, HlOb, H 10c, HlOe, and H10f
were not supported.

Please see Table 23 for a summary of the hypotheses along with whether or not they
were empirically supported.

Exploratory analyses: The other referenced negative emotions were examined using
an exploratory factor analysis of just those items. This factor analysis resulted in ﬁve factors
with eigenvalues over 1 and which explained 60.09% of the variance; however, an analysis of
the scree plot (see Figure 5) shows that an inﬂection point occurs after factor three. Further,
factors four and ﬁve only contributed minor amounts to the overall variance explained
(4.45% and 3.43% respectively), so another analysis was conducted in which the number of
factors to be extracted was limited to three. Please see Table 24 for this three factor
structure. With this constraint, the total amount of variance explained by the rotated three
factor structure was 50.09 %. This three factor structure appeared justiﬁable and will be
assumed for the remainder of these exploratory analyses.

All of the items that addressed internalizing emotions and involved feeling of having

“let someone down” factored on a Single dimension, regardless of direction or

87

 

Table 22

Time-Based WF C Extemalizing, Time-Based WF C
Intemalizing. Strain-Based WIF, Strain-Based F I W,
Behavior-Based WIF and Behavior-Based F I W Emotions
Regressed on Job Satisfaction

 

 

 

Variables Life Satisfaction
SE B B AR2

Step 1 .00
Gender -.08 .09 -.05
Age -.01 .01 -.r 1*
Step 2 .09"
Time Based WFC - .02
Extemalizing Emotions "02 '06
Time Based WFC - .10
Intemalizing Emotions ”'09 '08
Strain Based WIF All
Emotions -.02 .07 -.O3
Strain Based FIW _
AllEmotions -2] .08

.22"
Behavior Based WIF All
Emotions .04 .06 -.05
Behavior Based FIW All
Emotions -.04 .07 -.04

 

* p < .05, "p<.01, N=332

 

Figure 5: Scree plot for other referenced emotions

 

Scree Plot
14 -
12 ~
10 4

Eigenvalue

 

Othm
1

 

 

1 2 3 4 S 6 7 8 91011121314151617181920

Factor Number

 

 

 

88

WE

m

"‘7!-
5941])

'h ’25-,
.h‘~

3ilﬁl:

3i:-

 

 

type of WF C. This scale will be referred to as Other Intemalizing Emotions. The items that
addressed externalizing emotions, however, seemed to fall on two general factors that lined
up with direction of WFC. This makes sense when considering how the items were worded.
For the FIW direction of WFC, the items were worded such that individuals felt angry or
frustrated with family members for interfering with work while for the WIF direction, the
items were worded such that individuals felt angry or frustrated with coworkers or
supervisors. Thus, these dimensions seem to account for externalizing emotions as targeted
at family members or as targeted at coworkers. These scales will be referred to Other WIF
Extemalizing Emotions and Other FIW Extemalizing Emotions.

Although a priori hypotheses were not speciﬁed for these scales, exploratory analyses
were conducted. Three hierarchical linear regressions were conducted to examine the role
work family conﬂict may play in experiences of these three types of other directed negative
emotions. The results of these three regressions are reported in Table 25. For the Other
Intemalizing emotions, both directions of time and strain conﬂict are Signiﬁcantly related to
experiencing other referenced internalizing emotions. The behavior type of conﬂict,
however, is not related to these emotions. Because the Other Intemalizing emotions scale
spans directions and types of WFC it is possible that it relates to both either direction and all
three types of work-family conﬂict. The fact that the Other Intemalizing emotions scale is
related to time- and strain-based types of WFC but not behavior-based WFC suggests that
individuals may experience behavior-based conﬂict with different patterns of internalizing

emotions than they do for time- and strain-based conﬂict.

89

 

Table 23

Summary of new hypotheses and whether they were supported through analysis

 

Hypothesis Ia: Individuals who experience more WIF time will
experience more WFC time externalizing emotions.

Hypothesis 1 b: Individuals who experience more WIF time will
experience more WFC time internalizing emotions.

Hypothesis 2a: Individuals who experience more FIW time will
experience more WFC externalizing emotions

Hypothesis 2b: Individuals who experience more FIW time will
experience more WF C time internalizing emotions

Hypothesis 3a: Individuals who experience more WIF strain
will experience more WIF strain based negative emotions

Hypothesis 3b: Individuals who experience more FIW strain
will experience more F [W strain based negative emotions.

Hypothesis 4a: Individuals who experience more WIF behavior
will experience more WIF behavior based negative emotions

Hypothesis 4b: Individuals who experience more FIW behavior
will experience more FIW behavior based negative emotions.

Hypothesis 5a: The relationship between FIW time and WFC
time externalizing emotions will be moderated by work
importance, such that individuals who place greater importance
on work will be more likely to experience WFC time
externalizing emotions

Hypothesis 5b: The relationship between FIW time and WFC
time internalizing emotions will be moderated by work
importance, such that individuals who place greater importance
on work will be more likely to experience WFC time
internalizing emotions

90

Supported

Supported

Supported

Supported

Supported

Supported

Supported

Supported

Not Supported

Not Supported

Table 23 (continued)

Hypothesis 5c: The relationship between F [W strain and F [W
strain negative emotions will be moderated by work
importance, such that individuals who place greater importance
on their work will be more likely to experience negative FIW
strain based emotions.

Hypothesis 5d: The relationship between FIW behavior and
FIW behavior negative emotions will be moderated by work
importance, such that individuals who place greater importance
on their work will be more likely to experience negative F [W
behavior based emotions.

Hypothesis 6a: The relationship between WIF time and WFC
time externalizing emotions will be moderated by family
importance, such that individuals who place greater importance
on family will be more likely to experience WFC time
externalizing emotions

Hypothesis 6b: The relationship between WIF time and WP C
time internalizing emotions will be moderated by family
importance, such that individuals who place greater importance
on family will be more likely to experience WFC time
internalizing emotions

Hypothesis 6c: The relationship between WIF strain and WIF
strain negative emotions will be moderated by family
importance, such that individuals who place greater importance
on their family will be more likely to experience negative WIF
strain based emotions.

Hypothesis 6d: The relationship between WIF behavior and
WIF behavior negative emotions will be moderated by family
importance, such that individuals who place greater importance
on their family will be more likely to experience negative WIF
behavior based emotions.

Hypothesis 70: Gender will moderate the relationship between
FIW time and WFC time internalizing emotions such that
women will be more likely to experience WFC time
internalizing emotions in relation to F [W time than men.

Hypothesis 7b: Gender will moderate the relationship between
WIF time and WFC time internalizing emotions such that
women will be more likely to experience WFC time
internalizing emotions in relation to WIF time than men.

Not Supported

Not Supported

Not Supported

Not Supported

Not Supported

Not Supported

Not Supported

Not Supported

 

91

 

Table 23 (continued)

Hypothesis 8: Gender will moderate the relationship between
F [W time and WFC externalizing emotions such that men will
be more likely to experience WFC externalizing emotions in
relation to FIW time than women.

Hypothesis 9a: Individuals who experience more WFC time
externalizing emotions will experience decreased job
satisfaction

Hypothesis 9b: Individuals who experience more WFC time
internalizing emotions will experience decreased job
satisfaction

Hypothesis 9c: Individuals who experience more WIF strain
negative emotions will experience decreased job satisfaction

Hypothesis 9d: Individuals who experience more WIF behavior
negative emotions will experience decreased job satisfaction

Hypothesis 100: Individuals who experience more WFC time
externalizing emotions will experience decreased life
satisfaction

Hypothesis 10b: Individuals who experience more WFC time
internalizing emotions will experience decreased life
satisfaction

Hypothesis 10c: Individuals who experience more WIF strain
negative emotions will experienced decreased life satisfaction

Hypothesis 10d: Individuals who experience more FIW strain
negative emotions will experienced decreased life satisfaction

Hypothesis 10e: Individuals who experience more WIF
behavior negative emotions will experienced decreased life
satisfaction

Hypothesis 10f: Individuals who experience more FIW behavior
negative emotions will experienced decreased life satisfaction

Not Supported

Not Supported

Not Supported

Supported

Not Supported

Not Supported

Not Supported

Not Supported

Supported

Not Supported

Not Supported

 

92

 

Table 24

Varimax Rotated Factor Structure for Other Referenced Items, With

Factor Number Constrained to Three

 

 

Factor
I 2

FIW stress Worry Other .820
WIF stress Won'y Other .741
WIF beh Worry Other .698
F 1W time Worry Other .657
FIW beh Won'y Other .643
WIF time Worry Other .640
FIW stress Guilt Other .628 .308
FIW time Guilt Other .581
WIF stress Guilt Other .577
F [W beh Guilt Other .501
WIF time Guilt Other .463
WIF time Anger Other .823
WIF time Frust Other .792
WIF stress Anger Other .749
WIF stress Frust Other .677
WIF beh Anger Other .522
WIF beh Frust Other .491
FIW time Frust Other .81 1
FIW time Anger Other .802
FIW stress Anger Other .663
F [W stress Frust Other .654
FIW beh Anger Other .569
FIW beh Frust Other .519

 

N=347, factor loam under .30 Simpressed

 

93

For the Other WIF Extemalizing emotions, strain and behavior-based conﬂict is

related to Other WIF Extemalizing emotions, but time based WIF and all three types of FIW

were not related. For Other FIW Extemalizing Emotions, strain based FIW was related to

F1 W Extemalizing Emotions but FIW time and behavior based conﬂict and all three types of

WIF were unrelated to Other FIW Extemalizing Emotions. Again, because Other WIF and

FIW Extemalizing emotions scales span type, but not direction of WFC, it follows that they

would relate to the corresponding direction of WFC. Since all types are encompassed in the

 

Table 25

Time-Based WIF C onﬂict, Time-Based F I W Conﬂict, Strain-Based WIF Conﬂict, Strain-Based F] W
Conﬂict, Behavior-Based WIF Conﬂict and Behavior-Based F I W Conﬂict Regressed on Other WF C
Intemalizing Emotion, Other F I W Extemalizing Emotions and Other WIF Extemalizing Emotions

 

 

 

Variables Other WFC Intemalizing Other FIW Extemalizing Other WIF Extemalizing
Emotions Emotions Emotions

SE B B AR2 SE B B AR2 SE B B AR2

Step 1 .06* .00 .00
t

Gender

-. I 6 .06 -.1 l "‘ -.04 .06 -.03 .08 .07 .05
Age -

-.Ol .00 .00 .00 .00 -.00 .00 -.03

.16"
Step 2 .30“ .21“ .19"
It i #

Time-based
WIF Conﬂict .09 .03 .13“ .02 .03 .04 .02 .04 .03
Strain-based .3 1
WIF Conﬂict .20 .03 .31" .06 .03 .1 1 .20 .04 "
Behavior-based .2 I
WIF Conﬂict -.01 .06 -.01 -.04 .06 -.05 .19 .07 I”
Time-based
FIW Conﬂict .11 .04 .16" .03 .03 .05 .04 .04 .05
FIW Conﬂict .12 .04 .16” .21 .04 H .05 .05 .07
Behavior-based
FIW Conﬂict .06 .06 .07 .10 .05 .14 -.06 .07 -07

 

* p < .05, "p<.01, N=332

 

94

Other WIF Extemalizing and Other FIW Extemalizing scales, then the type distinction of
conﬂict may not be distinguished by the Other WIF and FIW Extemalizing emotions scales.
Further exploratory analyses were conducted examining the relationship of these
three scales to job and life satisfaction. See Table 26 for these regressions. Only Other WIF
Extemalizing emotions was signiﬁcantly related to job satisfaction. Both Other WFC
Intemalizing emotions and Other FIW Intemalizing emotions were related to life satisfaction,

but WIF Extemalizing emotions was not.

 

 

 

 

Table 26
Other Referenced W F C Emotion Scales Regressed on Job Satisfaction and Life Satisfaction
Variables Job Satisfaction Life Satisfaction

SE B B AR2 SE B B AR2
Step I .02“ .00
Gender -.07 .08 -.05 "04
Age .01 .00 .123 -.or
Step2 .10“ .11"
Other WFC .02 -.16*
Intemalizing Emotions '02 '07 '° 1 8 '08
Other WIF - -.02
Extemalizing -_ 34 .06 .34“I -_02 .07
Emotions
Other F IW
Extemalizing .05 .07 .04 -.30 .03 -.22**
Emotions

 

_*p < .05, "p<.01, N=332

95

Discussion

This study took a closer look at how individuals experience negative emotions in
relation to work-family conﬂict. The results supported a conceptualization of work-family
conﬂict negative emotions that incorporates additional emotions other than the traditionally
assumed guilt, namely worry, anger and frustration. Individuals seem to experience all these
emotions in relation to work family conﬂict, and do so idiosyncratically with regard to type
and direction of work-family cOnﬂict. For time-based conﬂict, individuals tend to experience
negative emotions within internalizing or externalizing categories regardless of direction of
conﬂict. For strain- and behavior-based conﬂict, people tend to experience the range of
emotions and the direction of conﬂict (whether it is work interfering with family or family
interfering with work) is more important. The frequency of experiencing these negative
emotions was also positively related to the level of corresponding conﬂict individuals
reported.

In addition to further illuminating the emotional spectrum of work family conﬂict,
this study found that strain-based negative emotions relate to job and life satisfaction,
although the other types of negative emotions did not. WIF strain-based negative emotions
were signiﬁcantly negatively related to job satisfaction while FIW strain based negative
emotions were Signiﬁcantly negatively related to life satisfaction. These ﬁndings open the
door for further research into what sorts of individual reactions to work-family conﬂict may
be the most problematic to the individual. All these ﬁndings will be discussed in more detail
below.

Contributions
Expanding on the emotion variables relevant to WF C . The main goal of this study

was to further explore the types of emotions that individuals experience in response to work-

96

family conﬂict. Although guilt is often assumed to be an emotional response to work-family
conﬂict, other emotions such as worry, anger and frustration can also be experienced in
relation to work-family conﬂict situations. As a ﬁrst step, this study served a descriptive role
to Show that individuals do indeed experience a wide variety of emotions in response to work
family conﬂict. It is clear that guilt is not the only relevant emotion to work-family conﬂict
and that other emotions are similarly prevalent.

Individuals did report feeling guilt in relation to work-family conﬂict most
frequently, especially in the work interference with family direction. This is in line with
many of the assumptions in the literature such that individuals experience guilt in relation to
signiﬁcant others in their lives, and family members tend to be more personally signiﬁcant
than coworkers or organizations (Baumeister et al, 1994). The assumption is that family
members are the most signiﬁcant others in an individual’s life and so when other obligations
interfere with family obligations, this sense of failure to meet obligations is accompanied by
feelings of guilt.

It is also interesting to note, however, that the means for frustration and anger items
were also quite high, with individuals reporting feeling frustrated and angry over both work
interfering with family and family interfering with work. These types of externalizing
emotions have not been studied with regard to work-family conﬂict before and through the
descriptive value of this study it would appear that these emotions are reasonably prevalent
responses to work-family conﬂict and should be acknowledged or addressed in future
research.

At a descriptive level, the means of individual items and speciﬁc emotions paint a
picture of emotional responses to work-family conﬂict that expand beyond the traditionally
referenced guilt emotion. The contention that emotions beyond guilt are experienced in

response to work family conﬂict is further enhanced through the factor structure that emerged

97

from the exploratory factor analysis. Instead of falling along hypothesized emotion lines,
which would indicate that each emotion functioned differently in regard to work-family
conﬂict, the emotions tended to factor together along WFC directions and types. This
suggests that guilt is clearly not the only emotion experienced in response to WF C and that
the other emotional responses to WFC are important parts of the individual experience of
conﬂict.

This ﬁnding is important because different emotions can have different effects on the
individual. Emotion researchers have noted that the regulatory functions and the subsequent
actions that are related to speciﬁc emotions have different impacts on the well-being of the
individual. Guilt, for instance, may serve a regulatory purpose to focus attention on
important others in one’s life (Baumeister, 1994). This increased focus could lead to
investing more time in these relationships which in turn reaps positive returns for the
individual (Zahn-Waxler, Kochanska, Krupnick, & McKnew, 1990). Therefore, individuals
and organizations may not be that concerned about alleviating feelings of guilt due to the
relatively smaller impact of that emotion on well-being.

Worry, on the other hand, has been more consistently linked to negative outcomes,
particularly mental health problems such as depression (Yates, 2007). Anger and other
externalizing emotions also have a more stable negative impact in the well-being of
individuals. Experiencing frequent feelings of anger or frustration can have negative impact
on mental and physical health outcomes (Player, King, Mainous & Geeney, 2007).

Because this study showed that a variety of emotions tend to relate to WF C, it
highlights the possibility WFC may contribute to experiencing some of the negative
outcomes associated with these emotions. There is some evidence in the literature that
Speciﬁc coping strategies, including emotion focused coping (Watson & Sinha, 2008) and

seeking social support (Torkelson & Muhonen, 2003) can help alleviate the negative effects

98

of experiencing these types of emotions. From this study, individuals and organizations
could begin to deal with some of the emotions that emerge from feelings of conﬂict through
education about these coping strategies. This in turn would potentially alleviate some of the
negative outcomes associated with emotions like worry, anger and frustration.

Highlighting different emotional experiences of types of conﬂict. In addition to
supporting the idea that multiple emotions are relevant to the experience of WF C, the factor
structure that resulted from the exploratory factor analysis is interesting for what it has to say
about the type and direction structure of work-family conﬂict. The emotions factored along
WFC type lines of time-, strain- and behavior-based. This lends further credence to the
theoretical argument that the types of work-family conﬂict are indeed distinct and that it is
reasonable to assert that emotional responses to these types are distinct as well. It appears
through this factor structure that individuals distinguish between time-, strain- and behavior-
based conﬂict and that they emotionally experience these events in different ways.

The most interesting aspect of the factor structure for the WFC emotion items is that
direction of WFC was idiosyncratically represented with regard to type of WFC. For time-
based conﬂict, the emotion WF C items factored along internalizing and externalizing lines
such that both FIW and WIF emotions fell on the same factor. Although not initially
hypothesized, this ﬁnding is logically supportable. This factor structure suggests that time
based conﬂict, whether it is experienced from work to home or home to work, affects
individuals emotionally in similar ways and that individuals have a typical way of responding
to time-based conﬂicts. As mentioned in the introduction, part of the deﬁnition of WFC
states that conﬂict events can essentially “go either way” (e.g. result in FIW or WIF) until a
decision by the individual is made (Greenhaus & Beutell, 1985). This is especially clear for
time-based conﬂicts. For example, if a person has a family event (e.g. child’s soccer game)

and a work event (e.g. conference call) scheduled at the same time, the time conﬂict only

99

l "4
- s .
O.

 

.9 .-
',-u
u .1
l"
.u l
3!}.
\.
--. .
la...
rp.‘
h:.\
1” l
‘ I-

r.
lll-—‘
‘ b

I“:
l

 

 

acquires direction when the person makes a choice. If the person decides to attend the soccer
game, this would be considered F 1W time based conﬂict, or WIF time based conﬂict if he/she
chooses to attend the conference call.

Perhaps this multi-potentiality of time-based conﬂict makes the emotional reactions
to WIF and FIW directions quite Similar. In both cases there is a speciﬁc type of conﬂict
(e.g. scheduling conﬂict). Individuals then emotionally respond to this scheduling conﬂict in
typical ways regardless of direction of WFC, with some individuals feeling worried and
guilty (internalizing) and some individuals feeling angry and frustrated (externalizing). These
emotional distinctions may take on more of an individual difference quality with some
individuals being predisposed to emotionally respond to time-based conﬂict in consistent
ways.

Alternatively, the nature of the speciﬁc questions asked for general recollections of
time-based conﬂict events. It may be easier for individuals to recall speciﬁc incidents of
time-based conﬂict than strain- or behavior-based conﬂict because the latter two may occur
in more ambiguous situations. The difference between externalizing and internalizing
emotions in reaction to time-based WFC may be due to individuals recalling different
speciﬁc time-based conﬂict episodes. Some may have recalled those that made them feel
more guilty and some may have recalled those that made them feel more angry or ﬁ'ustrated.
Further research is necessary to determine which of the above scenarios is occurring.

If individuals do tend to fall along the internalizing or externalizing types of
responses to WFC, then this has some implications to their mental and physical health.
Individuals who respond to Situations with internalized emotions such as guilt and worry tend
to experience more depression (Yates, 2007) and withdrawal from social Situations (Vasa &
Pine, 2006). Individuals who experience more externalizing emotions like anger and

frustration, however, are more vulnerable to certain health problems such as hypertension

100

 

 

 

 

\\

m-o.
1

I101 I

a...“

ALn.l

. run-

 

 

 

(Player et al, 2007) and alcoholism (Maurage, Campanella, Philppot, Pham & Joassin, 2007).
Therefore, the experience of WFC negative emotions may affect some of these
intemalizing/extemalizing emotion speciﬁc outcomes and the methods of dealing with these
feelings may differ between groups or situations.

For the strain- and behavior-based types of WFC, speciﬁc types of emotions become
less important but direction of WFC becomes more important in the factor structure. The
emotional responses tend to hang together within WI F and FIW directions. This suggests
that an individual tends to experience general negative emotion in response to strain or
behavior based conﬂict consistently within direction. This too can be justiﬁed. The nature of
strain and behavior based conﬂict is often less cut and dried than time-based conﬂict. The
situations for time based conﬂict are clear and speciﬁc (there is an acknowledged scheduling
conﬂict) while the situations for strain and behavior based conﬂict are more amorphous.
While a scheduling conﬂict can create a simple avenue for blame for negative feelings,
feeling “tired” or “stressed” from work is more difﬁcult to apply to a single direct cause.

Because the situations in which individuals experience WIF and F 1W may be more
ambiguous with regard to the strain- and behavior-based types, the emotional pattern of
responses may differ as well. The diffuse and idiosyncratic nature of strain and behavior
based conﬂict may well lead to a more amorphous emotional response, with individuals
indicating a general negative affectivity to these events where they might feel guilty, worried,
angry or frustrated with relatively equal frequency, either simultaneously or at different
times.

The factor structure that resulted from the WFC emotion items is important because it
further highlights the idea that differences exist between the types of WFC and that
individuals experience these types of conﬂicts differently. Knowing these differences can

point to different steps organizations and individuals might take to help alleviate some of the

101

51‘

A.

' .r

..r..

'3'

P
p

a.“

 

 

potential negative effect of these emotions. The impact of organizations on time-based
versus strain- and behavior-based conﬂict negative emotions may be differently constrained.
If an organization desires to reduce the negative emotions responses to time-based conﬂict,
structural or beneﬁts changes could be implemented to aid working adults in better balancing
their time commitments. For instance, at an organization level required overtime
commitment could be limited and ﬂextime beneﬁts could be implemented. Both of these
changes could help reduce time-based conﬂict and thus time-based conﬂict related negative
emotions.

Negative emotional reactions to strain- and behavior-based conﬂict may be more
difficult for an organization to address however. One way to reduce strain-based conﬂict
may be to reduce workload or time pressures, but the nature of certain jobs frequently
precludes this type of job change. Past research has revealed that individuals seem to
experience stronger negative outcomes from higher strain- than time-based conﬂict (Eby et
al, 2005). This relation may be due to the idea that strain-based conﬂict is more difficult to
effect from a structural level without doing a total job redesign. For these types of emotional
reactions to conﬂict, organizations may be better served at attempting to help individuals
cope with the resultant negative emotions rather than eliminating the conﬂict through job
changes.

Potential to identiﬁz individuals especially at risk for negative emotional
consequences of WF C. This study also contributes to the literature because it has linked the
main effects of experiences of WFC related negative emotions with experiences of WFC.
The support of hypotheses 1a, 1b, 2a, 2b, 3a, 3b, and 4b Shows that individuals report
experiencing WFC related negative emotions more frequently as they experience more WFC.
While ostensibly an Obvious conclusion, these ﬁndings suggest that low levels of conﬂict

seem to have relatively little emotional consequence. If an individual only experiences WFC

102

occasionally, they may not experience much negative emotion in response to these infrequent
conﬂict situations. If an individual experiences a lot of WFC, however, they also tend to
experience more negative emotions to that conﬂict. This in turn could relate to increased risk
for some of the negative effects of experiencing these emotions, such as increased depression
(Yates, 2007) or increased hypertension (Player et al, 2007).

Insight into the role of gender in WF C emotions. Much of the previous literature on
WFC and emotion has highlighted an assumption that women experience more WFC related
negative emotions, especially guilt, than do men. When looking at simple main effects
through the correlation table, this assertion is upheld, with women reporting a greater
frequency of experiencing negative emotions in response to both WFC time and WP C strain
types of conﬂict. In the hypothesis tests for H1, H2 , H3 and H4, gender also played a
signiﬁcant role, with women generally experiencing the emotions more frequently for all
types of the negative emotions scales.

It is interesting to note that while traditional emotional theory (e.g. Garside &
Klimes-Dougan, 2004) might suggest that men would respond to WFC events with more
externalizing emotions than women, the data do not uphold this suggestion. Instead, it seems
that women may experience more WFC time externalizing emotions than men, contradicting
the assumption that emotions such as anger and frustration would occur more frequently in
men. This ﬁnding suggests that in future analysis of WFC, externalizing emotions should not
be disregarded, and assumed gender relations to these types of emotions may not be correct.

In addition, these ﬁndings contribute to the idea that there is value in delineating the
types of WFC. Although there were gender differences for most of the emotions scales,
where women experienced more negative emotions in response to conﬂict than men, WIF
strain-based conﬂict showed no gender differences. This may suggest that men and women

might experience WFC time- and behavior-based conﬂict in emotionally different ways,

103

although the experience of WIF-strain is less gendered This could point to differences in the
ways men and women interpret time-based scheduling conﬂicts and behavior-based conﬂicts
and the ways they feel emotionally about these conﬂicts. Strain-based conﬂict, however,
seems to be interpreted and experienced emotionally in similar ways among the genders.

These ﬁndings further the literature on WFC in a couple of ways. First, there is an
assumption in the literature that women are more bothered when work interferes with family
while men are more bothered with family interferes with work (Grandey, Cordiero &
Crouter, 2005). This research would suggest that women are more emotionally bothered in
relation both types of WFC. Although this ﬁnding may be due to the tendency for women to
undertake more family responsibilities than men (Cinamon, 2006), thus resulting in more
F 1W, the correlation table in this study shows no gender differences for different types of
conﬂict. So it would appear that contrary to popular opinion, the negative emotional
experience of women in response to WFC may be greater regardless of direction of conﬂict.
This ﬁnding suggests that it might be useful for an organization to pay particular attention to
the role of conﬂict for women in the time and behavior types. Efforts to create programs or
beneﬁts that might lessen the effect of time and behavior based conﬂict at work may prove
more beneﬁcial to women than men. In the family domain, individual coping skills may also
be more beneﬁcial for women to use when attempting to deal with family interference with
work. In addition, this research may help couple realize that each spouse might experience
WFC in different ways. This could help married couples deal with conﬂict situations that
reduce the net amount of conﬂict related negative emotions experienced.

Second, unlike the other types of WFC, the domain of WIF-strain based negative
emotions showed no gender differences. This suggests that stressful job situations that
impact family life are just as negatively emotional for women and men. To help alleviate

some of the potential negative effects of WIF strain negative emotions, organizations could

104

help educate workers on effective coping strategies. The utility of interventions such as these
would be equally useful to men and women.

Practically, this research suggests that the ways men and women experience negative
emotions in response to WF C may have interesting similarities and differences. This has
implications for the WFC research literature. Meta-analyses in the ﬁeld have shown that the
samples of most WFC studies are mostly comprised of women (Byron, 2005; Ford et al,
2007). These studies may have lacked the statistical power to discern interesting differences
between men and women in WF C contexts. Future research Should be more cognizant of the
role gender may play and make concerted efforts to obtain a more gender balanced sample.

Examining Other and Self-Referenced Emotions: This study looked at emotions
through items that referenced both the self and other people. Although only the self-
referenced items were retained for the formal hypothesis testing, exploratory analyses
involving the other referenced items revealed interesting differences in the factor structures
the two reference points. While the self-referenced items fell into Six factors that were
distinguished on type of WFC, the other referenced items more clearly fell along factors that
related to internalizing or externalizing emotions. For other referenced internalizing
emotions, neither direction nor type of WFC played a signiﬁcant part in distinguishing that
factor. For other referenced externalizing emotions, direction of WFC was important but
type was less so. These differing factor structures between the different reference points of
the items suggests that reference of emotion may play an important role in the experience of
WFC.

It is possible that the reference point of conﬂict related negative emotions may impact
the severity of the emotions experienced. The guilt research would suggest that individuals
would feel more strongly negative about letting down a signiﬁcant other than letting down

oneself (Baumeister, I994). Self-recriminating anger, however, may be more detrimental to

105

the mental health and well being of the individual than other directed anger (Klonsky &
Muehlenkamp, 2007; Kulik and Brown, 1979). Future studies could make an effort to more
speciﬁcally compare self versus other directed emotions in relation to WFC and how these
might be differentially related to health and satisfaction outcomes.

The relation of the other-referenced items to job and life satisfaction is also
interesting. It would appear that feelings of “letting other people down” as referenced in the
Other WFC Intemalizing emotion scale, has little relation to job satisfaction but does relate to
decreased life satisfaction. Feeling angry or frustrated with family members about FIW
conﬂict is also related to decreased life satisfaction. Feeling angry or frustrated with
coworkers or bosses, however, is related to decreased job satisfaction but does not relate to
life satisfaction. These ﬁndings might indicate what sorts of emotions are more linked to
satisfaction in different domains and may eventually help point to positive interventions that
could decrease these emotions and their negative relationships with satisfaction.

Because job satisfaction is linked to a variety of organizationally relevant outcomes
(e.g. turnover, absenteeism, performance), organizations may take this information and use it
to help mitigate an individual’s negative emotional responses to coworkers and bosses as a
result of WFC. This has implications for the types of other-referenced emotions would be a
priority to the organization. Only Other WIF Extemalizing emotions were related to job
satisfaction, SO an organization may want to focus on ways of alleviating workers experiences
of anger or frustration at coworkers and bosses. The other-directed internalizing emotions
were not related to job satisfaction, so an organization could concern themselves less with
feelings Of worry or guilt.

This distinction between other and self-referenced negative emotions as responses to
WF C is wide open for future research. Studies that more speciﬁcally target instances of

WFC may be more able to get at these distinctions. For instance, instead of assessing a

106

general experience of type and direction of conﬂict, researchers may focus on a Speciﬁc
incident of conﬂict the participant remembers. From this Speciﬁc incident, the individual
could report the occurrence and extremity of different emotions felt in relation to themselves
and in relation of relevant others. This may more accurately assess the different relationships
between self and other-referenced negative emotions in relation to WFC. Although only
exploratory in this paper, the distinction deserves to be more closely examined.

Limitations and future directions

Although this study succeeds at adding to the work-family conﬂict and emotion
literature for the reasons listed above, there are limitations to this study.

WF C emotion measure: Although an acceptable and theoretically interesting factor
structure emerged from the exploratory factor analysis, one dimension, WFC time
internalizing emotion, had relatively low loadings on that factor and relatively high loadings
on other factors. This might suggest that the WFC time internalizing factor may not be well
enough deﬁned or the measures used were inadequately capturing this dimension.

In addition, this measure was designed to closely adhere to the presupposed structure
of work-family conﬂict. The wording of the questions may have contributed to the resultant
factor structure in the self-reference items, especially for strain- and behavior-based conﬂict.
This may not be a serious concern, however, because the time type of WFC emotions had a
different factor structure and the other-referenced items did not fall along traditional WFC
construct lines. Future researchers should be cognizant how the wording and framing of the
measures they use may impact factor structure and subsequent analysis possibilities. This
may require a different way of measuring WFC relevant emotions.

Finally, the measure only assessed individual recollections of general instances of
work-family conﬂict when asked to respond about related emotions. The generality of this

wording may not have been interpreted similarly across individuals. Additionally, the

107

generality of the wording may have made it difficult for participants to anchor their responses
in a speciﬁc recollected conﬂict context. Future research could help mitigate this problem
through prompting respondents to recall and report a speciﬁc conﬂict situation. This would
ground the emotional response anchors in a way that was not present in this study.

Failure tofind moderation hypotheses. This study failed to ﬁnd the hypothesized
moderated relationships. As argued in the introduction, the effects of higher levels of
individual work and family importance theoretically could impact emotional experience of
work-family conﬂict. That is, individuals who do not highly value a speciﬁc role would be
less likely to respond in a negative emotional way to circumstances that disrupt that role.

The failure to ﬁnd anticipated moderations may be due to the fact that, overall, the sample
reported placing high importance on both work and family domains. Family importance, in
particular, appears to have neared a ceiling effect, with the mean value of family role value
being very close to the top value of the scale. Because the means of both work and family
importance variables consistently fell a considerable amount higher than the midpoint, this
restriction of range variability may have made detection of potential moderators impossible.

Future research might try to select a sample based on more variability on the domains
of work and family importance in order to detect hypothesized moderations. In addition, the
measure for work and family importance may need to be reworked, perhaps with more focus
on behaviors than vague values. This may result in greater variability and be less vulnerable
to social desirability effects. Alternatively, the high endorsement of work and family
importance variables may indicate that they should be omitted in future research as they do
not vary substantially across people. Different measures of work and family involvement
could be utilized instead to more accurately assess involvement in different domains.

Third Variable Explanations of Findings. In accordance with other research on work-

family conﬂict, it was hypothesized that WFC negative emotions would be related to job and

108

.' .. 'Id‘l‘i
.' -,\ l
.“\."l ‘
'.-'.-r
5" ﬂ
'~..li~5'
I
.b

2“ 3* 1.9;

f' 7‘18 ll.

:— new
.13: LIL»

'.‘ —
2951'” 1

.l‘b .‘L >0:

Tillie~

nil

L‘s

i=3

 

 

life satisfaction. These relationships were supported for some but not all types and directions
of conﬂict.

Although most of the hypothesized relationships between WFC negative emotions
and job and life satisfaction were not signiﬁcant, FIW and WIF strain-based negative
emotions were signiﬁcantly related to life and job satisfaction, respectively. Individuals who
experience more FIW and WIF-strain negative emotions may also experience decreased life
and job satisfaction. While intriguing, it is important not to over generalize these ﬁndings.
The way the WFC emotion scales were worded solicited responses regarding frequency of
experience. Individuals who have stressful home lives may therefore experience more FIW
strain based negative emotions and more F 1W conﬂict but it might not be the conﬂict per se
that is affecting life satisfaction. Instead, the increased stress in the life domain, regardless of
its spillover into the work domain, may be causing the decreased levels of life satisfaction.
The same argument hold for experiences of stressful situations at work and WIF strain-based
emotions.

Future research should attempt to more completely understand how WFC and WP C
related emotions impact life and job satisfaction above and beyond situational concerns. This
could be accomplished through gathering more information about speciﬁc home and work
stressors and how they relate to the WFC strain negative emotions.

Methodological limitations. The data for this survey were collected at one point in
time from a single individual. Although several of the variables preclude methods other than
self-report (e.g. the emotion variables), concerns of common method variance may remain.
Common method variance may not be a large concern for this study, however. First,
differential relations and correlations were observed among variables. For example, only the
strain based WFC negative emotion scales signiﬁcantly related tojob and life satisfaction.

Other differential relationships were also evident, namely that the WP C negative emotion

109

scales tended to be signiﬁcantly related to the corresponding type and direction of WF C and
be unrelated to the other types and directions of WFC. Although the WFC emotion scales
and the WFC scales were related to each other, they were differently related to constructs like
gender. While the WF C emotion scales were largely Significantly correlated with gender,
none of the WFC scales showed signiﬁcant gender relationships. As discussed by Podsakoff
et al (2003), this would not have occurred if individuals were just responding within a limited
response structure.

Second, the factor structure of the WFC negative emotion scales was complex and
varied. Had common method variance been a seriously limiting factor, a single factor may
have emerged instead of the six that resulted from the EF A. This complex structure would
suggest that there was enough variability in individual responses to restrict concerns of
common method variance.

The threat of common method variance should be addressed more fully in future
studies, however. Future studies should seek alternative methods for assessing some of the
relevant variables in order to hedge against the possibility of common method variance
concerns. Other studies in the literature, for example, collect spousal ratings of participant
WFC in order to get a second method for assessing this variable. Also, emotion researchers
often counteract the problem of common method variable through the use of experience
sampling methodology in which they collect multiple assessments from the same individual
over time, thus reducing the temporal aspect of common method variance.

Experience sampling methodology might also be able to address another weakness in
this study. The survey method utilized required individuals to think back to experiences of
WFC and report on how frequently they tended to experience speciﬁc emotions in these
Situations. Although framing the questions within the context of a situation is thought to aid

in recall (Brown, Williams, Barker & Galambos, 2007), these retrospective reports can fall

110

victim to retrospective biases or simple forgetting (Reis & Gable, 2000). Experience
sampling methodology could be used such that individuals would respond to a short survey
each time they experienced a speciﬁc type of conﬂict, reporting emotions felt in situ. Future
research might wish to examine the issue of WFC related negative emotions using this
method.

Sample concerns: It is also important to note that the sample was mostly well
educated, mostly White and mostly middle to upper-middle class. Although much of WFC
focuses on the professional and semi-professional middle class (see Eby et al, 2000), the
sample demographics limit generalizability of the ﬁndings. Perhaps individuals in lower
income brackets experience emotions in different ways or experience different types of
emotions in relation to conﬂict. Conﬂict itself is less understood for individuals of lower
socioeconomic classes as well. Future research should make more targeted efforts to
understand how individuals of lower socioeconomic status experience conﬂict and associated
emotions.

Research and Practical Implications

Research implications. Despite some of the limitations mentioned above, this study
provides a valuable ﬁrst step in more fully understanding the emotional experience of
individuals who experience work-family conﬂict. Although the emotional component of
WFC has long been assumed, it has rarely been directly studied. Even the studies that do
address emotional responses to WFC directly tend to rather narrowly focus on guilt as the
only relevant emotion (e.g. Hochwarter et al, 2007). This study highlights that a variety of
emotions are important to the experience of WF C, and that the directions and types of WFC
affect the sorts of emotions experienced. Future research should be mindful of this expanded

criterion place for WFC relevant emotions.

lll

Future research can build upon the ﬁndings in this study to make emotion a more
relevant aspect of WFC research. Speciﬁcally, future studies could utilize a more qualitative
approach to determining when and how individuals experience negative emotions to work-
family conﬂict. This measure was only capable of assessing general tendencies of emotional
reaction and was not able to discern which Situations might be more or less distressing. If
distressing situations could be distinguished from the more minor annoying ones, this would
provide a tool to create effective interventions. Managers could be made more aware of
certain, especially problematic situations and work to decrease the potential for them to
occur. Additionally, targeted coping interventions could be created that focus narrowly on the
particularly distressing situations, thus improving efﬁciency and cost-effectiveness of these
interventions.

In addition, this study found some intriguing differences between self- versus other-
directed negative emotions. Although the analyses done with the other directed negative
emotion items were exploratory and parenthetical to the main thrust of this study, the
difference in factor structures between self- and other-directed emotions could provide
fruitful opportunity for future research. Determining what situational variables might lead to
self-blame versus other-blame for the conﬂict experience could provide valuable insight into
how individuals experience these negative emotions.

Future studies could more closely examine these differences. For example, person
level characteristics such as personality may relate to tendencies to blame the self versus
blame another person. The emotion literature suggests that there might be relatively stable
person level characteristics that speak toward inclinations of accepting blame and
responsibility or assigning it elsewhere (Martin & Dahlen, 2005 ). This could impact an
individual’s emotional responses to WFC and how they focus or attribute the negative

emotions they experience. In addition, individual coping strategies propensities could result

112

in individuals being more or less vulnerable to making self-blame attributions. Social
support, for instance, has been shown to relate to how individuals experience work-family
conﬂict (van Daalen, Willemsen, & Sanders, 2006). It is likely that it might impact
individual emotional reactions to conﬂict as well.

Practical implications. This study suggests that individuals who experience more
WF C also experience WF C negative emotions in response to that conﬂict more frequently.
This may indicate that the affective life of individuals who experience more conﬂict might be
more negative. Because increased negative affectivity is linked to a number of undesirable
outcomes, such as increased somatic complaints (Van Diest et al, 2005) and increased mental
health problems (Tarlow & Haaga, 1996), attempts to mitigate the negative affective
responses to WFC for individuals experiencing a large of amount of conﬂict might be
especially important.

Additionally, this study helps normalize the experience of different negative emotions
with regard to WFC. Individuals seem to experience a wide variety of negative emotions in
response to WFC. The popular press would suggest that guilt is the only relevant WF C
related emotion, which may leave individuals experiencing anger, frustration or worry with
less understanding of their own emotional responses. Through illuminating a larger
emotional criterion space for WF C, this study shows individuals that experiencing emotions
other than guilt in response to WFC is common. From this ﬁnding, future steps can be taken
to deal with speciﬁc emotions as they arise.

Conclusions

This study looked at individuals from a wide variety of jobs and family situations and
discovered that the conventional assumptions regarding emotional responses to WF C need
updating. By focusing exclusively on guilt, past studies have missed the richer reality of the

emotional backdrop to WFC and omitted potentially important considerations of frustration,

113

will I
1.. 3.
i
.» .:l\
n, c.

w'ﬁl
r».
\

‘U
‘.l .5.

anger and worry. The role of gender is also further illuminated in this study. For some
aspects of WFC, such as WIF and F 1W time- and behavior-based conﬂict, gender may impact
emotional experiences, but for other aspects of WFC, such as the strain-based type, gender is
of less importance in explaining the emotional responses to WFC. This study provides a ﬁrst
step in considering these varied emotions and suggests that the direction and type of WFC

could be of utmost importance in understanding individual emotional reactions to WFC.

114

Appendix A: Online consent form

Work and Family Roles

Please read the information below:

We will be asking you to respond to a series of questions about your feelings about and
experiences of balancing work and family roles. We are also asking you to respond to some
commonly used demographic questions that will help us interpret the meaning of your

responses to the questionnaire.

We expect that it will take about 20 minutes for you to complete this survey. You will receive
a ten-dollar ($10) gift certiﬁcate to a common online retailer via email upon the completion

of this survey

There are no foreseeable risks associated with participating in this study. Your name and
information will remain conﬁdential. Your privacy will be protected to the maximum extent
allowable by law. The data will be saved for at least ﬁve years after it is collected and will
only be accessible by the primary investigator and one graduate student. By writing your
name below you indicate that you are free to refuse to participate in this project or any part of
this project. You may refuse to participate in certain procedures or answer certain questions.
Participation is completely voluntary. You may choose not to participate at all and may

discontinue your participation at any time without penalty or loss of beneﬁts.

If you have any questions or concerns about your participation in this project, you can reach

115

Megan Huth by phone: (517) 614-5799, fax: (517) 353-4873, email: huthmeg1@msu.edu, or
regular mail: 348 Psychology Building, East Lansing, MI 48824. If you have questions or
concerns regarding your rights as a study participant, or are dissatisﬁed at any time with any
aspect of this study, you may contact - anonymously if you wish - Peter Vasilenko, Ph.D.
Chair ofthe Human Research Protection Program by phone: (517)355-2180, fax: (517)432-

4503, email: irb@msu.edu, or regular mail: 202 Olds Hall, East Lansing, MI 48824.

Please mark the box that says "I agree to give my consent to participate" if you agree to
participate in this study. Mark "1 do not want to participate" if you do not agree to participate
in this study. If you agree to participate, enter your name below and you will be taken to the

survey once this step is completed.

116

Willi I

‘‘‘‘‘
, r

Appendix B: Study advertisement to working parents distributed in a school newsletter.

Attention working parents! Our friends at Michigan State University would like to invite you
to participate in an online survey about how you balance work and family roles. If you work
for pay for at least 20 hours a week you are eligible to participate in this survey. The survey
Should take 15 to 20 minutes and you will receive a $10.00 gift certiﬁcate to an online retailer
(like Amazon.com) for your participation. All working adults in your household are eligible
to participate. Please go to https://psyclrologymsuedu/RyanResearch/ and enter the

password "working" to access the online survey. Thank you!

117

radii C '. l

2"'

.
,. we‘ ‘I
1 ‘ . I.
gran-uh!
-

.I.I rm
. "lg

‘4 ~v

tar-f )-.
au_b v

'u ,4. ~

 

.
\- ‘fi

. . T.
I

p-
‘ ‘51
.5 I

. 5y

 

 

 

Appendix C: Second wave recruitment email

I'm a graduate student here at MSU in the Industrial/Organizational Psychology program. As
part of my master's thesis research, I'm seeking a population of working adults with children
and/or spouses to complete an online survey about balancing work and family responsibilities
and the emotions associated with this effort. Participants will be compensated with a 10
dollar gift certiﬁcate upon completion of the 15 minute online survey. The survey should take
15 to 20 minutes and you will receive a $10.00 gift certiﬁcate to an online retailer (like
Amazon.com) for your participation. All working adults in your household are eligible to
participate. Please go to httpsz/ipsychology.msumlu/RyanResearch/ and enter the password

"working" to access the online survey. Thank you!

118

quill ll '

.4th i ll

-— can
591.5“

8

, .
-.-_ a 'V,
_-n h .\lll-

Iraq?!
.5 u. .

lltt‘tl

rilldr

 

 

Appendix D: Third wave recruitment email (men only)

Hi MSU l/O alumni,

I am currently a third year the Organizational Psychology program here at MSU and I need a
little extra help in getting the participants 1 need for my master's thesis. I'm sure you all
remember how difficult it is to ﬁnd a large sample of working adults. My population of
interest is even smaller, and I am asking your help in meeting the number I need for adequate

power.

I need working male participants who also have a family (are married and/or have

children) to ﬁll out an online survey.

If you meet these criteria, I would be grateful if you could ﬁll out the online survey yourself
and then pass it on to others that qualify. If you don't meet these criteria but know someone
(a coworker, spouse, relative, etc) who does, I would greatly appreciate you passing on the

link to them.

It will take about 15 minutes and you will receive a $10 Amazon.com gift certiﬁcate as a

thank you for your time.

The link is: https::’/psyclrolog\.msucdu"Ryachscarch

 

The access 1D is: workingmen

Thanks for any help you might be able to give. If you have any questions, please email me.

119

Go Green!

Megan Huth

120

Appendix E: Work-Family Interference and Negative Affect scale (new scale)

Sometimes time pressures from work can result in someone being unable to meet family
responsibilities. For example, the timing of work can interfere with family when a person

must work late and miss a family event, such as dinner or a child’s sporting event.

In thinking about times when work responsibilities interfered with time you normally spend
with your family, please respond to the following questions about how Often you generally
feel these emotions. Indicate your agreement with the statements on a 1-5 scale, with 1 being

never and 5 being Always or almost always.

When time pressures from work interfere with my family life, I feel:

1. Guilty for letting down myself, being unable to “do it all”.

2. Guilty for letting down my family, being unable to “be there for them”.

3. Angry at myself for my time management skills

4. Angry at my coworkers, supervisor or organization for interfering with my family
life.

5. Frustrated with my time management skills.

6. Frustrated with my coworkers, supervisor or organization for interfering with my
family life.

7. Worried that my inability to avoid time conﬂict means I am not a good parent,
spouse, or relative.

8. Worried that my inability to avoid time conﬂict means others will see me as not a

good parent, spouse or relative.

121

Sometimes time pressures from family can result in someone being unable to meet work

responsibilities. For example, responsibilities to your family can result in unexpected

absence or tardiness from work, such as when a child is sick or other family concerns keep

you home.

In thinking about times when family responsibilities interfered with time you normally spend

at work, please respond to the following questions about how often you generally feel these

emotions. Indicate your agreement with the statements on a 1-5 scale, with 1 being Never

and 5 being Always or almost always.

When time pressures from family interfere with my work, I feel:

1.

2.

Guilty for letting down myself, being unable to “do it all”

Guilty for letting down my coworkers, supervisor or organization, being unable to
“pull my weight”

Angry at myself for my time management skills

Angry at my family members for interfering with my work life.

Frustrated with my time management Skills.

Frustrated with my family members for interfering with my work life.

Worried that my inability to avoid time conﬂict means I am not a good employee

or coworker.

Worried that my inability to avoid time conﬂict means others will see me as not a

good employee or coworker.

Sometimes feelings of stress from work can result in someone feeling stressed or exhausted

at home. For example, the stress from a work deadline can interfere with family when a

122

person is so stressed out from work that they are unable to emotionally or physically

contribute to family members when at home.

In thinking about times when the stress from work responsibilities interfered with
interactions with your family, please respond to the following questions about how often you
generally feel these emotions. Indicate your agreement with the statements on a 1-5 scale,

with 1 being Never and 5 being Always or Almost Always.

When feelings of stress from work interfere with my family life, I feel:
1. Guilty for letting down myself, being unable to cope with the work stress.
2. Guilty for letting down my family, being unable to contribute to the family.
3. Angry at myself for my stress management Skills
4. Angry at my coworkers, supervisor or organization for causing me stress.
5. Frustrated with my stress management skills.
6. Frustrated with my coworkers, supervisor or organization for causing me stress.
7. Worried that my inabilities to cope with stress means I am not a good parent,
spouse, or relative.
8. Worried that my inability to cope with stress means others will see me as not a

good parent, spouse or relative

Sometimes feelings of stress from home can result in someone feeling stressed or exhausted
at work. For example, the stress from family arguments or tensions can result in a person
who is so stressed out from family factors that they are unable to emotionally or physically

contribute to work obligations when at work.

123

In thinking about times when the stress from family responsibilities interfered with
performance at yourjob, please respond to the following questions about how often you
generally feel these emotions. Indicate your agreement with the statements on a 1-5 scale,

with 1 being Never and 5 being Always or Almost Always.

When feelings of stress from family interfere with my work, I usually feel:

1. Guilty for letting down myself, being unable to cope with the family stress.

2. Guilty for letting down my coworkers or supervisors, being unable to perform at
work.

3. Angry at myself for my stress management skills

4. Angry at my family members for causing me stress.

5. Frustrated with my stress management skills.

6. Frustrated with my family members for causing me stress.

7. Worried that my inability to cope with stress means I am not a good employee or
coworker.

8. Worried that my inability to cope with stress means others will see me as not a

good employee or coworker.

Sometimes behaviors that are useful and appropriate with coworkers can be problematic at
home. For example, when a deadline is approaching, behaving in a demanding way may
work well with coworkers, but this behavior may not work well when responding to your

child(ren) and/or spouse.

In thinking about times when work behaviors proved to be ineffective at home, please

respond to the following questions about how often you generally feel these emotions.

124

Indicate your agreement with the statements on a 1-5 scale, with 1 being Never and 5 being

Always or Almost Always.

1. When work behaviors are ineffective at home, I usually feel: Guilty for letting
down myself, being unable to use effective behaviors

2. Guilty for letting down my family, being unable to effectively behave with them

3. Angry at myself for my behaviors

4. Angry at my coworkers, supervisor or organization for my behaviors.

5. Frustrated with myself for my behavior.

6. Frustrated with my coworkers, supervisor or organization for my behaviors

7. Worried that using the wrong type of behavior means I am not a good parent,
spouse or relative.

8. Worried that using the wrong type of behavior means others will see me as not a

good parent, spouse or relative.

Sometimes behaviors that are useful and appropriate with family can be problematic at work.
For example, being especially patient and accommodating may work well when responding
to your child(ren) and/or spouse, but may not work well when dealing with coworkers or

supervisors.

In thinking about times when family behaviors proved to be ineffective at work, please
respond to the following questions about how often you generally feel these emotions.
Indicate your agreement with the statements on a 1-5 scale, with 1 being Never and 5 being

Always or almost always.

125

When family behaviors are ineffective at work, I usually feel:

I.

2.

Guilty for letting down myself, being unable to use effective behaviors

Guilty for letting down my coworkers, supervisor or organization, being unable to

effectively behave with them

Angry at myself for my behaviors

Angry at my family members for my behaviors.

Frustrated with my behavior.

Frustrated with my family members for my behaviors

Worried that using the wrong type of behavior means I am not a good employee.

Worried that using the wrong types of behavior means others will see me as not a

good employee or coworker.

126

 

Appendix F: Work-Family Conﬂict Scale (from Carlson, Kacmar and
Williams, 2000)
Please indicate your agreement with the following items using the scale below.
Time Based Work Interference with Family
1. My work keeps me from my family activities more than I would like.
2. The time I must devote to my job keeps me from participating equally in household
responsibilities and activities.
3. l have to miss family activities due to the amount of time I must spend on work
responsibilities.
Time Based Family Interference with Work
4. The time I spend on family responsibilities often interfere with my work
responsibilities.
5. The time I spend with my family often causes me not to spend time in activities at
work that could be helpful to my career.
6. I have to miss work activities due to the amount of time I must spend on family
responsibilities.
Strain Based Work Interferences with Family
7. When I get home from work I am often too frazzled to participate in family
activities/responsibilities.
8. I am often so emotionally drained when I get home from work that it prevents me
from contributing to my family.
9. Due to all the pressures at work, sometimes when I come home I am too stressed to
do the things I enjoy.
Strain Based Family Interference with Work.

10. Due to stress at home, I am oﬂen preoccupied with family matters at work.

127

1 1. Because I am often stressed from family responsibilities, 1 have a hard time
concentrating on my work.
12. Tension and anxiety from my family life often weaken my ability to do my job.
Behavior Based Work Interference with Family
13. The problem solving behaviors I use in my job are not effective in resolving
problems at home
14. Behavior that is effective and necessary for me at work would be counterproductive
at home
15. The behaviors I perform that make me effective at work do not help me to be a better
parent or spouse.
Behavior-based Family Interference with Work.
16. The behaviors that work for me at home do not seem to be effective at work.
17. Behavior that is effective and necessary for me at home would be counterproductive
at work.
18. The problem-solving behavior that works for me at home does not seem to be as

useful at work.

128

Appendix G: PANAS (from Watson, Clark and Tellegen, 1988)
This scale consists of a number of words that describe different feelings and emotions. Read
each item and then, using the scale below, indicate to what extent you generally feel that
way, that is, how you feel on average:

1. Interested

2. Disinterested

3. Excited
4. Upset
5. Strong
6. Guilty
7. Scared
8. Hostile

9. Enthusiastic
10. Proud

II. Irritable
12. Alert

13. Ashamed
l4. Inspired
15. Nervous
16. Determined
l7. Attentive
18. Jittery

19. Active

20. Afraid

129

Appendix H: Work Importance— Adapted from Life Role Salience Scale by Amatea et al,

1986

Occupations Role Reward Value Subscale

1.

5.

Having work/a career that is interesting and exciting to me is my most important life
goal.

I expect my job/career to give more real satisfaction than anything else I do.

Building a name and reputation for myself through work/a career is not one of my life
goals (reverse coded)

It is important to me that I have a job/career in which I can achieve something of
importance.

It is important to me to feel successful in my work/career.

Occupational Role Commitment

1.

2.

I want to work, but I do not want to have a demanding career (reverse coded)

It is important to me to make as many sacriﬁces as are necessary in order to advance
in my work/career.

I value being involved in a career and expect to devote time and effort needed to
develop it.

It is important to me to devote a Signiﬁcant amount of my time to building my career
and developing the skills necessary to advance in my career

It is important to me to devote whatever time and energy it takes to move up in my

job/career ﬁeld.

130

Appendix I: Family Importance- Adapted from Life Role Salience Scale by Amatea et al,

1986

Family Role Reward Value Subscale

l.

2.

4.

5.

My life would seem empty if I never had a family.

Although having a family requires many sacriﬁces, the love and enjoyment of family
of one’s own are worth it.

It is important to me to feel I am an effective family member.

The whole idea of having a family and caring for them is not attractive to me.

If I chose not to have a family, I would regret it.

Family Role Commitment

1.

It is important to me to commit whatever time necessary to making my family feel
loved, support and cared for.

Devoting a signiﬁcant amount of time to being with or doing things with my family is
not something I ﬁnd important to do (reverse coded) "This item was excluded
from further analyses due to low reliability“

It is important to me to be very involved in the day-to-day details of maintaining
healthy family relationships

Really involving myself in family relationships involves costs in other areas of my
life that I am unwilling to accept.

I expect to work hard to build good family relationships even if it means limiting my

opportunities to pursue other personal goals.

13]

Appendix J: Job Satisfaction
Job Satisfaction: Please indicate your agreement with the following statements, as they
apply generally:

I. I feel enthusiastic about my work.

2. I feel fairly satisﬁed with my job.

3. Every minute of work seems like it will never end.

4. I ﬁnd real enjoyment in my work.

5. I consider my job rather unpleasant.

I32

Appendix K: Life Satisfaction
Lite Satisfaction Scale: Please indicate your agreement to the following statements, as
they apply generally:

1. In most ways my life is close to my ideal.

2. The conditions of my life are excellent.

3. I am satisﬁed with life.

4. So far I have gotten the important things I want in life.

5. If I could live my life over, I would change almost nothing.

I33

Appendix L: Demographics

In order to describe those who participated in this study, we would appreciate it if you would

answer the following questions about yourself.

Sex: El Male El Female

Age:

Education- Please indicate the highest level completed:
El Some high school

[:1 Completed high school

I] Some college

[:1 Completed college

[:1 Some graduate school

El Graduate degree

Marital status:

ElSingle/never married 1:! Divorced El Widowed EIMarried/partnered

Race/ethnicity

U American Indian or Alaska Native
1:] Asian

[J Black or African American

El Hispanic

134

El Native Hawaiian or other Paciﬁc Islander
[:1 White Non-Hispanic

Cl Other

 

Number of children at home

Age(s) of children at home (separate ages of children with commas)

Number of hours m work in paid employment per week:
El None

El 1-10

[:1 11-20

C] 21-30

13 31-40

CI 41-50

El More than 51

ﬂu; typical work schedule
Cl Day shift

[:1 Afternoon shift

Cl Night Shift

13 Variable

D Not employed for pay

Your job title:

 

135

Household yearly income:
El Less than 15,000
El 16,000-25,000
Cl 26,000-35,000
D 36,000-45,000
El 46,000-55,000
El 56,000-65,000
El 66,000-75,000
El 76,000-85,000
Cl 86,000-95,000
El 96,000-105,000
Cl 106,000-120,000

Cl More than 120,000

Do you have a spouse/partner? El Yes D No

Number of hours of paid employment your spouse/partner engages in per week.
El None

El 1-10

El 1 1-20

CI 21-30

CI 31-40

CI 41-50

El More than 50

I36

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:

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:

How many hours do you spend per week on household maintenance, including lawn care,

gardening, household repairs and improvements, painting, remodeling, ﬁxing appliances,

ﬁxing the car, or arranging for these types of tasks to be done by others:

137

 

References

Adams, G., King, L., King, D. 1996. Relationships of Job and Family Involvement, Family
Social Support, and Work-Family Conﬂict with Job and Life Satisfaction. Journal of
Applied Psychology, 81(4), 411-420.

Amatea, E.S., Cross, E. 0., Clark, J. E., & Bobby, C. L. (1986) Assessing the work and
family role expectations of career oriented men and women: the Life Role Salience
Scales. Journal of Marriage and Family, 48(4), 831-838.

Anderson, C. A. & Bushman, B. J. (2002) Human aggression. Annual Review of Psychology,
53(1), 27-51.

Anonymous. (2008) About WMAG. Retrieved April 15. 2008, Working Moms Against
guilt website: hm)://www.workingmomsagz_rinstguilt.com.

Anonymous. (2008) Employment status of the population by sex, marital status, and
presence and age of own children under 18, 2005-2006 annual averages. Retrieved
March 20, 2008, Bureau of Labor Statistics website:
http://wwwblsgw/newsrelease/famee.t05.htm

Archer, J. (2004) Sex differences in aggression in real-world settings: a meta-analytic
review. Review of General Psychology, 8(4), 291-322.

Aronson, E. (1992) The retum of the repressed: dissonancy theory makes a comeback.
Psychological Inquiry, 3(4), 303-31 1.

Batt, R. & Valcour, P. M. (2003). Human resources practices as predictors of work-family
outcomes and employee turnover. Industrial Relations, 42(2), 189-220.

Baumeister, R. F., Stillwell, A. M. & Heatherton, T. F. (1994) Guilt: an interpersonal
approach. Psychological Bulletin, 115(2), 243-267.

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.

Bennett, P, Lowe, R. & Honey, K. L (2003) Appraisals, core relationship themes, and
emotions: a test of the consistency of reporting their associations. Cognition and
Emotion, 17(3), 511-520.

Bemdsed, M., van der Pligt, J ., Doosje, B. & Manstead, A. S. R. (2004) Guilt and regret:
the determining role of interpersonal and intrapersonal harm. Cognition and Emotion,

18(1), 55-70.

Bizman, A., Yinon, Y. & Krotman, S. (2001) Group-based emotional distress: an

138

extension of self-discrepancy theory. Personality and Social Psychology Bulletin,
27(10), 1291-1300.

Blackburn, R. T., Horowitze, S. M., Edington, S. M., & Klos, D. M. (1986) University faculty
and administrator responses to job strains. Research in Higher Education, 25(1), 31-
41.

Bond, J. T., Thompson, C., Galinsky, E., & Prottas, D. (2003). Highlights of the 2002
National Study of the Changing Workforce. New York: Families and Work Institute.

Borkovec, T. D., Robinson, E., Pruzinsky, T. & DePree, J. A. (1983) Preliminary
exploration of worry: some characteristics and processes. Behavior Research and
Therapy, 21(1), 9-16.

Borkovec, T. D. & Roemer, L. (1995) Perceived functions of worry among generalized
anxiety disorder subjects: distraction from more emotionally distressing topics.
Journal of Behavior Therapy and Experimental Psychiatry, 26(1), 25-30.

Boyar, S., Maertz, C. & Pearson, A. (2005) The effects of work-family conﬂict an family-

work conﬂict on nonattendance behaviors. Journal of Business Research, 58, 919-
925.

Boyar, S., Maertz, C., Pearson, A. & 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), 175-191.

Brayﬁeld, A. H, & Rothe, H. F. (1951). An index of job satisfaction. Journal of Applied
Psychology, 35, 307-311.

Brown, N. R., Williams, R. L., Barker, E. T., & Galambos, N. L. (2007) Estimating
frequencies of emotions and actions: a web-based diary study. Applied Cognitive
Psychology, 21, 259-278.

Bruch, M. A., Rivet, K. M. & Laurenti, H. J. (2000) Type of self-discrepancy and
relationships to components of the tripartite model of emotional distress. Personality
and Individual Diﬂkrences, 29(1), 37-44.

Burley, K. A. (1995) Family variables as mediators of the relationship between work-
family conﬂict and marital adjustment among dual-career men and women. Journal
of Social Psychology, 135(4), 483-497.’

Byron, K. (2005) A meta-analytic review of work-family conﬂict and its antecedents.
Journal of Vocational Behavior, 67(2), 169-198.

Campbell, A. (2006) Sex differences in direct aggression: what are the psychological
mediators. Aggression and Violent Behavior, 11(3), 237-264.

Casper, W. J ., Eby, L. T., Bordeaux, C., Lockwood, A., & Lambert, D. (2007) A review

139

of research methods in lO/OB work-family research. Journal of Applied Psychology,
92(1), 28-43.

Carver, C. S., Lawrence, J. W., & Scheier, M. F. (1999) Self-discrepancies and affect:
incorporating the role of feared selves. Personality and Social Psychology Bulletin.
25(7), 783-792.

Carlson, D. S., Derr, C. B. & Wadsworth, L. L. (2003) The effects of internal career
orientation on multiple dimensions of work-family conﬂict. Journal of Family
and Economic Issues, 24(1), 99-116.

Carlson, D. & Kacmar, K. M. (2000) Work-family conﬂict in the organization: do life
roles make a difference? Journal of Management, 26(5), 1031-1054.

Carlson, D, Kacmar, K.M., & Williams, L. (2000) Construction and initial validation of a

multidimensional measure of work-family conﬂict. Journal of Vocational Behavior,
56, 249-276.

Casper, W., Martin, J., Buffardi, L., & Erdwins, J. (2002) Work-family conﬂict,
perceived organizational support, and organizational commitment among employed
mothers. Journal of Occupational Health Psychology, 7(2), 99-108.

Cinamon, R. G. (2006) Anticipated work-family conﬂict: effects of gender, self efﬁcacy and
family background. The Career Development Quarterly, 54(3), 202-216.

Clore, G. L. & Centerbar, D. B. (2004) Analyzing anger: how to make people mad.
Emotion, 4(2), 139-144.

Diener, D., Emmons, R. A., Larsen, R. J., & Grifﬁn, S. (1985) The satisfaction with life
scale. Journal of Personality Assessment, 49(1), 71-75.

Duxbury, L. E. & Higgins, C. A. (1991) Gender differences in work-family conﬂict.
Journal of Applied Psychology, 76(1), 60-73.

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.

Eysenk, M. W. (2004) Emotion, stress and coping. In M. W. Eysenk (ed), Psychology:
An International Perspective. London, UK, Psychology Press.

Feingold, A. (1994) Gender differences in personality: a meta-analysis. Psychological
Bulletin, 116(3), 429-456.

Festinger, L. (1957). A theory of cognitive dissonance. Oxford, England: Row, Peterson.
Finegold, D., Mohrrnan, S. & Spreitzer, G. M. (2002) Age effects on the predictors of

technical workers’ commitment and willingness to turnover. Journal of
Organizational Behavior, 23(5), 655-674.

140

Ford, M. T., Heinen, B. A. & 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.

Fox, S. & Spector, P. E. (1999) A model of work frustration-aggression. Journal of
Organizational Behavior, 20, 91 5-931 .

F riede, A. (2008). An intervention-based examination of work-family decision-making
self-efficacy and anticipated work-family conﬂict. Unpublished doctoral dissertation.
Michigan State University, East Lansing.

Frone, M., Russell, M., & Barnes, G. (1996) Work-family conﬂict, gender, and health related
outcomes: 3 study of employed parents in two community samples. Journal of
Occupational Health Psychology, 1(1), 57-69.

Prone, M., Russell, M., & Cooper, M.L. (1992a) Prevalence of work-family conﬂict: are
work and family boundaries asymmetrically permeable? Journal of Organizational
Behavior, 13(7), 723-729.

Frone, M, Russell, M., & Cooper, M. L. (1992b) Antecedents and outcomes of work-family

conﬂict: testing a model of the work-family interface. Journal of Applied Psychology,
77(1), 65-78.

Frone, M., Russell, M. & Cooper, ML. (1997) Relation of work-family conﬂict to health
outcomes: a four-year longitudinal study of employed parents. Journal of Occupational
and Organizational Psychology, 70, 325-335.

Garside, R. B. & Klimes-Dougan, B. (2004) Socialization of discrete negative emotions:
gender differences and links with psychological distress. Sex Roles, 4 7(3-4), 115-128.

Gilbert, P & Miles, J. N. V. (2000) Sensitivity to social put-down: it’s relationship to
perceptions of social rank, shame, social anxiety, depression, anger and self-other

blame. Personality and Individual Differences, 29, 757-774.
Goldberg, J. F. & Harrow, M. (2005) Subjective life satisfaction and objective functional

outcome in bipolar and unipolar mood disorders: a longitudinal analysis. Journal of
Aﬂective Disorders, 89(1), 79-89.

Grandey, A., Cordeiro, B., & Crouter, A. (2005) A longitudinal and multi-source test of
the work-family conﬂict and job satisfaction relationship. Journal of Occupational
and Organizational Psychology, 78, 305-323.

Grandey, A., Tam, A. P., & Brauburger, A. L. (2002) Affective states and traits in the
workplace: Diary and survey data from young workers. Motivation and Emotion,
26(1), 31-55.

Greenhaus, J. & Beutell, N. (1985). Sources of conﬂict between work and family roles.
Academy of Management Review, 10. 76-88.

141

Greenhaus, J., Parasuraman, S. & Collins, K. (2001) Career involvement and family
involvement as moderators of relationship between work-family conﬂict and
withdrawal from a profession. Journal of Occupational Health Psycholog, 6(2), 91-
100.

Greenhaus, J. H., Parasuraman, S., Granrose, C. S., Rabinowitz, & Beutell, N. (1989)
Sources of work-family conﬂict among two career couples. Journal of Vocational
Behavior, 34(2), 133-154.

Greenhaus, A. & Spector, J. (2006). Health consequences of work-family conﬂict: the

dark Side of the work-family interface. In Perrewe and Ganster (Eds) Research in
occupational stress and well-being, 5, 61-98. Amsterdam, JAI press/Elsevier.

Gutek, B. A., Searle, S., & Klepa, L. (1991) Rational versus gender role explanations for
work-family conﬂict. Journal of Applied Psychology, 76(4), 560-568.

Haines, V. Y., Marchand, A., & Harvey, S. (2006) Crossover of workplace aggression
experiences in dual-eamer couples. Journal of Occupational Health Psychology,
11(4), 305-314.

Harrison, D., Newman, D., & Roth, P. (2006) How important are job attitudes? Meta-analytic

comparisons of integrative behaviors outcomes and time sequences. Academy of
Management Journal, 49(2), 305-325.

Heller, D. & Watson, D. (2005) The dynamic spillover of satisfaction between work and
marriage: the role of time and mood. Journal of Applied Psychology. 90(6)5, 1273-1279

Higgins, C., Duxbury, L. & Lee, C. (1994). Impact of life-cycle stage and gender on the
ability to balance work and family responsibilities. Family Relations, 43(2), 144-150.

Higgins, E. T. (1987) Self-discrepancy: a theory relating self and affect. Psychological
Review, 94(3), 319-340.

Higgins, E. T, Shah, J. & Friedman, R. (1997) Emotional responses to goal attainment:
strength of regulatory focus as moderator. Journal of Personality and Social
Psychology, 72(3), 515-525.

Hochwarter, W. A., Perrewe, P. L., Meurs, J. A. & Kacmar, C. (2007) The interactive
effects of work-induced guilt and ability to manage resources on job and life
satisfaction. Journal of Occupational Health Psychology, 12(2), 125-135.

Huang, Y. H., Hammer, L. B., Neal, M. B. & Parrin, N. A. (2004) The relationship

between work-to-family conﬂict and family-to-work conﬂict: a longitudinal study.
Journal of Family and Economic Issues, 25 (1), 79-100.

Hunt, S., Wisocki, P., & Yanko, J. (2003) Worry and use of coping strategies among
older and younger adults. Journal of Anxiety Disorders, 1 7(5), 547-560.

Huth, M., Ryan, A. M., Ilies, R. & Dimotakis, N. (2007) [Work load, emotional

I42

exhaustion and work-family conﬂict in school staff]. Unpublished raw data.

Judge, T. A. & Ilies, R. (2004) Affect and job satisfaction: a study of their relationships at
work and at home. Journal of Applied Psychology, 89(4), 661-673.

Judge, T. A., Ilies, R., & Scott, B. A. (2006) Work-family conﬂict and emotions: effects
at work and at home. Personnel Psychology, 59(4), 779-814.

Kahn, R., Wolfe, D., Quinn, R., Snoek, J. & Rosenthal, R. (1964) Organizational Stress.
New York, NY, Wiley.

Kenworthy, J. B., Canales, C. J ., Weaver, K. D. & Miller, N. (2003) Negative incidental
affect and mood congruency in crossed categorization. Journal of Experimental
Social Psychology, 39(3), 195-219.

Kessler, R. C. (1979). Stress, social status, and psychological distress. Journal of Health and
Social Behavior, 20, 259—272.

Kliegel, M., Jager, T., & Phillips, L. H. (2007) Emotional development across adulthood:
differential age-related emotional reactivity and emotion regulation in a negative

mood induction procedure. International Journal of Aging and Human Development,
64(3), 217-244.

Klonsy, E. D. & Muehlenkamp, J. J. (2007) Self-injury: a research review for practitioners.
Journal of Clinical Psychology, 63(11), 1045-1056.

Kulik, J .A. & Brown, R. (1979) Frustration, attribution of blame, and aggression. Journal
of Experimental Social Psychology, 15(2), I83-194.

Kuppens, P. & Van Mechelen, I. (2007) Interactional appraisal models for the anger
appraisals of threatened self-esteem, other-blame, and frustration. Cognition and
Emotion, 21 (1), 56-77.

Kuppens, P.; Van Mechelen, 1., Smits, D. J. M., De Boeck, P. & Ceulemans, E. (2007)
Individual differences in patterns of appraisal and anger experience. Cognition
and Emotion, 21(4), 689-713.

Lambert, E. G., Hogan, N. L., Camp, S. D. & Ventura, L. A. (2006) The impact of work-
family conﬂict on correctional staff: a preliminary study. Criminology & Criminal
Justice: An International Review, 6(4), 317-387.
Landry-Meyer, L., Gerard, J. M., & Guzell, J. R. (2005) Caregiver stress among grandparents
raising grandchildren: the functional role of social support. Marriage and Family
Review, 37(1-2), 171-190.

Larson, C. L., Nitschke, J. B., & Davidson, R. J. (2007) Common and distinct patterns of
affective response in dimensions of anxiety and depression. Emotion, 7(1), 182-191.

Lazarus, R. S. (1991) Emotion and adaptation. New York:Oxford University Press.

143

 

Lewis, H. B. (1971) Shame and guilt in neurosis. New York: International Universities
Press.

Liu, Y., Prati, P. L, Perrewe, P. L., & Ferris, G. (2008) The relationship between emotional

resources and emotional labor. Journal of Applied Social Psychology, 38(10), 2410-
2439.

Livingston, B. A. & Judge, T. A. (2008) Examination of gender role orientation among
working men and women. Journal of Applied Psychology, 93(1), 207-216.

Marecek, J. & Ballou, D. J. (1981) Family roles and women’s mental health. Professional
Psychology, 12(1), 39-46.

Martin, R. C. & Dahlen, E. R. (2005) Cognitive emotion regulation in the prediction of

depression, anxiety, stress and anger. Personality and Individual Differences, 39(7),
1249-1260.

.Matjasko, J. L. & Feldman, A. F. (2006) Bringing work home: the emotional experiences
of mothers and fathers. Journal of Family Psychology, 20(1), 47-55.

Matthews, E. & Tiedeman, D. (1964). Attitudes toward career and marriage and the

development of life style in young adult women. Journal of Counseling Psychology,
11(4), 375-384.

Maurage, P., Campanella, S., Philippot, P., Pham, T. H. & Joassin, F. (2007) The crossmodal
facilitation effect is disrupted in alcoholism: a study with emotional stimuli. Alcohol
and Alcoholism, 42(6), 552-559.

Measelle, J. F., Ablow, J. C., Cowan, P. A. & Cowan, C. P. (1998) Assessing young
children’s views of their academic, social, and emotional lives: an evaluation

of the self-perception scales of the Berkeley Puppet Interview. Child Development,
69(6), 1556-1576.

Myers, L. B. (1998) Repressive coping, trait anxiety and reported avoidance of negative
thoughts. Personality and Individual Diﬂerence, 24(3), 299-303.

Parasuraman, S. & Simmers, C. (2001) Type of employment, work-family conﬂict and
well-being: a comparative study. Journal of Organizational Behavior, 22(5), 551-
568.

Parkinson, B. (1999) Relations and dissociations between appraisal and emotion ratings
of reasonable and unreasonable anger and guilt. Cognition and Emotion, 13(4), 347-
385.

Payne, R. L. & Morrison, D. (2002) The differential effects of negative affectivity on

measures of well-being versus job satisfaction and organizational commitment.
Anxiety, Stress & Coping: An International Journal, 15(3), 231-244.

144

Player, M. S., King. D. A., Mainous, A. G., & Geesey, M. E. (2007) Psychosocial factors and
progression from prehypertension to hypertension. Annals of Family Medicine, 5(5),
403-41 1.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y. & Podsakoff, N. P. (2003) Common
method biases in behavioral research: a critical review of the literature and
recommended remedies. Journal of Applied Psychology, 88(5), 879-903.

Reis, H. T. & Gable, S. L. (2000) Event-sampling and other methods of studying everyday
experience. In Reis, H. T. & Judd, C. M. (Eds), Handbook of Research Methods in
Social and Personality Psychology, pp 190-222, Cambridge, UK, Cambridge
University Press.

Roney, C. J. R. & Sorrentino, R. M. (1995) Self-evaluation motive and uncertainty
orientation: asking the “who” question. Personality and Social Psychology
Bulletin, 21 (12), 1319-1329.

Roseman, I. J ., Spindel, M. S. & Jose, P. E. (1990) Appraisals of emotion-eliciting
events: testing a theory of discrete emotions. Journal of Personality and Social
Psychology, 59(5), 899-915.

Santanello, A. W. & Gardner, F. L. (2007) The role of experiential avoidance in the
relationship between maladaptive perfectionism and worry. Cognitive Therapy
and Research, 31(1), 319-332.

Scheibe, S. & Blanchard-Fields, F. (2009) Effects of regulating emotions on cognitive
performance: what is costly for young adults is not so costly for older adults.
Psychology and Aging, 24(1), 217-223.

Schoenfelt, E. L. & Battista, L. (2004) A laboratory study of satisfaction effects on mood

state, withdrawal intentions and citizenship behavior. Psychological Reports, 95(3),
803-820.

Scott, L. & O’Hara, M. W. (1993) Self-discrepancies in clinically anxious and depressed
university students. Journal of Abnormal Psychology, 10(2), 282-287.

Shaffer, M. A., Harrison, D. A., Gilley, K. M., & Luk, D. M. (2001) Struggling for
balance amid turbulence on international assignments: work-family conﬂict,
support and commitment. Journal of Management, 27(1), 99-121.

Silfver, M. & Helkama, K. (2007) Empathy, guilt and gender: a comparison of two
measures of guilt. Scandinavian Journal of Psychology, 48(3), 239-246.

Song, 2., Foo, M. D., & Uy, M. A. (2008) Mood spillover and crossover among dual-

eamer couples: a cell phone event sampling study. Journal of Applied
Psychology, 93(2), 443-452.

145

Spector, P. E., Zapf, D., Chen, P. Y. & F rese, M. (2000) Why negative affectivity should not
be controlled in job stress research: Don’t throw out the baby with the bath water.
Journal of Organizational Behavior, 21(1), 79-95.

Stickler, G. B., Salter, M., Broughton, D.D. & Alario, A. (1991) Parents’ worries about
children compared to actual risks. Clinical Pediatrics, 30(9), 522-528.

Shaffer, M. A., Harrison, D. A., Gilley, K. M., & Luk, D. M. (2001) Struggling for
balance amid turbulence on international assignments: work-family conﬂict,
support and commitment. Journal of Management, 27, 99-121.

Simonds, V. M. & Whiffen, V. E. (2003) Are gender differences in depression explained
by gender differences in co-morbid anxiety. Journal of Aﬂective Disorders, 77(3),
197-202.

Smith, C. A., Haynes, K. N., Lazarus, R. S., & Pope, L. K. (1993) In search of “hot”
cognitions: attributions, appraisals, and their relation to emotion. Journal of
Personality and Social Psychology, 65(5), 916-929.

Smith, C. A. & Lazarus, R. S. (1993) Appraisal components, core relational themes, and
the emotions. Cognition & Emotion. Special Issue: Appraisal and Beyond, 7(3-4),
233-269.

Strauman, T. J. & Higgins, E. T. (1987) Automatic activation of self-discrepancies and
emotional syndromes: when cognitive structures inﬂuence affect. Journal of
Personality and Social Psychology, 53(6), 1004-1014.

Sugiura, Y. (2007) Responsibility to continue thinking and worrying: evidence of
incremental validity. Behavior Research and Therapy, 45(7), 1619-1628.

Tangney, J. P & Dearing, R. L. (2002). Gender differences in morality. In Bomstein, R.
F. & Masling, J. M. (Eds) The Psychodynamics of Gender and Gender Role. Empirical
Studies in Psychoanalytic Theories, 10. 251-269. Washington, DC: American
Psychological Association.

Tarlow, E. M. & Haaga, D. A. F. (1996) Negative self concept: speciﬁcity to depressive
symptoms and relation to positive and negative affectivity. Journal of Research in
Personality, 30(1), 120-1 27.

Thomas, L. & Ganster, D. (1995) Impact of family-supportive work variables on work-family
conﬂict and strain: a control perspective. Journal of Applied Psychology, 80(1), 6-15.

Torkelson, E. & Muhonen, T. (2003) Coping strategies and health symptoms among men and
women in a downsized organization. Psychological Reports, 92(3), 899-907.

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),
462-476.

146

Van Diest, I., de Peuter, S., Eertmans, A., Bogaerts, K., Victoir, A., & van den Bergh, O.
(2005) Negative affectivity and enhanced symptom reports: differentiating between
symptoms in men and women. Social Science and Medicine, 61(8), 1835-1845.

Vansteelandt, K. & Van Mechelen, l. (2006). Individual differences in anger and sadness:

in pursuit of active Situational feature and psychological processes. Journal of
Personality, 74(3), 871 -910.

Vasa, R. A. & Pine, D. S. (2006) Anxiety disorders. In Eds. Essau, C. A. Child and
Adolescent Psychopathology. New York, NY: Routledge/Taylor & Francis Group.

Watson, D. & Clark, L. E. (1991) Self- versus peer ratings of speciﬁc emotional traits:
evidence of convergent and discriminant validity. Journal of Personality and Social
Psychology, 60(6), 927-940.

Watson, D., Clark, L. E., & Tellegen, A. (1988) Development and validation of brief
measures of positive and negative affect: the PANAS scales. Journal of Personality
and Social Psychology, 54(6), 1063-1070.

Watson, D. & Pennebaker, J. W. (1989) Health complaints, stress and distress: exploring
the central role of negative affectivity. Psychological Review, 96(2), 234-254.

Watson, D. C. & Sinha, B. (2008) Emotion regulation, coping and psychological symptoms.
International Journal of Stress Management, 15(3), 222-232.

Weiss, H. M. & Cropanzano, R. (1996) Affective events theory: a theoretical discussion of
the structure, causes and consequences of affective experiences at work. In Staw, B.
M. & Cummings, L. L. (Eds) Research in Organizational Behavior: An Annual
Series of Analytical Essays and Critical Reviews, 18, 1-74. U.S.; Elsevier Science/J A
Press.

Wharton, A. S. & Erickson, R. J. (1993) Managing emotions on the job and at home:
understanding the consequences of multiple emotional roles. Academy of
Management Review, 18(3), 457-486.

Williams, K. J. & Alliger, G. M. (1994) Role stressors, mood spillover, and perceptions

of work-family conﬂict in employed parents. Academy of Management Journal,
37(4), 837-868.

Yates, A. (2007) Anxiety Disorders. Retrieved April 1, 2008, from eMedicine web site:
http://www.emedicine.com/medffOP1C152.htm.

Zahn-Waxler, C., Kochanska, G., Krupnick, J. & McKnew, D. (1990). Patterns of guilt in
children of depressed and well mothers. Developmental Psychology, 26, 51-59.

147