 

 

, . .w .
u
n 5.. ..
wﬁwﬁ nu ..
x
.

.
. i
3. q. a”,

2mm

.

. 2

., a
V awﬂﬁ

. .

:4 "41W“. .3.

3mm

. x
I

1...?

. 4

“an.“ H

mm»?

.— a.

.3..hul.ﬁpﬂf
s...ﬁ!..........
. V n . u.-
.. z r 3.3..
i. 592;...

In!!! 3 7.! a.
x1~p1<

 

 

LIBRARY
MiChlgal. State
University

 

 

 

This is to certify that the

thesis entitled

FAMILY INTERFERENCE WITH WORK
IN SPECIAL NEEDS CAREGIVERS

presented by

Kerrie L. Vanden Bosch

has been accepted towards fulfillment
of the requirements for

M.A. _ Psychology
degree 1n

 

 

   

Major professor

Date /Z,/Z 3/02’

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:lClRC/DatoDuo.p65—p.15

FAMILY INTERFERENCE WITH WORK IN SPECIAL NEEDS CAREGIVERS
By

Kerrie L. Vanden Bosch

A THESIS

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

MASTER OF ARTS
Department of Psychology

2002

ABSTRACT
FAMILY INTERFERENCE WITH WORK IN SPECIAL NEEDS CAREGIVERS
By

Kerrie L. Vanden Bosch

It has been demonstrated in the literature that the role of being a caregiver affects
the amount of family interference with work (FIW) a person experiences, which in turn
affects the amount of life satisfaction and job satisfaction that a person experiences. This
research will investigate the effects of caregiving status on the FIW-job/life satisfaction
relationships and work withdrawal behaviors. This study sampled 119 caregivers, 98 of
whom were employed at a large midwestern university. In order to capture an acceptable
amount of variance in caregiving, 21 of the caregivers were recruited through agencies
that provide support for people who care for chronically ill loved ones. The proposed
model predicts that Special needs caregiving, deﬁned as caregiving that involves unique
caregiving behaviors as well as large time demands, moderates the relationship between
FIW and job/life satisfaction. The model also lays out the mediating processes in this
relationship. In addition, the model predicts that those who are special needs caregivers
will exhibit more withdrawal behaviors than those who are not. In general, only the main
effects proposed in the model are supported. The only supported hypotheses are the are
the negative relationships between FIW and life satisfaction and FIW and job
satisfaction, the positive relationship between FIW and work withdrawal, and the positive
relationship between special needs caregiving and the experience of work as an escape.
However, due to low response rates and low power, it is likely that even if an effect was

present, this study would not have been able to detect it.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... vi
LIST OF FIGURES .............................................................................. viii
INTRODUCTION ..................................................................................... 1
The Nature of Work and Caregiving Responsibilities ................................ 1

Special Needs Caregiving ................................................................ 2

Purpose ..................................................................................... 4
Overview of Thesis ........................................................................ 4
Work-Family Conﬂict ..................................................................... 5

The Nature of the Work-Family Interface ........................................... 5

The Nature of Work-Family Conﬂict ................................................ 8

The Directional Nature of Work-Family Conﬂict .................................... 11

Two Domains of Work-Family Conﬂict ............................................ 11

Sources of Family Interference With Work ........................................ 12

Outcomes of Work-Family Conﬂict ................................................... 13

Job Satisfaction ........................................................................ 14

Life Satisfaction ........................................................................ 20

Work Withdrawal Behaviors ......................................................... 23

Model of Work-Related Outcomes for Special Needs Caregivers ................ 25
Hypotheses ................................................................................ 26
Hypothesis 1 ............................................................................ 26
Hypothesis 2 ............................................................................ 27
Hypothesis 3A .......................................................................... 27
Hypothesis 3B .......................................................................... 27
Hypothesis 4A .......................................................................... 28
Hypothesis 4B .......................................................................... 28
Hypothesis 5A .......................................................................... 28
Hypothesis SB .......................................................................... 28
Hypothesis 6 ............................................................................ 29
Hypothesis 7 ............................................................................ 29
Hypothesis 8A .......................................................................... 30
Hypothesis 8B .......................................................................... 3O
Hypothesis 9A ......................................................................... 30
Hypothesis 9B ........................................................................ 31
Hypothesis 10 ........................................................................... 31
METHOD ............................................................................................. 31
Participants ................................................................................ 31
Procedure .................................................................................. 34
Measures .................................................................................. 36
Caregiving Assessment ............................................................... 36

iii

Work-Family Conﬂict ................................................................. 39

Job Satisfaction ........................................................................ 39
Life Satisfaction ........................................................................ 40
Work Withdrawal Behaviors ......................................................... 40
Standards of Acceptability ............................................................ 40
Coworker Social Support ............................................................. 41
Work as an Escape ..................................................................... 41
Resilience ............................................................................... 42
Affect .................................................................................... 42
Demographics .......................................................................... 42
Order of Administration ............................................................... 42
RESULTS ............................................................................................ 43
Caregiving Assessments Analysis ...................................................... 43
Tests of Hypotheses ...................................................................... 51
Hypothesis 1 ............................................................................ 51
Hypothesis 2 ............................................................................ 51
Hypothesis 3A .......................................................................... 53
Hypothesis 3B .......................................................................... 55
Hypothesis 4A .......................................................................... 56
Hypothesis 4B .......................................................................... 57
Hypothesis 5A .......................................................................... 58
Hypothesis SB .......................................................................... 60
Hypothesis 6 ............................................................................ 61
Hypothesis 7 ............................................................................ 61
Hypothesis 8A .......................................................................... 63
Hypothesis 8B .......................................................................... 65
Hypothesis 9A .......................................................................... 66
Hypothesis 9B .......................................................................... 67
Hypothesis 10 ........................................................................... 68
Post Hoc Analyses ........................................................................ 69
Negative Affectivity ................................................................... 69
Relationship Between FIW and J DI Dimensions .................................. 71
DISCUSSION ....................................................................................... 73
Future Research ........................................................................... 78
APPENDIX A Power Analysis .................................................................... 80
APPENDIX B Recruitment Letter for Caregivers of the Chronically Ill .................... 83
APPENDIX C University Employee Recruitment Letter ...................................... 85
APPENDIX D Caregiving Assessment.......................; .................................. 87
APPENDIX E Complete Caregiving Measure .................................................. 91
APPENDIX F Work-Family Conﬂict (W IF and FIW) Items ................................. 97
APPENDIX G Job Satisfaction Items ............................................................ 99
APPENDIX H Life Satisfaction Items ............................................................ 102
APPENDIX I Work Withdrawal Items ........................................................... 104
APPENDIX J Standards of Acceptability Scale ................................................. 106

iv

APPENDIX K Coworker Social Support Items ................................................ 108

APPENDIX L Work as an Escape Items ....................................................... 110
APPENDD( M Resilience Items ................................................................. 112
APPENDIX N Positive and Negative Affect Items ........................................... 114
APPENDD( 0 Demographic Items ............................................................. 116
REFERENCES ..................................................................................... 118

Table 1.

Table 2.
Table 3.
Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.
Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.

Table 18.

LIST OF TABLES

Sample Characteristics: Mean Age, Weekly Hours of Work, Number of

Dependent Children ...................................................................... 32
Sample Characteristics: Race, Gender and Marital Status .......................... 33
Size and Response Rate of Subsarnples ............................................... 36
Caregiving Factors with Item Loadings ............................................... 46
Extreme Behaviors and Overall Caregiving Descriptives by Type of
Caregiving ................................................................................. 48
Means and Standard Deviations for and Correlations Between Age, Hours of
Work, Number of Children, and Caregiving Variables .............................. 49
Means, Standard Deviations, and Intercorrelations of all Subjective Report
Variables in the Study ................................................................... 50
Regression Relating FIW to Job Satisfaction ........................................ 51
Regression Testing the Hypothesis that Extreme Behaviors Moderates the
Relationship Between FIW and Job Satisfaction .................................... 52
Regression Testing the Hypothesis that Overall Caregiving Moderates the
Relationship Between FIW and Job Satisfaction .................................. 53
Regression Relating Extreme Behaviors to Standards of Acceptability ........ 54
Regression Relating Overall Caregiving to Standards of Acceptability. . . . . ...54
Regression Determining Whether Standards of Acceptability Moderates the
Relationship Between FIW and Job Satisfaction .................................. 55
Regression Relating Extreme Behaviors to Coworker Social Support ......... 56
Regression Relating Overall Caregiving to Coworker Social Support ......... 57
Regression Determining Whether Coworker Social Support Moderates the
Relationship Between FIW and Job Satisfaction .................................. 58
Regression Relating Extreme Behaviors to Work as an Escape ................. 59
Regression Relating Overall Caregiving to Work as an Escape ................. 59

vi

Table 19.

Table 20.

Table 21.

Table 22.

Table 23.

Table 24.

Table 25.

Table 26.

Table 27.

Table 28.

Table 29.

Table 30.

Table 31.

Table 32.

Table 33.

Regression Determining Whether Work as an Escape Moderates the

Relationship Between F IW and Job Satisfaction .................................. 60
Regression Relating FIW to Life Satisfaction ..................................... 61
Regression Testing the Hypothesis that Extreme Behaviors Moderates the
Relationship Between FIW and Life Satisfaction .................................. 62
Regression Testing the Hypothesis that Overall Caregiving Moderates the
Relationship Between FIW and Life Satisfaction .................................. 63
Regression Relating Extreme Behaviors to Resilience ........................... 64
Regression Relating Overall Caregiving to Resilience ........................... 64
Regression Determining Whether Resilience Moderates the Relationship
Between FIW and Life Satisfaction ................................................. 65
Regression Relating Extreme Behaviors to Positive Affect ...................... 66
Regression Relating Overall Caregiving to Positive Affect ...................... 67
Regression Determining Whether Positive Affect Moderates the

Relationship Between FIW and Life Satisfaction ................................. 68
Regression Relating FIW to Work Withdrawal .................................... 69
Regression Relating Extreme Behaviors to Negative Affect ..................... 70
Regression Relating Overall Caregiving to Negative Affect ..................... 70
Regression Determining Whether Negative Affect Moderate the Relationship
Between FIW and Life Satisfaction ................................................. 71
Intercorrelations Between JDI Job Satisfaction Dimensions and FIW ......... 72

vii

LIST OF FIGURES

Figure 1.
Figure 2.

Figure 3.

Figure 4.
Figure 5.
Figure 6.
Figure 7.

Figure 8.

The proposed model with hypotheses labeled on linkages .......................

The proposed interaction between caregiving status and job satisfaction. .

The proposed interaction between standards of acceptability and job
satisfaction .............................................................................

The proposed interaction between social support and job satisfaction. . .
The proposed interaction between work as an escape and job satisfaction. . ..
The proposed interaction between caregiving status and life satisfaction. .
The proposed interaction between resilience and life satisfaction ..............

The proposed interaction between positive affect and life satisfaction ........

viii

26

27

27

28

29

30

.30

.31

INTRODUCTION

To state the obvious, work and non-work activities encompass the whole of adult
life. Many people spend 20, 40, or even 60 hours per week at work. Employees’ work life
is often affected by their non-work life. They often think of their families while they are
at work, they may receive phone calls from home while they are at work, and they
sometimes have to leave work early to pick up their car from the shop. In the same way,
employees’ non-work life is often affected by their work life, whether it be because of the
schedule constraints that work places on non-work life, because employees bring work
home that they do not ﬁnish during the day, or because employees are distracted at home
by thoughts of what they need to do at work (Staines & Pleck, 1984; Zedeck, 1992).
Work life and non-work life, or particularly family life, often conﬂict. Work-family
conﬂict has been shown to negatively impact job satisfaction and life satisfaction (Kossek
& Ozeki, 1998). Work and non-work activities are important to people (Andrews &
Withey, 1976), their domains often overlap (Zedeck, 1992), and there are important
consequences when they conﬂict.

The Nature of Work and Caregiving Responsibilities

The largest part of non-work life for many people is their family. Many
employees provide a great deal of care for their families, and their families may iconsume
the majority of the non-work waking hours. Employees’ families often depend on them
for daily care, ﬁnancial support, and emotional support. Oﬁen, employees must provide
basic daily care for their children, and they nurture them towards personal growth. There
is a tension here; employees’ families need ﬁnancial support, which involves time spent

by the employee at work, and employees’ families also need emotional nurturance, which

involves time spent by the employee at home. Because of these things, the caregiver role
can require a lot of time and energy from the employee. The work role can help the
employee provide some of this care through salary, health beneﬁts, and family-related
employee assistance programs (Thomas & Ganster, 1995).

One factor that is increasing the non-work demands on employed caregivers is
aging parents. People of the “Baby-Boom Generation” are reaching their ﬁfties and
sixties, and as they do, many must provide care for their elderly parents, both in the form
of ﬁnancial assistance and in the form of assistance with the basic activities of daily life
(Coward, Home, & Dwyer, 1992). Thus, many caregivers, referred to as the “Sandwich
Generation,” are feeling pressure ﬁ'om both sides: pressure to care for their children, and
pressure to care for their aging parents (Buffardi, Smith, O’Brien, & Erdwins, 1999;
Stephens & Townsend, 1997). This is an example of a greater-than-normal caregiving
burden. Caregiving can occur in many forms. The form that will be discussed here will be
referred to as special needs caregiving and will be deﬁned in more detail below.

Special Needs Caregiving

Before developing the hypotheses, the deﬁnition of special needs caregiving must
be differentiated ﬁ'om what is meant by caregiving in general. The current literature does
not address special needs caregiving speciﬁcally, but it describes Situations that are
Similar to what this research aims to explore. For example, Kinney and Stephens (1989)
focus on primary caregivers who provide care to family members diagnosed with
probable Alzheimer’s disease. Marks (1998) deﬁnes the caregivers in her study as those
who assume the role of caring for a frail or disabled family member or friend, and she

indicates that these caregivers experience large amounts of stress and burden. Both of

these studies address providing care for a family member whose needs are greater than
those of normal, healthy children or adults. The case of special needs caregiving in the
present study is analogous to the caregivers in these studies, and the speciﬁc deﬁnition of
special needs caregivers used in this research will be outlined below.

Special needs caregiving, as deﬁned here, is composed of two conditions. First,
special needs caregiving involves an extraordinary time demand and may disrupt the
caregiver’s daily routines. In most cases, special needs caregivers experience higher
demands on their time than others because of the activities involved in caring for ones
with special needs, such as taking loved ones to frequent doctor’s appointments and
providing assistance with bathing, toileting and eating. Thus, the role of special needs
caregiving may disrupt the caregiver’s daily routines. Quittner, Espelage, Opipari, Carter,
Bid and Eigen (1998) found that this occurred in a sample of 33 couples, each of whom
cared for a child with cystic ﬁbrosis.

Second, the caregiving responsibility must involve behaviors that are out of the
ordinary due to the nature of the care recipient’s needs. More speciﬁcally, the caregivers
must be providing care to the care recipients that healthy people of the same age would
not need. For example, healthy two-year-old children cannot bathe themselves, so bathing
a two-year-old child because he or she cannot would not be considered special needs
caregiving. However, healthy sixty-year-old adults can bathe themselves, so providing
needed bathing assistance to a sixty-year old adult would be considered special needs
caregiving. The unique behaviors associated with Special needs caregiving can also
include illness- or disability-speciﬁc assistance, such as assisting with insulin shots and

administering oxygen.

In sum, the research that will be described here will consider special needs
caregiving as a special form of caregiving. Special needs caregiving involves an
extraordinary time demand with possible disruptions of daily routines and unique types of
caregiver behaviors because the problem is unusual.

Purpose

The purpose of this research is to investigate the work-related reactions and
experiences of Special needs caregivers. This study will focus on those who are employed
outside the home but are also providing care to a person with Special needs. It will be
argued that caring for persons with special needs affects the way that the caregiver
experiences the conﬂict between the work role and the family role and that this effect is
different from other forms of family time demands. It is predicted that special needs
caregiving inﬂuences the effects of family interference with work on job and life
satisfaction as well as the amount of work withdrawal the caregiver experiences.

Overview of Thesis

AS discussed earlier, the special needs caregiving role comes with heavy time
demands as well as unique responsibilities. Because non-work life and work life intersect,
it is assumed that the caregiving role will inﬂuence the caregiver’s work life. At times,
the caregiving role and the work role will come into conﬂict. This work-family conﬂict
has been the subject of research over the years. In the next part of this thesis, the
literature on work and family conﬂict will be reviewed to provide a background for the
general conditions in which special needs caregiving as it relates to work is embedded.
After addressing the general nature of the work-family interface and Speciﬁcally, work-

family conﬂict, the focus will turn to the research that has been done related to the type of

conﬂict likely to exist for working persons who serve as the primary caregivers for family
members with special needs: family interference with work (FIW). From there, I will
look at the antecedents and consequences of F IW due to the demands of caring for
special needs persons. In addressing the latter, hypotheses will be presented that relate to
how the special needs caregiving is predicted to impact aspects of work and non-work
life in ways that are different from the demands created by more general forms of
caregiving or by other sources of work-family conﬂict. This will conclude the
introduction to be followed by the method and data analytic plan.
Work-Family Conﬂict

The Nature of the Work-Family Interface

Zedeck and Mosier (1990) describe ﬁve models of the work-family interface. The
ﬁrst, compensation theory, postulates an inverse relationship between work and family
experiences. Thus, according to the position, when experiences in one realm are positive,
experiences in the other realm will be negative, and vice versa (e. g., Greenglass & Burke,
1988). The second model, segmentation, states that there is no conﬂict between work and
family; work and family environments are totally separate entities, and one’s functioning
in one environment bears no effect on one’s functioning in the other environment
(Lambert, 1990). Instrumental theory hypothesizes that a person obtains things in one
environment by means of the other. That is, work outcomes lead to family outcomes, and
vice versa (Evans & Bartolome, 1984). For example, money earned at work could be
used to pay for a family vacation, and relaxation provided by spending time with the
family could lead to better productivity at work. Conﬂict theory postulates that success or

satisfaction in one environment can only be achieved by making sacriﬁces in the other

environment (Crosby, 1984). Finally, spillover theory states that there is a Similarity in
the tenor of experiences at home and the tenor of experiences at work, such that
happiness in the home environment leads to happiness in the work environment, and
happiness in the work environment leads to happiness in the home environment (Zedeck
& Mosier, 1990). Spillover can be positive or negative; happiness at work could lead to
happiness at home, or the negative aspects of work could have an adverse effect on the
family (Barling, 1990). Most of the research in work-family conﬂict has focused on the
spillover theory (Zedeck & Mosier, 1990).

Researchers have not been very successful at ﬁnding support for most of these
models of the family-work interface. Pond and Green (1983) unsuccessfully searched for
support for compensation theory in a sample of 57 married couples with non-working
wives and 61 couples with working wives. London, Crandall and Seals (197 7) suggested
that their ﬁnding of no correlation between work satisfaction and leisure satisfaction in
their study of 1297 American working adults suggested that the two realms are
independent, and thus argued for segmentation theory, but as with any null ﬁnding, a
number of alternative explanations for the data were plausible. For example, it is possible
that the measures they used were not as reliable or valid as they should have been and
thus failed to detect a relationship. The only people to date to study instrumental theory
were Evans and Bartolome (1980). Their qualitative and descriptive study of study of 44
couples indicated that one sixth of them experienced instrumentation, but small sample
sizes and the lack of quantitative rigor makes it difﬁcult for us to make any deﬁnitive
conclusions. Of all the models, spillover is most consistent with the data. Yet, the

evidence for spillover theory has been mixed and suggests that it may operate differently

for different people. Wallace (1997) found that stress caused by the number of hours
worked did not translate into spillover for a sample of 512 male and female members of
the legal profession. Burley (1991) found in a sample of 277 partners in a dual-career
couple relationship that stress caused by time spent in family work was associated with
higher Spillover for women, but not for men. Barnett and Marshall (1992) studied 403
women who were licensed practical nurses or social workers and did not ﬁnd any
negative-spillover effects from job to parenting, but they did ﬁnd positive Spillover.
Higgins and Duxbury (1992) found in a sample of 136 dual-career couples that men in a
dual-career relationship experience signiﬁcant negative spillover such that negative
experiences at work lead to negative experiences at home.

In conclusion, most of the work on the models of the work-family interface is
conceptual, and where empirical data exists, it does not strongly and consistently support
any one theory. That is, there is little support for a position that the relationship between
work and the family is captured within any one of these models. The most reasonable
position is that there are elements of two of them that may be important to the proposed
study. First, the literature on spillover is generally supportive of the idea that job
satisfaction can affect the satisfaction that one feels with his/her life in general (Barnett &
Marshall, 1992; Higgins & Duxbury, 1992). Thus, it would seem reasonable that a person
could experience high life satisfaction due to spillover from satisfaction at the job, even if
he/she experienced stressors at home (e. g., a Special needs caregiving role). However,
even with these data, alternative explanations exist. In particular, the Spillover could arise
from dispositional factors. People who are generally more happy in everything

experience high levels of both job and life satisfaction (Judge & Watanabe, 1993).

Nevertheless, the possibility of a spillover process remains and is relevant to the study
proposed here. Second, the conﬂict model has some bearing on this study due to the
extremity of the demand that caring for special needs persons can create in the family
domain. As people push themselves to the limit in one realm, they may feel that they are
not reaching their goals in another realm. Since the time and energy that a person has
remains more or less constant, when he/she has to give more to one aspect of life, other"
aspects of life may struggle.
The Nature of Work-Family Conﬂict

Whereas Zedeck and others addressed the work family interface, others have
looked at the type of conﬂict that is generated when the two domains interfere with each
other. Work-family conﬂict occurs when the work role and the family role interfere with
each other (Adams & J ex, 1999). According to Greenhaus and Beutell (1985), work-
family conﬂict is caused by three types of conﬂict: time-based conﬂict, strain-based
conﬂict, and behavior-based conﬂict. Time-based conﬂict occurs when one does not have
time to fulﬁll all of his/her responsibilities in the work domain as well as those in the
family domain (Adams & J ex, 1999). Time is a limited resource (Frone, Yardley, &
Markel, 1997). At some point, the more time devoted to one role, the less time he/she has
to devote to the other role. Time commitments, therefore, are a direct cause of work-
family conﬂict (Greenhaus & Beutell, 1985; Parasuraman, Purohit, Godshalk, & Beutell,
1996).

The literature supports the relationship between time pressures and work-family
conﬂict. Gutek, Searle, and Klepa (1991) found in a sample of 423 psychologists as well

as in a sample of 209 senior managers that time spent in family work (i.e., childcare,

housework, Shopping, etc.) as well as time spent in paid work were positively related to
work-family conﬂict. Bartolome and Evans (1 97 9) found that time pressure can cause
people to be preoccupied with one role while physically acting in the other role; this is an
indication of the tension that individuals can feel between their work role and their family
roles. In a study of 522 working adults, Adams and J ex (1999) found that three types of
time management behaviors effectively reduced work-family conﬂict both directly,
through better use of time and the reduction of time pressures, and indirectly, through
increasing peoples’ perceptions of control over time.

Strain-based conﬂict occurs when strain in one role affects performance in the
other role (Greenhaus & Beutell, 1985). Thomas and Ganster (1995) deﬁned strain as the
psychological, physiological, and behavioral changes that occur when people are
continually exposed to stressors. Strain results from perceptions of challenges in the
environment that cause harm or difﬁculty, and these perceptions affect wellness-
outcomes for the person. Characteristics of one’s role can cause strain or distress in an
individual, and this strain in one role prevents the individual’s ability or willingness to
fulﬁll responsibilities in the other role (Frone, Yardley, & Markel, 1997). Thomas and
Ganster (1995), using a sample of 398 employees ﬁ'om health care facilities in Nebraska,
found that employees, who work for organizations with family-supportive elements (e.g.,
ﬂexible scheduling, supportive supervisors) that help reduce the strain of balancing work
and family life, experience lower levels of work-family conﬂict than those who work for
organizations without these strain-reducing family-supportive elements. Prone, Yardley

and Markel (1997) found that work distress and family distress, both of which can lead to

strain, contributed to work-family conﬂict in a sample of 372 working adults who had
spouses and/or children.

Behavior-based conﬂict occurs when the expectations for speciﬁc types of
behavior in one role contrast or are inconsistent in a negative way with the expectations
for behavior in another role (Greenhaus & Beutell, 1985). For example, a woman might
be expected to be self-reliant, powerful, and assertive at work but gentle, nurturing and
loving at home. If she cannot adjust her behavior easily between these two roles, she is
likely to feel behavior-based conﬂict. Bartolome (1972) found that many young male
managers felt conﬂict caused by the differing value systems and behavior expectations of
work and the family. Beyond the Bartolome study, little other data have addressed this
form of work-family conﬂict, to my knowledge.

Of the three sources of conﬂict, time-based conﬂict appears to be the most likely
to be experienced by special needs caregiver. Special needs caregivers provide large
amounts of assistance to those for whom they care. Special needs caregiving is also
consistent with strain-based conﬂict. Having to care for and worry about a loved one with
a chronic illness while trying to fulﬁll one’s job responsibilities causes stress in one’s life
(Stephens, Franks, & Atienza, 1997), and repeated exposure to that stress can lead to
strain (Thomas & Ganster, 1995). Special needs caregivers often experience difﬁculties
in their lives related to their caregiving role, and these difﬁculties can have psychological
or physiological outcomes over time (Lee, 1997). While not ruling out behavioral conﬂict
as a possible form of work-family conﬂict for special needs caregivers, it does not appear

to be a very good ﬁt, theoretically, to the problem. The acts of giving special care to

10

others off the job should not be at odds with work-role behaviors from a value
perspective.

The Directional Nature of Work-Family Conﬂict
Two Domains of Work-Family Conﬂict

Work-family conﬂict sometimes is approached directionally, that is, the conﬂict
of work with the family (work interference with family, or WIF) or vice versa (family
interference with work, or FIW) (Gutek, Searle, & Klepa, 1991; Perrewe, Hochwarter, &
Kiewitz, 1999). WIF involves the interference of the work role with the family role, and
FIW involves the interference of the family role with the work role (Gutek, Searle, &
Klepa, 1991). Although many studies conceptualize FIW and WIF as part of a global,
two-directional construct (Frone, Russell, & Cooper, 1992; Gutek, Searle, & Klepa,
1991), recent research conceptualizes the two as distinct, yet related constructs
(O’Driscoll, Ilgen, & Hildreth, 1992; Kossek & Ozeki, 1998; Perrewe, Hochwarter, &
Kiewitz, 1999).

WIF and FIW are likely to impact job satisfaction and life satisfaction in different
ways, and empirical ﬁndings regarding these differences will be discussed in more detail
later. In most of the work-family literature, the source of the conﬂict is work (Frone,
Russell, & Cooper, 1995; e.g., Zedeck, 1992). Yet, in some of the literature, FIW is the
main construct of interest. Given my interest in the special needs caregiver, FIW is the
direction of the conﬂict in which I am most interested. Therefore I will limit my

discussion to this form of conﬂict.

ll

Sources of Family Interference With Work

Marks (1998) investigated the effect of caring for the disabled on FIW. She found
that a sample of 5782 caregivers of disabled spouses, children, parents and other family
members experienced more family-stress spillover to work (i.e., FIW) than non-
caregivers, and that men who were caring for their spouses experienced it more than
women. Stephens, Franks, and Atienza (1997) measured spillover from work to home
and spillover from home to work and found that some employed adult caregivers of
elderly parents experienced positive spillover from the car'egiver role to the employment
role, that is, positive feelings from the caregiver role carried over into the employment
role. Likewise, negative feelings also appeared to generalize from home to work. The
spillover of feelings from positive experiences at work to home and from positive
experiences at home to work was positively related to the caregivers’ well being, and the
negative spillover of feelings was negatively related to the caregivers well being.
Experiences in the family setting can affect experiences in the work setting; thus, the
difﬁculties or beneﬁts experienced by people because of their caregiving role can affect
their work role. However, the information in this study was all gathered via self-report
and thus is subject to response-response bias, making it likely that the measures of home
life and work life will be related even if the constructs are not. Taking this into account,
we must realize that although their arguments seem reasonable, it is possible that the
results that Stephens et al. (1997) found are merely an artifact of their methodology.

Eagle, Icenogle, Maes, and Miles (1998) classiﬁed people according to marital
status, working or nonworking Spouse or partner, and the presence of children. In this

study, they found that divorced women with children, who experienced large amounts of

12

time pressure because of their family demands, experienced more F IW than men, those
without children, or those who were not divorced. When work-family conﬂict is
differentiated by direction, it is clear that FIW occurs and that time pressure, stress, and
other effects originating at home impact the work environment.

Outcomes of Work-Family Conﬂict

Earlier, the case was made that F IW and WIF are two separate forms of work-
family conﬂict. However, much of the literature treats work-family conﬂict as an
undifferentiated form of conﬂict and then relates that form to outcomes of the conﬂict.
Thus, the discussion in this section will treat WIF and FIW separately when they are
treated as such in the literature, but, because the differentiation is rarely made in the
literature, the discussion will have to rely on a general work-family conﬂict construct for
much of what is reported. Furthermore, caregiving status is rarely measured when
conducting work-family conﬂict research; thus, the samples used may include people
with no children, caregivers of healthy children, and special needs caregivers alike.

The three outcomes of work-family conﬂict that have been studied most
frequently are job satisfaction, life satisfaction, and withdrawal. While withdrawal can
occur in either the work setting or the family setting, this discussion will focus on
withdrawal from work because that is the domain that has received attention ﬁ'om
researchers. The literature regarding these outcomes will be discussed. Most of the
literature is not focused on special needs caregiving, but I believe much of it is relevant to
it. Thus, I will begin each section by reviewing the literature without reference to special
needs caregiving and then end each section by raising the implications that this literature

has to Special needs caregiving.

l3

Job Satisfaction

As previously mentioned, not all of the literature has made the distinction
between WIF and FIW. Thus, in the following review of the literature, I will use the term
“work-family conﬂict” when the authors did not differentiate between the two forms, and
F [W and WIF when they did.

Many studies have exhibited a negative relationship between the general work-family conﬂict
construct and job satisfaction. Netemeyer, Boles, and McMurrian (1996) tested the
relationship between work-family conﬂict and job satisfaction for three different samples,
including elementary and high school teachers and administrators, small business owners,
and real estate sales employees, and found a signiﬁcant negative relationship in each
case. The work-family conﬂict-job satisfaction relationship has been observed
internationally as well; Ahmad (1996) found the signiﬁcant work-family conﬂict-job
satisfaction relationship in a sample of female Malaysian researchers. Bacharach,
Bamberger, and Conley (1991) found job satisfaction to be related to work-home conﬂict
in two samples: one of 430 engineers, and another of 21 5 nurses. This study assumed that
work-home conﬂict caused lower levels of job satisfaction, but this is an inappropriate
conclusion because the data were correlational self-reports. Thompson and Blau (1993)
found that the extent to which the work role interfered with the family role and the
amount of incompatibility between the parent and work roles were Signiﬁcantly related to
job satisfaction in a sample of 234 employees. Judge, Boudreau, and Bretz (1994) studied
WIF; they found job satisfaction to be signiﬁcantly and negatively related to WIF in a
sample of 1388 male executives. This study asserted that this relationship was a causal

one; WIF caused people to experienced lower levels of job satisfaction. Again, however,

14

since the data were correlational self-reports, this assertion was inappropriate. A meta-
analysis of 50 studies veriﬁed the work-family conﬂict-job satisfaction relationship with
a weighted mean correlation of -.31, regardless of the type of measured used (general
construct of work-family conﬂict, F 1W , or WIF; Kossek & Ozeki, 1998).

Researchers have also demonstrated a relationship between job satisfaction and
the way that parents carry out their caregiving activities. In a study of 195 fathers of
elementary school children, Stewart and Barling (1996) found that lower levels of job
satisfaction were related to signiﬁcantly higher levels of parental rejection of their child’s
demands for time (e.g., ignoring the child when the child wanted the father’s attention). It
is likely that parents who reject their children’s demand for time will Spend less time with
their children than those who do not, and they are also likely to experience less time-
based FIW. Thus, it seems that the correlation between FIW and job satisfaction should
be higher for those who are Special needs caregivers in this category than those who are
not. However, this situation is not as clear as Stewart and Barling make it out to be. Hulin
and Judge (2001) indicate that job satisfaction is dispositional, as are social attitudes;
those who experience dispositionally low job satisfaction will also be likely to experience
negative social attitudes. These attitudes may extend to how people feel about their
personal relationships with others, even their children. Thus, Stewart and Barling may be
capturing disposition rather than behavior-attitude links.

Perrewe, Hochwarter, and Kiewitz (1999) suggested that value attainment
mediates the relationship between WIF and FIW and job satisfaction. Speciﬁcally, they
argued that experiencing WIF or F 1W reduces one’s value attainment, which in turn,

lowers job satisfaction. A person compares his/her life activities to the standards that

15

he/she has for his/her life (e. g. success or virtue) and if the life activities match the
standards, then the person has reached value attainment (Tatarkiewicz, 1976). James and
James (1989) stated that values determine the meaning that work and organizational
experiences have for people. If work is instrumental in reaching these values, then job
satisfaction is higher. Conversely, if work is not instrumental in reaching these values,
then job satisfaction is lower (Locke, 1976). Perrewe, Hochwarter, and Kiewitz (1999)
tested value attainment as a mediator. In a sample of 270 hotel managers, they found that
value attainment fully mediated the relationship between work-family conﬂict and job
satisfaction.

Some researchers have explored F IW and WIF separately as each relates to job
satisfaction. The results have generally indicated that the FIW relationship is the stronger
of the two. Perrewe, Hochwarter, and Kiewitz (1999) studied a sample of hotel managers
and found that both WIF and F IW were negatively related to job satisfaction, but the FIW
relationship is stronger than the WIF relationship when value attainment is entered into
the equation as a mediator. They used Gutek, Searle, and Klepa’s (1991) Work-Family
Conﬂict Scale, which measured both WIF and F IW with four-item scales designed to
measure perceptions of work interfering with family and family interfering with work.
The internal consistency reliability estimates for WIF and FIW were .85 and .84,

' respectively. Adams and J ex (1999) found both constructs to be negatively related to job
satisfaction in a sample of 572 working adults, and they also found the FIW-job
satisfaction relationship to be the stronger of the two.

All of the research reported thus far was conducted in the context of various

family demands ranging from none to normal to extreme. The case of a special needs

16

 

caregiver, however, involves extreme family demands in which the caregiver must
provide unique, emotionally intensive, time-consuming care for his/her loved one. It is
possible that the F IW-job satisfaction relationship may play out differently in this
population.

Standards of Acceptability. One possible reason for the expectation of a different
F IW -job satisfaction relationship for special needs caregivers is that the special needs
caregiver may have lower standards of what is an acceptable number of work hassles in a
job. It is likely that they arrive at these standards of acceptability by comparing their life
outcomes to the work outcomes. Comparison levels have been the subject of research for
many years. In the perception literature, Helson and Rohles (1959) found that people
judged the same Shade of gray to be darker when it was overlaid with black lines than
when it was overlaid with white lines, indicating that one’s judgment of color contrast is
not independent; it depends on which colors surround it. Ilgen and Hamstra (1972) found
a similar phenomenon in performance judgments. Using a sample of 150 male
undergraduates, they found that participants’ satisfaction with their performance was
affected by the discrepancy between their expected performance and how they were told
they performed more when they are told they perform high or low than at an average
level. This indicates that the participants’ satisfaction with performance was a function of
the comparison of their expected performance and their reported performance as well as
the comparison of their reported performance to some external performance standard.

Hulin (1969) studied 470 Canadian white-collar workers working in two different
communities. He found that there was a signiﬁcant positive relationship between people’s

satisfaction with the economic conditions (cost of living and cost of housing) in their

17

 

,

town and their satisfaction with their pay. The author argues that the workers use their
contexts when making judgments about things; when judging whether they were satisﬁed
with their pay, they compared the money they earned with the money that they spent to
maintain their lifestyle in their towns. Those who were more satisﬁed with cost of
maintaining their lifestyles in their towns were more satisﬁed with their pay. When
making judgments about things, people often use their other experiences as a ﬂame of
reference.

Thus, when thinking about the problems they have at work, special needs
caregivers may consider their life problems when making those judgments. Since their
lives are likely to contain a large number of hassles (Kinney & Stephens, 1989) and they
may have adapted to that level of hassle, they may have lower standards of acceptability
for what they expect ﬁom their job (e. g., they will tolerate more job hassles, they will not
expect as many beneﬁts, etc.). It is likely that people who have lower standards of
acceptability will experience less of an effect of FIW on job satisfaction because there are
fewer qualiﬁcations that their jobs must meet in order for them to be satisﬁed.

Coworker social support. A second possible reason to expect that the F IW-job
satisfaction relationship may be different for special needs caregivers than caregivers of
healthy children is that special needs caregivers who are struggling with an extreme
family demand may receive social support from their coworkers. Beehr and McGrath
(1992) afﬁrmed that meaningful social support can come from work sources. It usually
takes the form of either emotional support (e. g., listening, compassion, empathy) or
instrumental support (e. g., covering the caregiver’s job tasks so he/she can attend a

doctor’s appointment with his/her loved one). Coworkers will likely be aware of a

18

person’s status as a special needs caregiver. If they are, they may provide encouragement
and a forum for expressing ﬁ'ustrations related to this role as well as instrumental help in
ﬁrlﬁlling the responsibilities associated with either role. Since special needs caregivers
are likely to experience more stressors than those who are not special needs caregivers,
special needs caregivers are also likely to need, seek out, and receive more coworker
social support. McCann, Russo, and Benjamin (1997) found in a sample of 159 lawyers
that social support is related to higher levels of job satisfaction. Thus, it is likely that
those who experience more coworker social support will experience less of a negative
effect of F IW on job satisfaction.

Work as an escape. As previously discussed, special needs caregivers experience
large amounts of emotional turmoil and stress (Marsh & Johnson, 1997). People who
experience such large amounts of stress are likely to need to have a break ﬁom this stress.
This phenomenon is not speciﬁcally explored in the literature, but it is analogous to what
can be observed in the work environment when job stress is high. Della-Rocco,
Comperatore, Caldwell, and Cruz (1999) studied air trafﬁc controllers, a population of
workers who are under extreme levels of stress. They found that those who had nap
breaks during the workday performed better on a vigilance test because they had an
opportunity to refresh themselves. Brody, Kleban, J ohnsen, Hoffman, and Schoonver
(1987) found that employed caregivers tend to experience less caregiver strain and better
emotional health than caregivers who are not employed; thus, the employment role seems
to provide relief from the strain of caregiving. Thus, it is likely that the job role can

provide an escape from the stresses of the special needs caregiving role. Because work

19

can have these therapeutic effects, it is also likely that F IW will not affect job satisfaction
as negatively for those who view work as more of an escape.
Life Satisfaction

Researchers have also found a negative relationship between life satisfaction and
the general work-family conﬂict construct in many studies. Bedeian, Burke, and Moffett
(1988) studied a sample of 423 male and 335 female accountants and found that work-
family conﬂict was negatively and signiﬁcantly associated with life satisfaction.
Appropriate non-causal inferences were made in this study. International samples have
yielded similar results. Richardsen, Burke, and Mikkelsen (1999) studied a sample of 191
professional and managerial women aged 24-55 years in Norway. Women experiencing
high role conﬂict and work-family pressures also tended to experience low life
satisfaction. Ahmad (1996) found a signiﬁcant negative relationship between work-
family conﬂict and life satisfaction in a sample of 82 female Malaysian researchers
employed at research institutions. Judge, Boudreau, and Bretz (1994) studied FIW and
WIF separately in relation to life satisfaction in a sample of 1388 male executives. They
found that both constructs were Si gniﬁcantly related to life satisfaction in the negative
direction. However, they made inappropriate casual inferences; the data used were
correlational self-reports. Kossek and Ozeki’s (1998) meta-analysis of 50 studies
conﬁrmed the work-family conﬂict-life satisfaction relationship with a weighted mean
correlation of -.36, regardless the measure of work-family conﬂict (as a general construct,
as WIF, or as FIW).

Perrewe, Hochwarter, and Kiewitz’s (1999) previously described theory also

suggests that value attainment mediates the relationship between WIF and FIW and life

20

satisfaction. Again, they argued that experiencing WIF or FIW reduces one’s value
attainment, which in turn, lowers life satisfaction. In their study of 270 hotel managers,
they found that value attainment partially mediates the relationship between WIF and life
satisfaction, and it fully mediates the relationship between FIW and life satisfaction.

Some researchers have considered WIF and FIW separately in relation to life
satisfaction, and many found that both WIF is more negatively related to life satisfaction
than is FIW. Adams, King, and King (1996), in a sample of 163 working-adult students,
found only WIF to be negatively related to life satisfaction. Perrewe, Hochwarter, and
Kiewitz (1999), in their study of 270 hotel managers, found signiﬁcant relationships
between both WIF and FIW with life satisfaction. In summary, work-family conﬂict in
general is negatively related to life satisfaction, and both WIF and FIW are as well.

Just as in the discussion of the literature on job satisfaction, the literature related
to life satisfaction focuses on a variety of levels of family demands. When one considers
the case of the special needs caregiver, the family demands are extreme. Since the family
environment within which the F IW-life satisfaction relationships play out changes when
special needs caregivers are considered, it is important to consider whether the
relationships change as a result or whether they remain the same. It is expected that
special needs caregivers will experience the FIW-life satisfaction relationship differently
than caregivers of healthy children. Reasons for this expectation are discussed below.

Resilience. One factor that may account for the different FIW-life satisfaction
relationship for special needs. caregivers is resilience. Families who have to deal with a
loved one with a chronic illness learn to adapt and change in constructive ways, and they

have family resilience (Marsh & Johnson, 1997). Perlesz, Kinsella, and Crowe (1999)

21

report a similar resilience in families of people with traumatic brain injuries. Thus, it is
likely that those who are special needs caregivers will have more resilience than those
who are not. This resilience is a hardiness, toughness or resistance that minimizes the
negative impacts of hardships on families. It may develop because families learn, adapt,
grow and change because of the hardships that they face. This resilience sometimes even
causes the chronic illness to be a positive inﬂuence on the family members. For instance,
Marsh, Leﬂey, Evans-Rhodes, Ansell, and Doer'zbacher (1996) found that families of
mentally ill patients reported that they had stronger family commitments, an expanded
base, increased compassion, and more personal resilience since their loved one was
diagnosed. Thus, it is likely that people who experience this resiliency will experience the
FIW-life satisfaction relationship differently than those who do not.

Aﬂect. Another factor that may account for the different experience of the FIW-
life satisfaction relationship that special needs caregivers have may be the affect
associated with the role of caregiver. Caregivers often experience positive affect from
their caregiving roles (Stephens, Franks, & Townsend, 1994). Even when the stress of the
caregiver, mother and wife roles were accounted for, Stevens, Franks, and Atienza (1994)
found that women caring for their aging and ill parents feel satisfaction in being a Special
needs caregiver that is related to well-being. Thus, it is likely that those who are special
needs caregivers will experience more positive affect than those who are not. This
positive affect and sense of well-being can contribute to life satisfaction (Diener, 1994).
Thus, it is likely that people who experience more positive affect will experience a

weaker relationship between FIW and life satisfaction than those who do not.

22

Work Withdrawal Behaviors

Withdrawal behaviors refer to a set of behaviors employees use in attempts to
remove themselves from their jobs or avoid work tasks (Hanisch & Hulin, 1991). The
general population tends to withdraw due to reduced job satisfaction and the perception
that there are better job alternatives available (Mobley, 1977). The intent of the
withdrawing person is usually to ﬁnd another job (Mobley, 1977). Researchers have
traditionally found withdrawal behaviors, including things like absenteeism, tardiness,
and reluctance to do organizational citizenship behaviors to be positively related to work-
family conﬂict in the general population (e.g., Gignac, Kelloway, & Gottlieb, 1996; Goff,
Mount & J amision, 1990).

Because F IW is partially time-based (Greenhaus & Beutell, 1985), employees
who are caregivers may need to sacriﬁce time at work to tend to some aspect of their
caregiving role. Gignac, Kelloway, and Gottlieb (1996) found that this is true. They
studied 396 women and 316 men from eight different organizations and found that FIW
is positively and Signiﬁcantly related to absenteeism for both women and men; many of
those people may have been caregivers who were absent due to their caregiving
responsibilities. Goff, Mount, and Jamison (1990) studied 62 users and 191 nonusers of
an on-Site company childcare center, and they found that work-family conﬂict was
related to higher levels of absenteeism, and that use of an on-site childcare center did not
reduce this absenteeism. Barling, MacEwen, Kelloway, and Hi gginbottom (1994) studied
a sample of 141 university employees and found work-family conﬂict to be positively
related to partial absenteeism, including coming to work late, leaving work early, and

taking or making phone calls at work related to the caregiving role. Greenhaus,

23

Parasuraman, and Collins (2001) studied 135 male and 64 female Certiﬁed Public
Accountants who were married or in a long-term relationship and had one or more
children, and they found WIF to impact withdrawal intentions. The FIW-withdrawal
behaviors relationship is also evident in the special needs caregiver population. In a
sample of 67 employees of a ﬁnancial institution, Lee (1997) found that caregivers of
elderly parents were absent signiﬁcantly more than those who did not care for their
elderly parents.

When considering special needs caregivers, the reasons for withdrawal behaviors
are different than those for the general population. It is likely that withdraw because they
have to tend to caregiving responsibilities (Lee, 1997). The literature is unclear about the
relationship between caregiver withdrawal and job satisfaction, but because these
caregivers may have to attend to their caregiving roles during working hours, it is likely
that the cause of absenteeism and tardiness for some is because the caregiving role
sometimes requires immediate attention and not because the caregivers are unsatisﬁed
with their jobs. Much of the literature focuses on job dissatisfaction as a source of work
withdrawal (Hulin, Roznowski, & Hachiya, 1985). However, this is a different sort of a
cause. The characteristics of the job are not likely to be the cause of withdrawal; the
characteristics of the family are. The employee likely withdraws because their family
demands are interfering with their work role, and they need to leave their work role for a
while to deal with these pressing family demands.

Thus, the role of being a special needs caregiver affects aspects of a person’s

experiences such as job satisfaction, life satisfaction, and work withdrawal behavior.

24

These three outcomes ﬂow directly out of the work-family conﬂict; many researchers
have focused on these phenomena.
Model of Work-Related Outcomes for Special Needs Caregivers

Figure 1 below depicts the proposed model and the hypotheses that correspond to
each linkage. To summarize what was stated earlier, job satisfaction is negatively related
to FIW in the literature, but most of the studies have looked at a variety of levels of
family demand. Special needs caregivers have an extreme family demand, so the
relationship between F IW and job satisfaction for them may be different than it is for
caregivers of healthy children. Because of the large amounts of stress ﬁom the family
role as compared to the work role, coworker social support, and the escape from their
stressful caregiving role that their jobs provide, it is likely the relationship between FIW

and job satisfaction may be different based on the type of care that a person provides.

25

 

Job

 

 

 

     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Satisfaction
4B] 5B
Standards of Cowqr'ker Work as
FIW Acceptability 5811:3121“ an escape
3ANAT 5A T
T I Type of ' Work

  

Caregiving __, Withdrawal
10 Behaviors

 

 

 

 

 

 

 
  
 
  

 

 

8A 9A

Resilience Positive
Affect

 

 

 

 

 

 

 

 

9B

 

 

 

Life
Satisfaction

 

 

Figure 1. The proposed model with hypotheses labeled on linkages.
Hypotheses
Because these hypotheses are largely exploratory, they reﬂect tendencies that are
expected to occur in the data.
Hypothesis 1

As FIW increases, job satisfaction decreases.

26

 

 

Hypothesis 2

FIW affects job satisfaction differently based on type of caregiving such as the

negative relationship between FIW and job satisfaction is weaker for those who are

special needs caregivers than those who are not.

A

...................... ’ special Needs Caregivers

 

Caregivers of Healthy
Children

Job Satisfaction

 

‘
7

 

Family Interference With Work

Figure 2. The proposed interaction between caregiving status and job satisfaction.

Hypothesis 3A

Type of caregiving affects the standards of acceptability such that those who are

special needs caregivers have lower standards of acceptability than those who are not.

Hypothesis 3B

FIW affects job satisfaction differently based on how low a person’s standard of
what is an acceptable number of work hassles in a job such that the negative relationship

between FIW and job satisfaction is weaker for those who have lower standards of

acceptability. A

....................... ’ Low Standards of Acceptability

High Standards of Acceptability

Job Satisfaction

 

 

F amily-Interference With Work

Figure 3. The proposed interaction between standards of acceptability and job
satisfaction.

27

Hypothesis 4A

Type of caregiving affects coworker social support such that those who are
special needs caregivers experience more coworker social support than those who are not.
Hypothesis 4B

FIW affects job satisfaction differently based on how much coworker social
support a person experiences such that the negative relationship between FIW and job

satisfaction is stronger for those who receive less coworker social support.
it

 

 

 

: .......................

o ooooooooooooooooooooo - .

.§ «u. ngh Socral Support

“-4

"3

m .

'8 Low Social Support
Family Interference With work

Figure 4. The proposed interaction between social support and job satisfaction.

Hypothesis 5A

Type of caregiving affects the experience of work as an escape such that those
who are Special needs caregivers are likely to view work as an escape ﬁom troubles in
their non-work life.

Hypothesis 53

F IW affects job satisfaction differently based on to what extent a person views

work as an escape such that the negative relationship between FIW and job satisfaction

will be weaker for those who View work as an escape more.

28

.................. 5 High Work as an Escape

Job Satisfaction

 

Low Work as an Escape

 

‘
7

Family Interference With Work

Figure 5. The proposed interaction between work as an escape and job satisfaction.

The literature has also found that FIW is negatively related to life satisfaction.
Again, however, most of these studies occurred in a variety of family demands, and it is
likely that the work-family conﬂict-life satisfaction relationship will play out differently
for Special needs caregivers, who all experience extreme family demands. Because
special needs caregivers acquire family resilience through their struggles and they
experience positive affect as a result of their extreme caregiving role, it is likely that the
relationship between FIW and life satisfaction will be different based on the type of care
that a person provides.
Hypothesis 6

As FIW increases, life satisfaction decreases.
Hypothesis 7

FIW affects life satisfaction differently based on type of caregiving such as the
negative relationship between FIW and life satisfaction is weaker for those who are

special needs caregivers than those who are not.

29

................... > Special Needs Caregivers

 

Caregivers of Healthy
Children

‘
7

Life Satisfaction

 

Family Interference With Work

Figure 6. The proposed interaction between caregiving status and life satisfaction.

Hypothesis 8A

Type of caregiving affects resilience such that those who are Special needs
caregivers have more resilience than those who are not.

Hypothesis SB

F IW affects life satisfaction differently based on how resilient a person is such

that the negative relationship between FIW and life satisfaction is weaker for those who

are more resilient.

.................. ‘ High Resilience

Low Resilience

Life Satisfaction

 

 

L
V

Family Interference With Work

Figure 7. The proposed interaction between resilience and life satisfaction.

Hypothesis 9A

Type of caregiving affects affect such that those who are special needs caregivers

experience more positive affect than those who are not.

30

Hypothesis 9B
FIW affects life satisfaction differently based on how much positive affect a
person experiences such that the negative relationship between FIW and life satisfaction

is weaker for those who are experience more positive affect.

Ii

 

 

 

t: ...............
O ..........................
g """"" > High Positive Affect
U)
’5
m
53—: Low Positive Affect
...1
Family Interference With Work

Figure 8. The proposed interaction between positive affect and life satisfaction.
Hypothesis 10

Type of caregiving affects work withdrawal behaviors such that special needs
caregivers will exhibit more work withdrawal behaviors than those who are not.

Method
Participants

Participants included 119 caregivers; 98 who at a large Midwestern university and
21 who worked elsewhere. All participants worked at least 20 hours per week. An a priori
power analysis indicated that 190 participants were necessary to provide an adequate
amount of power (see Appendix A). Initial sampling strategies including running ads in
newspapers and contacting local support groups for those with Special needs failed to
obtain the number of cases needed to reach the desired level of power. Therefore, another
attempt was made to return to the samples with follow-up contacts, seeking more cases.

These resulted in over 100 cases but still below the desired level. Thus, after several

31

attempts to strengthen power, it was decided to go forward with the existing sample, in
part, because of the fact that special needs caregiving is a low base rate phenomenon and
because of what we felt was the importance of knowing more about the problems
caregivers faced at work, even if the potential for learning about their Situation was
limited by a small sample. Therefore, it was decided to continue with the smaller sample
and take into account its limitations when the data are discussed. A post hoc power
analysis is included in Appendix A, indicating that the size of the effect that would be
detectable in this study was .14. Participants were compensated $10 for their
participation. Characteristics of the sample are listed in Tables 1 and 2.

Table 1

Sample Characteristics: Mean Age, Weekly Hours of Work, Number of Dependent

 

 

Children

Variable Mean SD N
Age 42.81 8.01 116
Hours ofwork 38.89 8.81 111
Number of children dependent on 1.66 1.00 1 10
participant

 

32

Table 2

Sample Characteristics: Race, Gender and Marital Status

 

 

Variable Percentage
Race“
White/Caucasian 90.76%
Aﬁican-American 3.36%
Hispanic 3.3 6%
Asian/Paciﬁc Islander 0.84%
Other 0.84%
Gender“
Male 19.33%
Female ' 79.83%
Marital Status”
Single 2.52%

Single, lives with romantic partner 5.04%

Married 79.83%
Separated 0.84%
Divorced 1 1.76%

 

‘One participant declined to indicate his/her race.

l’One participant declined to indicate his/her gender.
°Percentages do not add up to 100.00% due to rounding error.
dN=113

°N=119

33

Participants were recruited from two samples in order to attempt to capture an
acceptable amount of variance in the caregiving load with which the caregiver was faced.
First, a group of people who cared for chronically ill family members was sampled. This
sample was recruited through four different agencies: a center that provides respite care
for parents of chronically ill children (hereafter referred to as subsample A), a center that
provides assistance to family members of patients who have a chronic illness and have
six months or less to live (hereafter referred to as subsample B), a clinic that provides
care for children with pulmonary disorders (hereafter referred to as subsample C), and a
group of parents that participate on an advisory board that expresses the concerns of
parents with chronically ill children to the Michigan state government, each of whom had
a child of his or her own with a chronic illness (hereafter referred to as subsample D).
The second sample consisted of university employees who cared for at least one
dependent family member (hereafter referred to as subsample E).

Procedure

Participants from subsamples A-D were recruited using a contact person from
each of the aforementioned agencies who sent out a letter to each of the families served
by that agency. The letter introduced the experimenter and the study. With the letter was
included a letter from the experimenter introducing the study (see Appendix B) and a
stamped, self-addressed postcard. The caregivers were instructed to return the postcard to
the experimenter if they were interested in participating. The experimenter sent out a
survey along with a self-addressed business reply return envelope upon receipt of the

postcard. The participants were asked to return the survey within two weeks.

34

The second sample consisted of university employees (subsample E). Nine
hundred and seventy-ﬁve university employees were randomly selected out of the
employee directory in two waves. They received a letter through the campus mail system
describing the study and asking them to participate (see Appendix C). If they did not feel
they ﬁt the qualiﬁcations for the study, they were asked to indicate that. As in the sample
of participants who cared for chronically ill loved ones, the university employees
indicated their agreement to participate (or their failure to qualify for the study) by
returning a self-addressed postcard to the experimenter. Upon receipt of the postcard, the
experimenter sent out the survey along with a self-addressed business reply return
envelope to those who indicated that they ﬁt the qualiﬁcations for participation and were
willing to participate. Participants ‘were instructed to return it within two weeks. Four
hundred and eighty-six potential participants received a follow-up letter with an
additional postcard, reminding them to reply. Participants in this sample who agreed to
ﬁll out the survey received an email reminding them to return the survey every two
weeks for twelve weeks after the survey was initially sent out until either the survey was
returned or until twelve weeks were over. There were 298 people who indicated that they
did not qualify for the study; this left a total of 677 people, 100 of whom participated in
the study, and 577 of whom did not respond at all to the recruitment letter(s). The
response rate for the university employees sample was calculated based on these 677
people. Of the 100 that agreed to participate, 33 indicated that they were the primary or
co-caregiver for a family member with a chronic illness. Of the university employees
who were recruited, two cases were dropped because they did not have any dependents.

Thus, the data from 98 of the people who were recruited through the university were

35

analyzed in this study. The number of people in each subsample who were recruited as
well as the number who actually participated and the response rate are summarized in
Table 3. Of the 100 returns from the university sample, two failed to notice that we asked
that all participants work at least 20 hours per week. These two cases were dropped.
Similarly, of the 16 returns in subsample A, one person was not employed, so that case
was dropped.

Table 3

Size and Response Rate of Subsamples

 

Subsample Recruited Returned Surveys Used in Final Sample Response Rate

 

 

A 93 16 l 5 17.20%
B 3 0 0 0.00%
C 69 2 2 2.90%
D 1 l 4 4 36.36%
E 677 100 98 14.77%
Total 853 122 l 19 14.30%
Measures
Caregiving Assessment

This study sought to explore the nature of family interference with work for those
with a heavy caregiving demand—Special needs caregivers. In order to assure that the
sample contained those who were special needs caregivers and that our measures
captured these behaviors, this type of caregiving was assessed in several ways. First,

participants were selected ﬁ'om groups of people who had family members with chronic

36

illnesses to ensure that a heavy caregiving demand could potentially exist. However, a
large amount of variance existed in how much care a chronically ill person needs,
depending on the illness and how far it had progressed. Therefore, the level of caregiving
in which the participant engaged was measured in a three-part fashion (see Appendix D).
Each participant completed all three parts of this assessment, regardless of whether or not
he/she provided care for a chronically ill loved one. The ﬁrst part of this assessment
process was Simply to ask the participants if they provided care for a chronically ill loved
one (hereafter referred to as type of caregiving; Part I). A second was to inquire how
many unique caregiving behaviors they provided for their loved ones and how often they
provided them (Part 11). Unique caregiving behaviors were deﬁned as providing
assistance with things with which a healthy person the same age would not need help. A
third was to administer a level of caregiving measure, asking the participants how often
they provided assistance to their loved ones with the basic and instrumental activities of
daily life (Part 111). These measures are described in more detail below. Analyses were
performed to determine which of these measures (or combination thereof) best captured
the essence of special needs caregiving and therefore should be used in testing the
hypotheses, and that procedure will be presented after a description of the components of
the caregiving measures.

Part 1: Type of Caregiving. Caregivers were asked to respond to one item
indicating whether or not they provided care for a chronically ill loved one.

Part 11: Unique Caregiving Behaviors. Unique caregiving behaviors were
measured using an instrument constructed for the purposes of this study. Caregivers were

asked to list the kinds of care they provided for their loved ones that were not required by

37

most individuals that were the ages of the loved ones for whom they provided care;
caregivers could list up to 10 such kinds of care. They were then asked to indicate how
often they provided these kinds of care using a 7-point Likert-type scale ranging from
“Never” to “Daily.”

Two scores were derived from this measure. The ﬁrst was the number of unique
behaviors, which was determined by counting the number of unique caregiving behaviors
that caregivers listed, from 0 to 10 (M=2.78, SD=3.56). The second was the frequency of
unique behaviors, which was determined by adding up the frequency ratings given to
each of the behaviors listed (M=16.57, SD=22.3 8). The number of frequency ratings
summed was equal to the number of caregiving behaviors that the caregivers listed. For
example, if the caregiver listed 7 unique caregiving behaviors and rated each of those
behaviors with regard to the frequency with which they performed them, then these 7
frequency ratings would be added together to make up the frequency of unique behaviors
score, for a minimum possible score of 7, and a maximum possible score of 49. If the
caregiver did not list any unique caregiving behaviors, the maximum possible score for
the ﬁ'equency of unique behaviors would be 0, and if the caregiver listed 10 unique
caregiving behaviors, the maximum possible score for the frequency of unique behaviors
would be 70.

Part III: Level of Caregiving. The level of caregiving provided was measured
using a 17-item measure adapted from the Caregiving Hassles Scale (Kinney & Stephens,
1989). The instructions were altered so that caregivers were asked to indicate the
frequency with which behaviors were performed rather than how much the behaviors

were a hassle, and one item was added to assess how often caregivers assisted care

38

recipients with homework or read to the care recipients. This scale asked participants to
indicate the frequency with which they provided assistance with the basic and
instrumental activities of daily life. Participants responded using a 7-point scale ranging
from never to daily. Sample items included, “About how often do you bathe the care-
recipient?” and “About how often do you prepare meals for the care recipient?” Internal
consistency reliability was .90 for the basic activities of daily life (BADL; n=103) and .83
for the instrumental activities of daily life (IADL; n=105). The overall scale had an
internal consistency reliability of .91 (n=99).a The score for this scale and all that follow
was calculated by taking the mean of the items that were answered in each subscale.
Work-Family Conﬂict

The extent to which family responsibilities interfered with work was assessed
with the four-item FIW subscale by Gutek, Searle & Klepa (1991). Participants were .
asked to respond using S-point scales ranging from strongly agree to strongly disagree to
items such as “After work, I come home too tired to do some of the things I’d like to do,”
and “My personal life takes up time that I’d like to spend at work.” The alpha coefﬁcient
was .71 for the FIW items (see Appendix F; n=119).
Job Satisfaction

Job satisfaction was measured by the Job Descriptive Index (Bowling Green State

University, 1997). This measure asked participants to indicate their level of satisfaction

 

' Additional caregiving items were administered in addition to these that were not included in the analyses.
The complete measure that was administered, including all of the items that were analyzed as well as those
that were not (which appear in boldface), is located in Appendix E. Footnotes at the end of the appendix

indicate the rationale for not analyzing the items that were excluded.

39

(satisﬁed, dissatisﬁed, or not sure) with various aspects of six dimensions of their job,
including (1) work on the present job, (2) present pay, (3) opportunities for promotion,
(4) supervision, (5) co-workers, and (6) the job in general (see Appendix G). Coefﬁcient
alphas for those subscales were .88, .83, .84, .87, .88, and .86, based on 108, 113, 115,
111, and 109 cases respectively. Only the job in general scale was used here because all
of the hypotheses were for overall job satisfaction and were not broken down by
subscales.
Life Satisfaction

Life satisfaction was measured by the Satisfaction With Life Scale (Diener,
1985), a 5-item scale that asked participants to respond using a 7-point scale ranging
from strongly disagree to strongly agree to questions such as, “In most ways, my life is
close to my ideal,” and “I am satisﬁed with my life.” The coefﬁcient alpha for this scale
was .89 (see Appendix H; n=119).
Work Withdrawal Behaviors

Withdrawal behaviors were measured with an 18-item scale by Roznowksi and
Hanisch (1990). Participants responded using a 7-point scale ranging from daily to never,
indicating how often they engaged in certain behaviors or thoughts. The participants
responded to statements such as, “It is easy for me to be on time for work,” and “I would
like to quit my job.” The alpha coefﬁcient for this scale was .77 (see Appendix I; n=93).
Standards of Acceptability

Standards of acceptability, deﬁned as the number of work-related hassles
acceptable in a job, were measured using the work hassles subscale of the Daily Hassles

Scale—Revised (Holm & Holroyd, 1992). This scale asked participants to respond to

.40

nine potential work hassles (e. g., job dissatisfaction, problems with employees) using the
following 6-point scale: 0 = “did not occur”; 1 = “occurred, not severe”; 2 = “occurred,
somewhat severe”; 3 = “occurred, moderately severe”; 4 = occurred, very severe”; 5 =
“occurred, extremely severe.” The alpha coefﬁcient for this scale was .63 (see Appendix
J; n=1 18).
Coworker Social Support

Coworker social support was measured using a scale written by Ducharme and
Martin (2000). This IO-item scale asked participants to indicate their agreement with
statements about both affective support, such as “My coworkers really care about me”
and instrumental support, such as “My coworkers will pitch in and help.” Participants
used a 5-point scale to respond to these questions. The coefﬁcient alpha for the affective
support scale was .93 (n=116), .87 (n=l 18) for the instrumental support scale, and .93 for
the overall scale (see Appendix K; n=115). The overall scale was used in the analyses
because the hypotheses were regarding the effects of coworker social support in general
rather than the effects of a Speciﬁc dimension of coworker social support.
Work as an Escape

Work as an escape was measured using a scale designed for the purposes of this
study. Participants used a 7-point scale to respond to 4 items such as “I enjoy going to
work because it allows me to spend some time away from the stresses of home,” and “My
time at work provides an escape from the stresses of home (see Appendix L).” The alpha

for this scale was .86 (n=119).

41

Resilience

Resilience was measured using the Resilience Inventory (W agnild & Young
(1993). Participants used a 7-point scale to respond to 25 items such as “I usually take
things in stride,” and “I am determined.” The coefﬁcient alpha for this scale was .88 (see
Appendix M; n=l 16).
Aﬂect

Affect was measured using the Positive and Negative Affect Schedule (PANAS;
Watson, Clark, & Tellegan, 1988). Participants used a 5-point scale to indicate how often
they experience 20 different feelings and emotions (e.g., interested, distressed, excited,
distressed). Coefficients alphas were .90 for positive affect (n=112) and .88 for negative
affect (see Appendix N; n=112). Caregiving can make caregivers feel good about
themselves and evoke positive feelings, and the hypotheses suggest that these positive
feelings may improve job and life satisfaction. It is possible that caregiving could evoke
negative feelings, but it is unlikely that these negative feelings would improve job or life
satisfaction. Thus, the negative affectivity data were not used.
Demographics

Finally, participants were asked to provide information about their living
arrangement, dependents, gender, race, marital status, income, and employment. These
appear in Appendix 0.
Order of Administration

The measures were administered in the following order: job satisfaction, life

satisfaction, work withdrawal, work-family conﬂict, caregiver time demands, level of

42

caregiving, unique caregiving behaviors, coworker social support, work as an escape,
work hassles, resilience, affect, demographics.
Results

Analyses were completed in two steps. The ﬁrst set of analyses was used to
determine what appeared to be the most appropriate way to assess special needs
caregiving. Second, the hypotheses presented earlier were tested using regression
methodology to see if there are main effects, moderation, and/or mediation. Because it
was necessary to determine which caregiving measures should be used in the analyses in
order to conduct them properly, I will begin by presenting these analyses.

Caregiving Assessment Analysis

In order to address the issue of which caregiving measure (or combination
thereof) should be used to index Special needs caregiving, a factor analysis was
conducted. The 17 items from the Level of Caregiving Scale, the Type of Caregiving
score, and the two scores from the Unique Caregiving Behaviors Scale were analyzed
using exploratory factor analysis with a maximum likelihood extraction followed by a
varimax rotation and Kaiser normalization. The scree test criterion indicated that a four or
ﬁve factor solution was most likely to be appropriate. The Kaiser criterion of considering
all factors that have eigenvalues of greater than 1.00 indicated that four factors should be
considered. However, when an examination of the factor loadings of a four-factor
solution rotated to a varimax criterion indicated that no item loaded highest on the fourth
factor, and since the three-factor solution appeared reasonably interpretable, three factors

were used.

43

Examining the three-factor solution and considering only the items that had
loadings of at least .50 on a factor revealed a fairly clear factor solution (see Table 4).
The ﬁrst factor includes items that describe basic activities of daily life; these are the
hands-on, fundamental activities of life that most healthy adults and children can perform
for themselves, including things like walking, toileting, bathing, moving from one place
to the next, eating, and dressing. The second factor includes items that describe
instrumental activities of daily life; these are the activities of life that healthy adults and
children often have others perform for them to some extent, including things like laundry,
picking up the house, preparing meals, and assistance with homework or paperwork. The
third factor describes things that would contribute to the Sheer volume of caregiving,
including the type of caregiving and both of the unique caregiving behavior scores. Five

items did not load clearly higher on one dimension than another.

Table 4

Caregiving Factors with Item Loadings

 

 

Items Factors 1 II III
Assist care-recipient with walking .70 -.04 .32
Assist with care-recipient’s toileting .77 .26 .28
Bathe care-recipient .81 .37 .07
Assist care-recipient with exercises/therapy .51 .22 .32
Lift or transfer care-recipient .71 .07 .07
Help care-recipient eat .73 .20 .24
Dress care-recipient .83 .37 .20
Provide care-recipient assistance during the night .72 .28 .16
Do care-recipient’s laundry .16 .78 .27
Pick up after care-recipient .31 .76 .04
Preparing meals for care-recipient .04 .63 -.02
Provide daytime supervision to care-recipient .31 .55 .07
Type of caregiving .14 -.16 .59
Number of unique behaviors .20 .12 .95
Frequency of unique behaviors .29 .22 .93
Assisting care-recipient with health aids (e. g., braces) .27 .27 .41
Transport care-recipient to doctor/other places .39 .21 .17
Giving medications to care-recipient .43 .33 .40
Read to care-recipient/help with homework or paperwork .12 .47 .00
Experience extra expenses due to caﬁgiving .37 .43 .17

 

 

46

Caregiving scores that were used in the regressions were calculated. The ﬁrst
caregiving score was calculated using the items that loaded highly on the ﬁrst factor. This
score, which will be hereafter referred to as “extreme behaviors,” was calculated by
taking the mean of all of the variables that loaded highly on Factor I. This was
determined to be a good proxy of special needs caregiving not only because it was the
ﬁrst and strongest factor that came out of the factor analysis but also because it
encompassed the activities that special needs caregivers perform that are quite hands-on
and can usually be performed by healthy children and adults for themselves. Coefﬁcient
alpha for this score was .93 (n=102).

The second caregiving score, which will hereafter be referred to as “overall
caregiving,” was calculated by taking the mean of all of the variables that initially entered
into the factor analysis. This was determined to be a good proxy of special needs
caregiving because it combined all of the aspects that were theorized to be part of special
needs caregiving and because the items had intercorrelations sufﬁcient enough to produce
a coefﬁcient alpha of at least .70. Coefﬁcient alpha for this score was .70 (n=99).

In order to assess the validity of these scores, a t-test was conducted to compare
the extreme behaviors and overall caregiving between those who indicated that they
provided care for a chronically ill loved one and those who indicated that they do not (see
Table 5). It was expected that there would be a signiﬁcant difference in the means for
extreme behaviors and overall caregiving between these two groups. The results indicate
that those who provide care for a chronically ill loved one provide signiﬁcantly more
overall caregiving than those who do not (t:-5.84, p<.01; d=1.01). Those who provided

care for a chronically ill loved one did not engage in more extreme behaviors than those

47

who do not (t=-1.35, p=.18; d=.25). Yet Since this was the most direct measure of Special
needs caregiving and since it was a self nomination, it was decided that the hypotheses
involving caregiving will each be tested twice: once with extreme behaviors as a proxy
for caregiving, and once with overall caregiving as a proxy for caregiving.

Table 5

Extreme Behaviors and Overall Caregiving Descriptives by Type of Caregiving

 

Special Needs Caregivers Caregivers of Healthy Children

 

Variable Mean SD N Mean SD N d
Extreme behaviors 3.49 2.15 54 2.96 2.1 1 65 .25
Overall caregivinga 5.44 2.52 54 3.47 1.32 65 1.01

 

" The difference between the means for the two groups is signiﬁcant at the p<.01 level.

In order to determine whether age, hours of work, or number of children varied
systematically with either of the caregiving variables, a correlation was conducted (see
Table 6). Extreme behaviors and age were signiﬁcantly and negatively correlated (r=-.42,
p<.01), indicating that those who are younger engage in more extreme behaviors. This
correlation is somewhat problematic. Age correlates negatively with extreme behaviors,
but that relationship may be caused by the fact that people who are younger tend to have
younger children. Younger children have more extreme needs; it is normal for them to
need assistance with walking, toileting, or eating. Thus, younger people may engage in
these extreme caregiving behaviors as a byproduct of the age of their children, not their
involvement in the special needs caregiving role. In order address this issue, I controlled
for the relationship between age and extreme behaviors in the analyses by entering age

into the ﬁrst step of any regression involving extreme behaviors.

48

Table 6
Means and Standard Deviations for and Correlations Between Age, Hours of Work,

Number of Children, and Caregiving Variables

 

 

Variable Mean SD N 1 2 3 4 5
1 Age 42.8 8.01 116 1.00
2 Hours ofwork 38.89 8.81 111 .00 1.00

3 Numberofchildren 1.65 1.00 117 -.18 -.15 1.00
4 Extremebehaviors 3.21 2.14 118 -.42* -.16 .08 1.00

5 Overallcaregiving 4.36 2.18 119 -.12 -.18 .10 .75* 1.00

 

‘p<.01

Table 7 provides the means, standard deviations, and intercorrelations among all

of the variables used in this study.

49

SVQ: .moVQn

 

dd.— :Rd. Snead. 2d nummd- 2d. 2.9th newmd ne—mdi ew_d.. nuemd mod wdd. m: mod no.“ b338b< gumwoZﬂ
dd; .3th mod 8.. 3d 33d 2 d Ed- Snead :d nmmd. bod odd. m : dud dmd bm>woob< o>£mom§

84 mod .3..de 2..de d— d nod. a... _ md Ed 2 d- odd mod. a: cod new ouuoamom:

do; «Ed 2..de odd. odd. *3 d. emmd nnmmd .3de «ind a: 34 mod 38mm .8 mm 3.532

dd; aimed «awmd nebmdi :vmd .3..de ammo- :d Cd a: Ned add toaqamﬁmoom BxuoBovgeoa—Bbmsd

do; eeemduevmd. eenNd :Nvd 2d. 2d 2d a: dud ems.” :39m38m goo—3300 grout/xx

 

cog aided. «2d nemmd «ammd. mod- Nod. w: Rd Sid hum—BSQoooado Endgame
do; tend- 33.? said ddd nod- w: ewd end 3355;» {03¢
do; :VNd nevmd. 2d- 3d a: me; :6 nouoﬂmuam 83m
do; :vmd. add :d a: 3d 3mm eouoﬂmaam no?
8; 33d :d a: dud mm.~ 2mm
do; aimed a: wﬁm end wﬁﬁwouao =8o>ON
do; w: EN _N.m 303305 vacuum:
2 S S 2 d w e c m v m N ~ Z Om =32 oimtmxw

 

 

Beam. 2: S $33.33 “Salem méaomwdam. he we Behﬁmhoeamds LEE .weehSSQ naewzeuwlmlaem:

N. 055.

50

Tests of Hypotheses
The tests of the hypotheses will be presented in numerical order.
Hypothesis 1
The ﬁrst hypothesis predicted that as F IW increases, job satisfaction decreases.
Linear regression analyses indicate that a signiﬁcant relationship does exist (B=-.24,
p<.01; see Table 8). Hypothesis 1 was strongly supported.
Table 8

Regression Relating F] W to Job Satisfaction

 

 

 

Variable a Adj. R2 AR2
FIW —.24* .05* .O6*
‘p<.01

Note: n=l 19. B is the standardized regression coefﬁcient. Increments for variables entered at the R2
signiﬁcance levels re based on F tests for that step.

Hypothesis 2

Hypothesis 2 stated that there would be an interaction between FIW and type of
caregiving such that the relationship between FIW and job satisfaction is weaker for
those who are Special needs caregiver than for those who are not. This hypothesis was
tested twice: ﬁrst with extreme behaviors as a proxy for caregiving, and second with
overall caregivers as a proxy for caregiving.

Extreme Behaviors. To test this hypothesis, a three-step hierarchical regression
was performed, with age in the ﬁrst step, FIW and extreme behaviors in the second step,
and the interaction between F IW and the extreme behaviors in the third step. The beta

coefﬁcient for FIW was signiﬁcant (B=—.33, p<.05), but the beta coefﬁcients for age

51

(B=.08, ns), extreme behaviors (B=-.10, ns), and the interaction between FIW and
extreme behaviors (B=.37, ns) were not (see Table 9). Because the interaction did not
account for signiﬁcantly more variance than FIW and extreme behaviors alone,
Hypothesis 2 was not supported.

Table 9

Regression Testing the Hypothesis that Extreme Behaviors Moderates the Relationship

Between FI W and Job Satisfaction

 

Variable a Adj.R2 AR2

 

Step 1

Age .08 -.01 .00
Step 2

F IW -.33*

Extreme Behaviors -.10 .03 .09*
Step 3

F IW x Extreme Behaviors .37 .04 .00

 

‘p<.05

Note: n=115. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. To test this hypothesis, a two-step hierarchical regression
was performed, with FIW and overall caregiving in the ﬁrst step and the interaction
between the two in the second step. Only the beta coefﬁcient for FIW (B=—.46, p<.05),
was signiﬁcant; the beta coefﬁcients for overall caregiving(B=-.21, ns), and the

interaction between the two (B=.48, ns) were not. Since the interaction did not account

52

for Signiﬁcantly more variance than FIW and type of caregiving did alone, Hypothesis 2
was not supported (see Table 10).

Table 10

Regression Testing the Hypothesis that Overall Caregiving Moderates the Relationship

Between F I W and Job Satisfaction

 

Variable p Ade2 AR2

 

Step 1

FIW -.46*

Overall Caregiving -.21 .02 .04
Step 2

FIW x Overall Caregiving .48 .02 ‘ .01

 

‘p<.05
Note: n=l 19. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.
Hypothesis 3A

Hypothesis 3A suggested that type of caregiving will affect the standards of
acceptability such that those who are special needs caregivers will have lower standards
of acceptability than those who are not. This hypothesis will be tested twice; once with
extreme behaviors as a proxy for special needs caregiving, and once as overall caregiving
as a proxy for special needs caregiving. Hypotheses 4A, 5A, 8A, and 9A will be tested
and presented in this format as well.

Extreme Behaviors. To test this hypothesis, a linear regression analysis was

performed. The results indicate that after controlling for age, the relationship is not

53

Signiﬁcant (B=.00, ns; see Table II). Hypothesis 3A was not supported using extreme
behaviors as a proxy for special needs caregiving.
Table 11

Regression Relating Extreme Behaviors to Standards of Acceptability

 

 

Variable p Adj. R2 AR2
Step 1

Age .02 -.01 .00
Step 2

Extreme Behaviors .00 . -.02 .00

 

Note: n=117. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. A linear regression analysis was conducted to test this
hypothesis. The results indicate that the relationship is not signiﬁcant (B=—.01 , ns; see
Table 12). Therefore, Hypothesis 3A was not supported using overall caregiving as a
proxy for special needs caregiving.

Table 12

Regression Relating Overall Caregiving to Standards of Acceptability

 

Variable a Adj. R2 AR2

 

Overall Caregiving -.01 -.01 .00

 

Note: n=118. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

54

Hypothesis 3B

Hierarchical regression methodology was used to determine whether standards of
acceptability moderated the relationship between FIW and job satisfaction. FIW and
standards of acceptability were entered into the ﬁrst step, and the interaction between the
two was entered into the second step. None of the beta coefﬁcients for FIW (B=-.28, ns),
standards of acceptability (B=.22, ns), or the interaction between the two (B=.24, ns) was
signiﬁcant; since the interaction was not signiﬁcant, standards of acceptability does not
moderate the relationship between FIW and job satisfaction. The results of the regression
are presented in Table 13.
Table 13
Regression Determining Whether Standards of Acceptability Moderates the Relationship

Between F1 W and Job Satisfaction

 

Variable p Ade2 AR2

 

Step 1

FIW -.28

Standards of Acceptability .22 .11* .11*
Step 2

FIW X Standards of Acceptability .24 .l 1* .00

 

*p<.01

Note: n=l 18. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

55

Hypothesis 4A

Hypothesis 4A said that type of caregiving affects coworker social support such
that those who are Special needs caregivers experience more coworker social support than
those who are not.

Extreme Behaviors. To test this hypothesis, 3 linear regression analysis was
performed. The results indicate that after controlling for age, the relationship is
signiﬁcant (B=.23, p<.05; see Table 14). Hypothesis 4A was supported using extreme
behaviors as a proxy for special needs caregiving.

Table 14

Regression Relating Extreme Behaviors to Coworker Social Support

 

 

 

Variable a Adj. R2 AR2
Step 1
Age .23* .01 .02*
Step 2
Extreme Behaviors .23* .05 .04*
*p<.05

Note: n=118. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. A linear regression analysis was conducted to test this
hypothesis. The results indicate that the relationship is not signiﬁcant (B=.17, ns; see
Table 15). Therefore, Hypothesis 4A was not supported using overall caregiving as a

proxy for special needs caregiving.

56

Table 15

Regression Relating Overall Caregiving to Coworker Social Support

 

Variable a Adj. R2 AR2

 

Overall Caregiving .17 .02 .03

 

Note: n=118. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.
Hypothesis 4B

This hypothesis was tested using a two-step hierarchical regression in order to
determine whether coworker social support moderated the relationship between FIW and
job satisfaction. F IW and coworker social support were entered into the ﬁrst step, and the
interaction between the two was entered into the second step. The betas associated with
FIW (B=.29, ns) and the interaction between FIW and coworker social support (B=—.39,
ns) are not signiﬁcant; only the beta associated with coworker social support was
signiﬁcant (B=.59, p<.01). Since the interaction was not signiﬁcant, coworker social
support does not moderate the relationship between FIW and job satisfaction. The results

of the regression are presented in Table 16.

57

Table 16

Regression Determining Whether Coworker Social Support Moderates the Relationship

Between F1 W and Job Satisfaction

 

 

Variable p Adj.R2 AR2
Step 1

FIW .29

Coworker Social Support .59* .15** .16"
Step 2

FIW X Coworker Social Support -.39 .15** .00

 

*p<.05, "p<.01

Note: n=1 19. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Hypothesis 5A

Hypothesis 5A said that type of caregiving affects the experience of work as an

escape such that those who are Special needs caregivers are likely to view work as an

escape from troubles in their non-work life.

Extreme Behaviors. To test this hypothesis, a linear regression analysis was

performed. The results indicate that the relationship is signiﬁcant (B=.36, p<.01; see

Table 17). Hypothesis 5A was supported using extreme behaviors as a proxy for special

needs caregiving.

58

Table 17

Regression Relating Extreme Behaviors to Work as an Escape

 

 

 

Variable a Adj. R2 AR2
Step 1

Age .13 -.Ol .00
Step 2

Extreme Behaviors .36* .09* .1 1*
*p<.01

Note: n=117. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. A linear regression analysis was conducted to test this
hypothesis. The results indicate that the relationship is signiﬁcant (B=.43, p<.01; see
Table 18). Therefore, Hypothesis 5A was not supported using overall caregiving as a
proxy for special needs caregiving.

Table 18

Regression Relating Overall Caregiving to Work as an Escape

 

 

 

Variable p Adj. R2 AR2
Overall Caregiving .43 * .18 .18
*p<.01

Note: n=118. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

59

Hypothesis 5B

A two-step hierarchical regression was conducted to determine whether work as
an escape moderated the relationship between FIW and job satisfaction. FIW and work as
an escape were entered into the ﬁrst step, and the interaction between the two was entered
into the second step. None of the betas associated with FIW (B=-.55, ns), work as an
escape (B=.04, ns), or the interaction between the two (B=.47, ns) was signiﬁcant. Work
as an escape does not moderate the relationship between FIW and job satisfaction. The
results of the regression are presented in Table 19.
Table 19
Regression Determining Whether Work as an Escape Moderates the Relationship

Between F1 W and Job Satisfaction

Variable a Adj.R2 AR2

 

Step 1

FIW -.55

Work as an Escape .04 .08* .11*
Step 2

FIW X Work as an Escape .47 .08* .00

 

*p<.01
Note: n=118. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

60

Hypothesis 6

The sixth hypothesis predicted that as F IW increases, life satisfaction decreases.
Linear regression analyses provided evidence in support of this hypothesis (B=-.34,
p<.01; See Table 20). Hypothesis 6 was strongly supported.
Table 20

Regression Relating F1 W to Life Satisfaction

 

 

 

Variable [3 Adj. R2 ARz
FIW -.34* .11* .12*
*p<.01

Note: n=l 19. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.
Hypothesis 7

Hypothesis 7 stated that there was an interaction between FIW and type of
caregiving such that the relationship between FIW and life satisfaction is weaker for
those who are Special needs caregiver than for those who are not. This hypothesis will be
tested twice: ﬁrst with extreme behaviors as a proxy for caregiving, and second with
overall caregivers as a proxy for caregiving.

Extreme Behaviors. To test this hypothesis, a three-step hierarchical regression
was performed, with age in the ﬁrst step, FIW and extreme behaviors in the second step,
and the interaction between FIW and the extreme behaviors in the third step. None of the

beta coefﬁcients for FIW (B=-.26, ns), age (B=-.12, ns), extreme behaviors (B=.19, ns), or

the interaction between FIW and extreme behaviors (B=-.21, ns) was signiﬁcant (see

61

Table 21). Because the interaction did not account for signiﬁcantly more variance than
FIW and extreme behaviors alone, Hypothesis 7 was not supported.

Table 21

Regression Testing the Hypothesis that Extreme Behaviors Moderates the Relationship

Between F1 W and Life Satisfaction

 

Variable a Adj.R2 AR2

 

Step 1

Age -.12 .00 .01
Step 2

FIW -.26

Extreme Behaviors .19 .1 1* .12*
Step 3

F IW x Extreme Behaviors -.21 .l 1* .00

 

‘p<.01

Note: n=115. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. A two-step hierarchical regression was used to test this
hypothesis. FIW and overall caregiving were entered into the ﬁrst step, and the
interaction between the two was entered into the second step. None of the beta
coefﬁcients for F IW (B=—.04, ns), overall caregiving(B=.31, ns), or the interaction
between the two (B=-.55, ns) was signiﬁcant. The interaction between FIW and overall
caregiving did not account for signiﬁcantly more variance than either FIW or overall

caregiving did alone; therefore Hypothesis 7 was not supported (see Table 22).

62

Table 22
Regression Testing the Hypothesis that Overall Caregiving Moderates the Relationship

Between F] W and Life Satisfaction

 

Variable a Adj.R2 AR2

 

Step 1

FIW -.04*

Overall Caregiving .31 .11** .12**
Step 2

FIW x Overall Caregiving -.55 .12** .02

 

‘p<.05, "p<.01
Note: n=119. B is the standardized regression coefﬁcient Increments for variables entered at the R2
signiﬁcance levels are based on F tests for that step.

Hypothesis 8A

Hypothesis 8A proposed that the type of caregiving in which someone is involved
affects resilience such that those who are special needs caregivers have more resilience
than those who are not.

Extreme Behaviors. A linear regression was conducted in order to test this
hypothesis. Results indicated that the proposed relationship was not signiﬁcant (B=.04,
ns; see Table 23). Therefore, this hypothesis was not supported when extreme behaviors

was used as a proxy for special needs caregiving.

63

Table 23

Regression Relating Extreme Behaviors to Resilience

 

 

 

Variable a Adj. R2 AR2
Step 1

Age .25 .05 .06*
Step 2

Extreme Behaviors .03 .04 .00
‘p<.01

Note: n=l 18. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. This hypothesis was tested using linear regression
methodology. Results indicated that the proposed relationship was not signiﬁcant. (B=.06,
ns; see Table 24). Therefore, this hypothesis was not supported when overall caregiving
was used as a proxy for special needs caregiving.

Table 24

Regression Relating Overall Caregiving to Resilience

 

Variable ‘ [3 Adj. R2 AR2

 

Overall Caregiving .06 -.01 .00

 

Note: n=119. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

64

Hypothesis 8B

This hypothesis was tested using a two-step hierarchical regression to determine
whether resilience moderated the relationship between F [W and life satisfaction. F IW and
resilience were entered into the ﬁrst step, and the interaction between the two was entered
into the second step. None of the betas associated with FIW (B=—.30, ns), resilience
(B=.27, ns), or the interaction between the two (B=-.01, ns) “was Signiﬁcant; since the
interaction was not signiﬁcant, resilience does not moderate the relationship between
FIW and life satisfaction. The results of the regression are presented in Table 25.
Table 25
Regression Determining Whether Resilience Moderates the Relationship Between FI W

and Life Satisfaction

 

Variable p Adj.R2 AR2

 

Step 1

FIW -.3O

Resilience .27 .I7* .19*
Step 2

FIW X Resilience -.01 .17* .00

 

*p<.01
Note: n=1 19. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

65

Hypothesis 9A
Hypothesis 9A said that type of caregiving affects positive affect such that those
who are special needs caregivers experience more positive affect than those who are not.
Extreme Behaviors. A linear regression was conducted in order to test this
hypothesis. Results indicated that the proposed relationship was not signiﬁcant (B=-.06,
ns; see Table 26). Therefore, when extreme behaviors was used as a proxy for special
needs caregiving, this hypothesis was not supported.
Table 26

Regression Relating Extreme Behaviors to Positive Aﬂect

 

 

 

Variable a » Adj. R2 AR2
Step 1

Age .25 .04 .05*
Step 2

Extreme Behaviors .05 .04 .00
*p<.05

Note: n=118. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. This hypothesis was tested using linear regression
methodology. Results indicated that the proposed relationship was not Signiﬁcant. (B=.O7,
ns; see Table 27). Therefore, this hypothesis was not supported when overall caregiving

was used as a proxy for special needs caregiving.

66

Table 27

Regression Relating Overall Caregiving to Positive Affect

 

Variable p Adj. R2 AR2

 

Overall Caregiving .07 -.01 .00

 

Note: n=l 19. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.
Hypothesis 9B

A two-step hierarchical regression was conducted to determine whether positive
affect moderated the relationship between FIW and life satisfaction. FIW and positive
affect were entered into the ﬁrst step, and the interaction between the two was entered
into the second step. None of the betas associated with F [W (B=—.02, ns), positive affect
(B=-.47, ns), or the interaction between the two (B=—.28, ns) was signiﬁcant; since the
interaction was not signiﬁcant, positive affect does not moderate the relationship between

F IW and life satisfaction. The results of the regression are presented in Table 28.

67

Table 28
Regression Determining Whether Positive Aﬁfect Moderates the Relationship Between

F1 W and Life Satisfaction

 

 

Variable p Adj.R2 AR2

 

 

Step 1

FIW -.02

Positive Affect .35 .11* .13*
Step 2

FIW X Positive Affect -.28 .1 1* .00

 

*p<.01
Note: n=119. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.
Hypothesis 10

The tenth hypothesis predicted that as FIW increases, work withdrawal increases.
Linear regression analysis found support for this hypothesis (B=.4l, p<.01; See Table

29). Hypothesis 10 was strongly supported.

68

Table 29

Regression Relating F1 W to Work Withdrawal

 

 

 

Variable a Adj. R2 AR2
FIW .41* .16* .17*
‘p<.01

Note: n=118. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.
Post Hoc Analyses
Negative A ﬂectivi ty

Although negative affectivity was not part of the proposed model, it seemed likely
to have some relationships with the variables of interest here. Therefore, regression
analyses were conducted to determine whether type of caregiving affects negative affect
such that those who are special needs caregivers experience more or less negative affect
than those who are not as well as whether negative affect moderates the relationship
between F IW and life satisfaction.

Extreme Behaviors. A linear regression was conducted in order to test whether
type of caregiving might affect negative affectivity when extreme caregiving was used as
a proxy for special needs caregiving. Results indicated that the proposed relationship was
not Signiﬁcant (B=-.08, ns; see Table 30). Therefore, when extreme behaviors was used
as a proxy for special needs caregiving, negative affectivity does not relate to Special

needs caregiving.

69

Table 30

Regression Relating Extreme Behaviors to Negative Aﬂect

 

 

 

Variable p Adj. R2 AR2
Step 1

Age -.01 -.Ol .00
Step 2

Extreme Behaviors -.08 -.01 .00
‘p<.05

Note: n=1 1 l. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

Overall Caregiving. A linear regression was conducted in order to test whether
type of caregiving might affect negative affectivity when overall caregiving was used as a
proxy for Special needs caregiving. Results indicated that the proposed relationship was
not signiﬁcant. (B=.03, ns; see Table 31). Therefore, negative affectivity does not relate
to special needs caregiving when overall caregiving was used as a proxy for special needs
caregiving.
Table 31

Regression Relating Overall Caregiving to Negative Aﬂect

 

 

Variable a Adj. R2 ARZ

Overall Caregiving .03 -.Ol .00

 

Note: n=l 14. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.

70

A two-step hierarchical regression was conducted to determine whether positive
affect moderated the relationship between F IW and life satisfaction. F IW and positive
affect were entered into the ﬁrst step, and the interaction between the two was entered
into the second step. Only the betas associated with FIW (B=—.64, p<.05) and positive
affect (B=-.70, p<.05) were signiﬁcant. Since the interaction between the two was not
Signiﬁcant (B=—.72, ns), negative affect does not moderate the relationship between FIW
and life satisfaction. The results of the regression are presented in Table 32.

Table 32
Regression Determining Whether Negative Aﬁ’ect Moderates the Relationship Between

F] W and Life Satisfaction

 

Variable a Adj.R2 AR2

 

Step 1

FIW -.64*

Negative Affect -.70* .13** .15**
Step 2

FIW X Negative Affect .72 .14** .00

 

*p<.01
Note: n=114. B is the standardized regression coefﬁcient. Increments for variables entered at the R2

signiﬁcance levels are based on F tests for that step.
Relationship Between FI W and JDI Dimensions

Although the model did not Specify any hypothesized relationships between F 1W
and the separate dimensions of the Job Descriptive Inventory (J DI), it seemed interesting

to explore these relationships to determine whether any particular dimensions of job

71

satisfaction are reduced more or less than others by F IW . A correlation was conducted to
determine the relationship between the J DI dimensions and F IW (see Table 33).
Table 33

Intercorrelations Between JDI Job Satisfaction Dimensions and F1 W

 

 

 

Variable N F 1W
Work 119 -.10
Pay 119 -.08
Opportunities for Promotion 119 -. l 3 -
Supervisors l 19 -.27**
Coworkers 1 l8 -.1 8*
JobinGeneral 119 -.14
*p<.01

The correlation between F 1W and satisfaction with supervisors is signiﬁcant (r=-
.27, p<.01), as is the correlation between F IW and satisfaction with coworkers (r=-.l 8,
p<.01). These relationships exist because the amount of FIW that people experience can
indirectly impact their satisfaction with their supervisors and coworkers. For instance,
people who experience more FIW exhibit more withdrawal behaviors (r=.4l , p<.01).
These withdrawal behaviors include behaviors such as being late to work and being
absent from work. Employees may feel that supervisors and coworkers are evaluating
them negatively because of these behaviors. The evaluation literature has demonstrated
that a person’s perception that an evaluator will rate him/her poorly negatively affects the

person’s satisfaction with the evaluator (Chambers & Schmitt, 2002). According to this

72

position, when the person perceives that the evaluator will evaluate him/her negatively,
the person will be less satisﬁed with the evaluator and in turn, will tend to rate the
evaluator negatively The fact that withdrawal correlates negatively and signiﬁcantly with
both satisfaction with one’s supervisor (r=-.34, p<.01) and satisfaction with one’s
coworker (r=-.27, p<.01) lends support to the argument that this might be occurring.
Thus, it seems likely that people who experience more FIW exhibit more withdrawal
behaviors, and as a result, they feel that they are being negatively evaluated by their
coworkers and supervisors, in turn negatively impacting their satisfaction with their
coworkers and supervisors.

Discussion

The main goal of this study was to examine the relationship between family
interference with work and its outcomes in order to see if those relationships differed
occurred to one’s caregiving status. Overall, this study did not detect any differences in
job and life satisfaction based on caregiving status. However, support was found for the
hypotheses regarding the negative relationships between F IW and life satisfaction and job
satisfaction as well as the positive relationship with, and work withdrawal. This is
consistent with a multitude of previous research on the effects of family interference with
work (e.g., Kossek & Ozeki, 1998; Lee, 1997).

Overall caregiving and extreme types of caregiving behaviors were also found to
be positively related to work as an escape. That is, those who had greater caregiving
demands appreciated the opportunity to escape the pressures and stresses of home life by
going to work more than those who had lower caregiving demands. The construct of

work as an escape has not been explored in the literature, and the measure used in this

73

study was original. Thus, the fact that the measure had good reliability and a positive
relationship with caregiving demands was detected was a meaningful contribution of this
study.

Even though support for some of the hypothesized relationships was found, it is
possible that this support is merely an artifact of response-response bias. Response-
response bias is the tendency for two variables collected through self-report surveys to
correlate simply because they were collected through the same self-report methodology.
Thus, it is possible that even though the measures of these variables were related, the
constructs were not. If support were found for the moderated relationships in this thesis,
that would help to mitigate this concern because moderated relationships are less likely to
fall victim to response-response bias than main effect relationships. However, the
moderated relationships were not supported. Thus, this is a difﬁculty. However, some of
the main effect hypotheses that received support in this literature replicated relationships
that have been established in the literature. In addition, the hypothesized relationships
were supported by theory, the measures all had good reliabilities, and with the exception
of the standards of acceptability measure, all were published in previous studies and have
good validity evidence. Thus, we have every reason to believe that the response-response
bias was not the sole cause of the signiﬁcant relationships found.

There are several plausible reasons why differences in the relationship between
FIW and its outcomes according to caregiving demands were not detected. The ﬁrst has
to do with the selection of participants. Of the 176 participants in subsamples A-D who
were recruited because it was thought likely that they would have a higher caregiving

demand than normal and thus score highly on caregiving measures, only 21 participated,

74

for a resulting response rate of 11.93%. This response rate is low, leading one to wonder
how those that responded to the survey may have differed from those that did not.
Looking at the pattern of correlations between extreme behaviors and overall caregiving
and the rest of the self-report variables in the data set (see Table 7), it appears that there
was not a strong pattern of relationships between the two caregiving variables and any of
the other variables except for work as an escape. In this sample, those with high
caregiving demands did not differ signiﬁcantly from those with low caregiving demands
on any of the variables of interest other than work as an escape.

In addition, the majority of the responses to the overall caregiving and extreme
behaviors scales were at the low end of the scales; the median score for extreme
behaviors was 2.25 on a scale from 1 to 7, and the median score for overall caregiving
was 3.89 on a scale from .9 to 10.05. This, in addition to the pattern of correlations, leads
to the conclusion that perhaps those who were recruited ﬁom subsamples A-D and did
not respond had heavier caregiving demands than anyone who did.

It seems likely that people who have high caregiving demands and must devote a
lot of time and energy to their caregiving might not feel like they wanted to devote any of
their precious time to ﬁlling out a survey and thus chose not to participate in the study.
One possible reason for this is that the data collection occurred during the summer; it is
likely that special needs caregivers with school-aged children have higher caregiving
demands during the summer months because their children are not at school during the
day and thus require more attention. If data collection had occurred during the school
year, these caregivers may have had time to participate in the study, thus increasing the

variance on the caregiving variables.

75

Another plausible explanation for our failure to capture those with high
caregiving demands as part of our sample may have to do with the fact that the people
who were recruited because we hoped that they would have high caregiving demands
were recruited through agencies that provide support for caregivers with care recipients
who have a chronic illness or disability. The support that the agencies provided may have
served to mitigate the caregiving demands placed on the caregivers. To address this issue,
caregivers facing similar issues (e. g., a care recipient with a chronic illness or disability)
could be recruited who were not involved with one of these agencies and thus did not
receive this auxiliary support. Such recruiting could be conducted through doctors’
ofﬁces or hospitals. Some of the participants recruited through the university sample
provided care for loved ones with chronic illnesses or disabilities; however, it is not clear
whether or not they received support ﬁom caregiving support agencies. In order to have
been able to test the viability of this proposed explanation, we should have asked the
university participants if they received any such support.

The fact that the relationships between F IW and job satisfaction, life satisfaction,
and work withdrawal played out in the data even though the selection of participants was
problematic is reasonable. These main effect relationships have been established through
meta-analytical methods (Kossek & Ozeki, 1998), which combine data ﬁ'om a number of
different studies to gain a more accurate estimation of the true relationships that occur in
the population (Hunter & Hirsch, 1987). That is, these relationships have been
established to be consistent and strong in the population at large, and restriction of range
caused by a small amount of variance on the caregiving variable would not be enough to

prevent the detection of those relationships. However, an effective test of the mediation

76

hypothesis requires that there be an adequate amount of variance on the caregiving
variable; one cannot test the difference between those with low and high caregiving
demands if enough of those with high caregiving demands are not included in the sample.

A second plausible reason why the mediation did not play out in the data is that of
the unclear nature of the construct of special needs caregiving. In this study, special needs
caregiving was operationalized by factor analyzing different caregiving items to
determine which ones hung together, and thus comprised the essence of special needs
caregiving. However, it seems that Special needs caregiving might be much more
complex than that. In this study, special needs caregiving was essentially just large
caregiving demands. A single mom of three children under age ﬁve would have large
caregiving demands, as would a parent of two teenagers who was caring for her elderly
parents, a parent of an only child who had Down’s syndrome, a parent of a teenager who
was quadriplegic, or the spouse of a person in the end stages of cancer. It is likely that all
of these people would have scored highly on the extreme behaviors and overall
caregiving measures used in this study, yet the experiences of all of these people are
qualitatively different.

The relationship between the caregiver and the care recipient, whether the care
recipient’s needs are permanent or temporary, whether the care recipient’s needs are age-
appropriate expected (e. g., a two-year-old needing to be fed) or not (e.g., a sixty-year-old
needing to be fed), and whether the care recipient has an illness, physical disability, or
mental disability are all dimensions upon which the nature of the caregiving can differ. In
order to study the realm of special needs caregiving precisely, these dimensions Should

all be studied individually.

77

Previous studies in this area have focused more narrowly on caring for loved ones
with Speciﬁc chronic illnesses or disabilities. Marsh and Johnson (1997) studied the
effects of caring for a loved one with a chronic illness. Perlesz, Kinsella, and Crowe
(1999) studied families who cared for family members with traumatic brain injuries.
Marks (1998) was even more speciﬁc; She studied those who provided care for a disabled
family member, and she compared caregivers of disabled spouses, children, parents, and
other family members to see how the effects of caregiving may vary by the relationship
between the caregiver and care recipient. The study reported here took a broader
approach, deﬁning a Special needs caregiver as anyone with a caregiving demand that
required lots of time and unique caregiving behaviors. However, in the course of
attempting to generalize across relationships and special needs, the speciﬁcity of the
construct was lost, reducing our ability to describe its effects. Combining data across
types of special needs and relationships between the caregiver and care recipient may
have been too diffuse to have lead to a consistent set of responses.

A third possible explanation regarding why the hypothesized relationships in the
model were not supported regards the level of analytical power in this study. It has
previously been mentioned that the power in this study is low. Thus, one might expect
that the lack of results in this study may have been due to low power.

A fourth plausible explanation for the lack of support for this model is the poor
reliability of some of the measures used. The coefﬁcient alphas for FIW, standards of
acceptability, and work withdrawal were .71, .63 and .77, respectively. While these

reliabililities are not high and may have contributed to the lack of ﬁndings, they are

78

unlikely to have been the primary cause of so few signiﬁcant effects. Nevertheless, when
combined with low power, low reliabilities may have been a factor.
Future Research

Future research is needed to address some of the limitations of this study. First
and foremost is the need to more clearly deﬁne the construct of special needs caregiving.
The construct needs to be narrowed to include only Speciﬁc types of care recipient needs
and speciﬁc relationships between the caregiver and care recipient; it is not meaningﬁll to
analyze all care recipient needs and relationships between the caregiver and care recipient
together because as discussed earlier, they are too different qualitatively. After the
construct is narrowed, this study could be replicated using a larger number of participants
to increase the precision and power of this study. Perhaps with more participants and a
clearer idea of what special needs caregiving is, support could be found for the
moderation hypotheses.

Another promising area for future research surrounds the construct of work as an
escape. As previously mentioned, no literature to date has explored the issue of using
work as an escape ﬁom the pressures and stresses of home life. This study employed an
original measure that had good reliability. Additional research could further explore this

construct as well as the psychometric qualities of the scale.

79

APPENDIX A

Power Analyses

80

A Priori Power Analysis

Conducting a power analysis is an important activity prior to conducting research,
because you can reduce the probability of making a Type 11 error. The statistical power of
a test refers to the probability that the test will make the correct decision of rejecting the
null hypothesis when it is in fact wrong (Murphy & Myors, 1998). It is computed using
the sample Size, the effect size, alpha, and the desired power of a statistical test. For this
study, power was set at the conventional level of .80 for each of the hypotheses.

Because none of my hypotheses have been tested in the work-family literature so
far, I will infer correlation values based on a small-to-moderate effect size. For a small-
to—moderate effect size (R2? = .08, RZR = .04; Murphy & Myors, 1998) testing one
interaction, the F-value would be: '

((R’p - RZR)v2)/((1 - Rzp)vl) = 2.69

where: v1 = 3 because there are three predictors in the hypotheses (e.g., in

Hypothesis 2: FIW, job satisfaction, and the interaction between the two)
v2 = 186 because V2 = n — # of predictors — 1

This value exceeds the critical F -value of 2.662 for 190 participants (Murphy &
Myors, 1998). Therefore, this power analysis demonstrates that a sample size of at least
190 participants should be adequate to test the all of the hypotheses that use only one
one-way interaction. More participants will be necessary to obtain precise tests of the

more subtle mediating hypotheses.

81

Post Hoc Power Analysis
Because this study did not have enough participants to detect a small to medium
effect based on the a priori power analysis, it was deemed necessary to conduct a post
hoc power analysis to determine what the effect size would have to be in order to detect
an effect with 119 participants. Based on the following equation, the effect size would

have to be about .14 for a Signiﬁcant relationship to be detected.
r2 = (vl . F)/vz= .1364

where v; = 3 because there are three predictors in the hypotheses (i.e., in
Hypothesis 7: FIW, type of caregiving, and the interaction between the
two)
v2 = 115 because v2 = n — # ofpredictors —1

F = (R2 * v2)/((1- R2) v1) = 5.23 because Adj. R2 for Hypothesis 7 = .12

82

 

APPENDIX B

Recruitment Letter for Caregivers of the Chronically Ill

 

83

Dear Caregiver:

My name is Kerrie Vanden Bosch, and I am a graduate student in psychology at
Michigan State University. As my Masters’ Thesis, I am investigating how providing
care for a chronically ill family member affects a person’s experience of the relationship
between work and farme life.

I would appreciate it if you would consider participating in this study. If you agree to
participate, I will send you a packet of questionnaires in the mail, and you will be asked
to ﬁll them out and send them back in a postage-paid envelope. The questionnaires
should take approximately 30 minutes to complete and will ask you questions about your
caregiving role, your work life, and your family life. All of your answers will be
conﬁdential. The ﬁrst 100 people to return their surveys will receive $10 for their
participation. If you are interested in participating, please ﬁll out the postcard provided
and return it to me in the mail.

Please note that in order to qualify for this study, you must work at least 20 hours per
week and be the primary or co-caregiver for at least one chronically ill family member. In
addition, please note that your decision to participate or not to participate in this study
will not affect the quality of care that your child receives.

Thanks for your consideration!

Kerrie L. Vanden Bosch
Daniel R. Ilgen, Ph.D.

84

APPENDIX C

University Employee Recruitment Letter

85

July 19, 2002
Dear (insert ernployee’s name),

You have been selected to participate in an important research study. Michigan State
University researchers are investigating how providing care for a chronically ill family
member affects a person’s experience of the relationship between work and family life.
You will serve as the comparison group; you need not be a caregiver for a chronically
ill family member to participate. In order to qualify for this study, you must work at least
twenty hours per week and be the primary or co-caregiver for some dependent family
member. However, if you do provide care for a chronically ill family member, please
check that line on the postcard; we would greatly appreciate your participation as well.

Your help in this study would be greatly appreciated. We are asking you to ﬁll out and
return the enclosed postcard via campus mail if you qualify and are interested in
participating. Alternatively, you can indicate your willingness to participate by emailing
dikkerri@msu.edu with the information requested on the postcard. If you agree to
participate, we will send you a survey through your choice of either campus mail to your
work address or US Postal Service to your home address. The survey will take
approximately 30 minutes to complete, and you will be sent a check for $10 for your
participation.

Please note that in order to qualify for this study, you must work at least 20 hours per
week and either be the primary caregiver or equally share caregiving responsibilities with
someone else for at least one dependent family member. If you do not qualify for this
study, please check that line on the postcard and return it so that we do not contact
you again.

We hope that you choose to participate in this study!

Thanks,

Kerrie L. Vanden Bosch

Daniel R. Ilgen, Ph.D.

Department of Psychology

129 Psychology Research Building
Michigan State University

East Lansing, MI 48824

Email: dikkerri@msu.edu

Phone: 517-432-7069

Fax: 517-353-4873

86

APPENDIX D

Caregiving Assessment

 

 

87

 

Part]: Type of Caregiving.

Do you provide care for someone with a chronic illness (circle one)? Yes No

Part 11: Unique Caregiving Behaviors.

Think about the people for whom you provide care. Think about the needs that most
individuals their ages have. Now think about the needs that the people you provide care
for have. What kinds of things (if any) do you do to help the people that you provide care
for that most people their ages wouldn’t necessarily need? List one item next to each
number in the space provided below. List as behaviors as you engage in, up to a total of

10.

For example, most 60-year-old people can make their own meals. Thus, if you have to
prepare meals for a 60-year-old that you provide care for, you would want to list it here.
Also, most l3-year-olds can dress themselves. Thus, if you have to help a l3-year-old
dress him/herself, you would want to indicate that here as well.

“OWSQLLLLL

.0 0

Now, use the following scale to indicate how often you do the things that you listed
above. Write the number corresponding to your answer in the line next to each numbered

item.

1 = Daily

2 = Several times per week
3 = Once per week

4 = Several times per month
5 = Once per month

6 = Several times per year

88

 

 

 

Part III: Level of Caregiving.

Please focus on all of your non-paid caregiving activities, including caring for your
children, your grandchildren, your chronically ill loved ones, etc. Do not include any
caregiving activities that are related to your employment. Please answer the questions

below using the following scale.
1 = Daily
2 = Several times per week
3 = Once per week
4 = Several times per month
5 = Once per month
6 = Several times per year
7 = Never

 

About hOW often do you:
Assist care-recipient with walking

Experience extra expenses due to caregiving
Assist with care-recipient’s toileting

Transport care-recipient to doctor/other places
Bathe care-recipient

Assist care-recipient with exercises/therapy
Do care-recipient’s laundry

Lift or transfer care-recipient

Help care-recipient eat

10. Pick up after care-recipient

11. Dress care-recipient

12. Give medications to care-recipient
13. Preparing meals for care-recipient
14. Assisting care-recipient with health aids (e.g., braces, dentures, etc.)

15. Provide daytime supervision to care-recipient

16. Provide care-recipient assistance during the night

17. Read to care-recipient or help care-recipient with paperwork or

homework

PWNQMPwN_

 

89

 

APPENDIX E

Complete Caregiving Measure

90

 

91

 

How many hours per week do you spend caring for your loved ones?

 

Do you provide care for someone with a chronic illness (circle one)? Yes No

If yes, how many hours per week do you spend providing care for this
pcrson?__"

If yes, how often does caring for this person interrupt your daily routine? °

1 = Always

2 = Often

3 = Sometimes
4 = Seldom

5 = Never

 

92

 

 

This measure seeks to ﬁnd out about the kind of caregiving behaviors that you provide
for that are not normally needed by people the age of your care recipient.

Think about the people for whom you provide care. Think about the needs that most
individuals their ages have. Now think about the needs that the people you provide care
for have. What kinds of things (if any) do you do to help the people that you provide care
for that most people their ages wouldn’t necessarily need? List one item next to each
number in the space provided below. List as behaviors as you engage in, up to a total of

10.

For example, most 60-year-old people can make their own meals. Thus, if you have to
prepare meals for a 60-year-old that you provide care for, you would want to list it here.
Also, most l3-year-olds can dress themselves. Thus, if you have to help a l3-year-old

dress him/herself, you would want to indicate that here as well.

10.

Now, use the following scale to indicate how often you do the things that you listed
above. Write the number corresponding to your answer in the line next to each numbered

item.
1 = Daily
2 = Several times per week
3 = Once per week
4 = Several times per month
5 = Once per month
6 = Several times per year

93

 

 

 

Please answer the following questions about the type of caregiving activities in which
you engage.
Are you the caregiver responsible for a person with chronic illness? Yes / No
If yes, How is this person related to you?cl
l = He/she is my parent
2 = He/she is my child
3 = He/she is my spouse
4 = Other (please describe)

 

What is the illness?
Are you the caregiver responsible for any other dependents? Yes/No

 

If yes, How many people total do you provide care for?

Please list their ages and how they are related to you.
Age Relationship

 

3"!"

 

“2“

Please focus on all of your non-paid caregiving activities, including caring for your
children, your grandchildren, your chronically ill loved ones, etc. Do not include any
caregiving activities that are related to your employment. Please answer the questions
below using the following scale.
1 = Daily
2 = Several times per week
3 = Once per week
4 = Several times per month
5 = Once per month
6 = Several times per year
7 = Never
About how. often do you:
1 Assist care-recipient with walking
2 Experience extra expenses due to caregiving
3 Assist with care-recipient’s toileting
4. Transport care-recipient to doctor/other places
5. Bathe care-recipient
6
7
8
9

 

Assist care-recipient with exercises/therapy
Do care-recipient’s laundry
Lift or transfer care-recipient
. Help care-recipient eat
10. Pick up after care-recipient
1 1. Dress care-recipient
12. Give medications to care-recipient
13. Preparing meals for care-recipient
14. Assisting care-recipient with health aids (e.g., braces, dentures, etc.)
15. Provide daytime supervision to care-recipient
16. Provide care-recipient assistance during the night

 

94

 

17. Read to care-recipient or help careorecipient with paperwork or
homework
If you are the caregiver of a chronically 111 family member, please answer the
following 17 questions. If not, please skip to the next section. For these questions,
please tell us about the caregiving activities that you provided for your loved onc(s)
before the diagnosis of the chronic illness. Please focus on all of your non-paid
caregiving activities, including caring for your children, your grandchildren, your
chronically ill loved ones, etc. Do not include any caregiving activities that are
related to your employment. Please answer the questions below using the following
scale.“
1 = Daily
2 = Several times per week
3 = Once per week
4 = Several times per month
5 = Once per month
6 = Several times per year
7 = Never
About how often do you:
18. Assist care-recipient with walking
19. Experience extra expenses due to caregiving
20. Assist with care-recipient’s toileting
21. Transport care-recipient to doctor/other places
22. Bathe care-recipient
23. Assist care-recipieht with exercises/therapy
24. Do care-recipient’s laundry
25. Lift or transfer care-recipient
26. Help care-recipient eat
27. Pick up after care-recipient
28. Dress care-recipient
29. Give medications to care-recipient
30. Preparing meals for care-recipient
31. Assisting care-recipient with health aids (e.g., braces, dentures, etc.) _

32. Provide daytime supervision to care-recipient

 

 

'Some of the answers given to this question were exaggerated (e. g. some people indicated that they were
involved in caregiving behaviors 168 hours per week). Thus, it was determined that Unique Caregiving
Behaviors and Level of Caregiving may be a better proxy for volume of caregiving.

l’It was determined that the more important contributor to job- and life-related outcomes was total hours
spent caregiving, rather than the hours spent caring for one particular person, so the data from this question
were not analyzed.

”The fact that caregiving may interrupt someone’s daily routine depends not only on the caregiving
demands but also on the nature of the routine. Thus, the data from this question were not analyzed because
the variable was not as straightforward as initially thought.

dIt was recommended to me by a caregiving researcher to collect data on the individuals for whom the
caregiver provides care. However, it did not seem necessary to use this data to test the hypotheses in this
study.

°Special needs caregivers were also asked to respond to level of caregiving items again, keeping in mind
the caregiving activities they participated in before their loved ones were diagnosed with the illness. The
purpose of this was to assess the change in caregiving behaviors that occurred after the chronic illness was
diagnosed. However, these data were not used because it was not deemed necessary to use the change in
caregiving behaviors in order to address the research questions hypothesized.

95

 

 

 

33. Provide care-recipient assistance during the night
34. Read to care-recipient or help care-recipient with paperwork or homework

 

96

APPENDIX F

Work-Family Conﬂict (WIF and F IW) Items

97

Please answer the following questions about the extent to which your work life and your
family life interfere with each other using the scale below.

1 = Strongly Agree

2 = Agree
3 = Neither Agree nor Disagree
4 = Disagree

5 = Strongly Disagree

Family Interference With Work:

1. I’m often too tired at work because of the things I have to do at home.

2. My personal demands are so great that it takes away from my work.
3. My superiors and peers dislike how often I am preoccupied with my personal life

while at work.
4. My personal life takes up time that I’d like to spend at work.

 

Work Interference With Familyﬂ
5. After work, I come home too tired to do some of the things I’d like to do.

6. On the job, I have so much work to do that it takes away from my personal

interests.
7. My family/friends dislike how oﬁen I am preoccupied with my work when I am at

home.
8. My work takes up time that I’d like to spend with family/friends.

 

a'FIW is more relevant to situations involving special needs caregivers because they have large family

demands, and all of the hypotheses are relevant to FIW; therefore, the data collected for WIF was not used.

98

APPENDIX G

Job Satisfaction Items

 

 

99

 

 

Please answer the following questions about your satisfaction with your work.

Think of the work you do at the present. How well does each of the following words or
phrases describe your work? Please place an “X” in the appropriate box next to word

or phrase.

 

 

 

 

 

 

 

 

 

 

Yes No Can’t Yes No Can’t
decide decide
Fascinating Pleasant
Routine Useful
Satisfying Challenging
Boring Simple
Good Repetitive
Gives sense of Uses my abilities
accomplishment
Respected Dull
Uncomfortable Uninteresting
Can see results Creative

 

 

 

 

 

 

 

 

 

 

Think of the pay you get now. How well does each of the following words or phrases
describe your present pay? Please place an “X” in the appropriate box next to each

word or phrase.

 

 

 

 

 

Yes No Can’t Yes No Can’t
decide Decide
Income adequate Income provides
for normal luxuries
exﬂises
Fair Insecure
Barely live on Less than I
income deserve
Bad Well paid

 

 

 

 

 

Underpaid

 

 

 

 

 

 

Think of the opportunities for promotion that you have now. How well does each of
the following words or phrases describe these? Please place an “X” in the appropriate

box next to each word or phrase.

 

 

 

 

 

Yes No Can’t Yes No Can’t
decide decide
Good opportunities Unfair promotion
for promotion policy
Opportunities Inﬁ'equent
somewhat limited promotions
Promotion on Regular
ability promotions
Dead-end job Fairly good
chance for
promotion

 

 

 

 

 

Good chance for
romotion

 

 

 

 

 

100

 

 

Think of the kind of supervision that you get on your job. How well does each of the
following words or phrases describe this? Please place an “X” in the appropriate box
next to each word or phrase.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yes No Can’t Yes No Can’t
decide Decide
Asks my advice Tells me where I
stand
Hard to please Annoying
Irnpolite Stubborn
Praises good work Knows job well
Tactful Bad
Inﬂuential Intelligent
Up-to-date Poor manner
Doesn’t supervise Around when
enough needed
Has favorites Lazy

 

 

 

Think of the majority of the people that you work with now or meet in connection
with your job. How well does each of the following words or phrases describe them?
Please place an “X” in the appropriate box next to each word or phrase.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yes No Can’t Yes No Can’t
decide Decide

Stimulating Talk too

much
Boring Smart
Slow Lazy
Helpful Unpleasant
Stupid Gossipy
Responsible Active
Fast Narrow

interests
Intelligent Loyal
Eapy to make enemies Stubborn

 

Think of the your job in general. All in all, what is it like most of the time? Please

place an “X” in the appropriate box next to each word or phrase.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yes No Can’t Yes No Can’t
decide decide
Pleasant Superior
Bad Better than most
Ideal Disagreeable
Waste of time Makes me content
Good Inadequate
Undesirable Excellent
Worthwhile Rotten
Worse than most Enjoyable
Acceptable Poor

 

101

 

 

 

APPENDIX H

Life Satisfaction Items

102

Please answer the following questions about your satisfaction with your life, using the

scale below.

1 = Strongly Disagree
2 = Disagree
3 = Slightly Disagree
4 = Neither Disagree Nor Agree
5 = Slightly Agree
6 = Agree
7 = Strongly Agree

1. In most ways, my life is close to my idea].

2. The conditions of my life are excellent.

3. I am satisﬁed with my life.

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

5. If I lived my life over, I would change just about nothing.

103

APPENDIX I

Work Withdrawal Items

104

 

Please indicate how often the following statements are true for you using the scale below:

1 = Daily

2 = Several times per week
3 = Once per week

4 = Several times per month
5 = Once per month

6 = Several times per year

7 = Never

I use the work phone for personal phone calls.

I am unconcerned about personal appearance or manners at work.
I make excuses to get out of work.

I do work that is less than my best.

I use work equipment for personal purposes without permission.
I drink or use illicit drugs after work because of things that occur at work.

I am often late for work.

It is easy for me to be on time for work.

I often think about being absent for work.
10. I am often absent from work.

11. It is easy for me to attend work.

12. I often think about quitting my job.

13. I would like to quit my job.

14. I would retire if I had no ﬁnancial worries.
15. I often think about retiring.

16. I wish I could retire from my current job.

17. I wish I could retire from work in general.
18. If I won the lottery or received a large inheritance, I would retire.

105

 

APPENDIX J

Standards of Acceptability Scale

106

O = Did not occur
1 = Occurred, not severed
2 = Occurred, somewhat severe

3 = Occurred, moderately severe
4 = Occurred, very severe

5 = Occurred, extremely severe

Hassles are irritants that can range from minor annoyances to fairly major pressures,
problems, or difﬁculties. They can occur few or many times. Listed below are a number

of ways in which a person can feel hassled by his/her job. Please use the scale above to
indicate how hassled you have felt by these job-related concerns:

Job dissatisfaction
Hassles from boss or supervisor
Don’t like current work duties
Don’t like fellow workers

Worries about decisions to change jobs

Problems with employees

107

Customers or clients giving you a hard time
Problems getting along with fellow workers
Problems on job due to being a man or a woman

APPENDIX K

Coworker Social Support Items

108

5 = Strongly Agree

4 = Agree
3 = Neither agree nor disagree
2 = Disagree

1 = Strongly Disagree

Please use the scale above to indicate the extent to which you agree with the following
statements.

My coworkers really care about me.

I feel close to my coworkers.

My coworkers take a personal interest in me.

I feel appreciated by my coworkers.

My coworkers are ﬁiendly to me.

My coworkers would ﬁll in while I’m absent.

My coworkers are helpful in getting the job done.
My coworkers give useful advice on job problems.
My coworkers assist with unusual work problems.
My coworkers will pitch in and help.

109

APPENDIX L

Work as an Escape Items

110

Please indicate how often the following statements are true for you, using the scale
described below:

5‘3”!"

1 = Strongly Disagree

2 = Disagree

3 = Slightly Disagree

4 = Neither Disagree Nor Agree
5 = Slightly Agree

6 = Agree

7 = Strongly Agree

I enjoy going to work because it allows me to spend some time away from the
stresses of home.

My time at work provides an escape from the stresses of home.
Going to work gives me psychological and emotional beneﬁts.
The break ﬁ'om my caregiving duties that work provides is important to me.

 

111

APPENDIX M

Resilience Items

112

Please indicate how often the following statements are true for you, using the scale
described below.

1 = Strongly Disagree

2 = Disagree

3 = Slightly Disagree

4 = Neither Disagree Nor Agree
= Slightly Agree

6 = Agree

7 = Strongly Agree

When I make plans, I follow through with them.

I usually manage one way or another.

I am able to depend on myself more than anyone else.
Keeping interested in things is important to me.

I can be on my own if I have to.

I feel proud that I have accomplished things in my life.
I usually take things in stride.

I am friends with myself.______

I feel that I can handle many things at a time.

10.1 am determined.

11. I seldom wonder what the point of it all is.

12. I take things one day at a time.

13. I can get through difﬁcult times because I’ve experienced difﬁculty before.

99°99‘P'PP’PZ‘

 

14. I have self-discipline.

15. I keep interested in things.

16. I can usually ﬁnd something to laugh about.

17. My belief in myself gets me through hard times.

18. In an emergency, I’m someone people generally rely on.
19. I can usually look at a situation in a number of ways.
20. Sometimes I make myself do things whether I want to or not.
21. My life has meaning.—

22.1 do not dwell on things that I can t do anything about.
23. When I’m in a difﬁcult situation, I can usually ﬁnd my way out of it.
24. I have enough energy to do what I have to do.

25. It’s okay if there are people who don’t like me.

113

APPENDIX N

Positive and Negative Affect Items

114

 

This scale consists of a number of words that describe different feelings and emotions.
Read each item and then mark the appropriate answer in the space next to that word.
Indicate to what extent you generally feel this way, that is, how you feel on the average.
Use the following scale to record your answers.

1 = Very slightly or not at all

2 = A little
3 = Moderately
4 = Quite a bit
5 = Extremely
__ Interested __ Irritable
__ Distressed __ Alert
__ Excited __ Ashamed
__ Upset __ Inspired
Strong __ Nervous
__ Guilty __ Determined
__ Scared __ Attentive
__ Hostile __ Jittery
Enthusiastic _ Active

Proud Afraid

115

 

APPENDIX 0

Demographic Items

116

Please answer the following questions about your personal characteristics.

1. How many children live in your home?

 

2. What is your gender? Male/Female

3. How would you describe your race?
White/Caucasian
African-American

Hispanic

Asian/Paciﬁc Islander

Native American/Alaskan Native
Other

wpepge

 

4. What is your marital status?

Single

Single, but live with romantic partner
Married

Widowed

Separated

Divorced

Hop???

5. How would you describe your annual income?
a. Less than $25,000
b. $25,000-$50,000
c. $50,000-$75,000
d. $75,000- $100,000
e. $100,000-$150,000
f. Over $150,000

6. Are you employed outside the home? Yes / No

7. In the past ﬁve years, have you changed jobs? Yes / No

8. If yes, did the change in jobs entail a move to a new community?
Yes / No / Not Applicable, didn’t change jobs
9. If you were offered a job involving an attractive promotion to a new community,

how likely would you be to consider it?
Very Likely

Somewhat Likely

Neither Likely Nor Unlikely
Somewhat Unlikely

Very Unlikely

sup-99‘s»

117

References

Adams, G. A., & J ex, S. M. (1999). Relationships between time management,
control, work-family conﬂict, and strain. Journal of Occupational Health Psychology, 4,
72-77.

Adams, G. A., King, L. A., & King, D. W. (1996). Relationships of job and
family involvement, family social support, and work-family conﬂict with job and life
satisfaction. Journal of Applied Psychology, 81, 411-420.

Ahmad, A. (1996). Work-family conﬂict among married professional women in
Malaysia. Journal of Social Psychology, 136, 663-665.

Andrews, F ., & Withey, S. (1976). Social indicators of well-being. New York:
Plenum Press.

Bacharach, S. Bamberger, P., & Conley, S. (1991). Work-home conﬂict among
nurses and engineers: Mediating the impact of role stress on burnout and satisfaction at
work. Journal of Organizational Behaviour, 12, 39-53.

Barling, J. (1990). Employment, stress and family ﬁmctioning. Chichester,
England: Wiley.

Barling, J ., MacEwen, K. E., Kelloway, E. K., & Higginbottom, S. F. (1994).
Predictors and outcomes of eldercare-based interrole conﬂict. Psychology and Aging, 9,
391-397.

Barnett, R. C., & Marshall, N. L. (1992). Worker and mother roles, spillover
effects, and psychological distress. Women and Health, 18, 9-40.

Bartolome, F. (1972). Executives as human beings. Harvard Business Review, 50,

62-69.

118

Bartolome, F., & Evans, P. A. L. (1980). Professional lives versus private lives—
Shifting patterns of managerial commitment. Organizational Dynamics, 7, 3-29.

Bedeian, A. G., Burke, B. G., & Moffett, R. G. (1988). Outcomes of work-family
conﬂict among married male and female professionals. Journal of Management, 14, 475-
491.

Beehr, T. A., & McGrath, J. E. (1992). Social support, occupational stress and
anxiety. Anxiety, Stress & Coping: An International Journal, 5, 7-19.

Bowling Green State University. (1997 ). The Job Descriptive Index (2nd revision).
Bowling Green, OH: Author.

Brody, E. M., Kleban, M. H., Johnsen, P. T., Hofﬁnan, C., & Schoonver, C.
(1987). Work status and parent care: A comparison of four groups of women.
Gerontologist, 27, 201-208.

Buffardi, L. C., Smith, J. L., O’Brien, A. S., & Erdwins, C. J. (1999). The impact
of dependent-care responsibility and gender on work attitudes. Journal of Occupational
Health Psychology, 4, 356-367.

Burley, K. A. (1991). Family-work spillover in dual-career couples: A
comparison of two time perspectives. Psychological Reports, 68, 471-480.

Caplan, R. D. (1971). Organizational Stress and individual strain: A social
psychological study of risk factors in coronary heart disease among administrators,
engineers and scientists. Ann Arbor: Institute for Social Research.

Chambers, B. A, & Schmitt, N. (2002). Inequity in the performance evaluation
process: How you rate me affects how I rate you. Journal of Personnel Evaluation in

Education, I 6, 103-1 12.

119

 

Conrad, R. T.,'Horn, C., & Dwyer, J. W. (1992). Demographic perspectives on
gender and family caregiving. In J. W. Dwyer & R. T. Coward (Eds), Gender, families,
and elder care (pp. 18-33). Newbury Park, CA: Sage.

Coward, R. T., Home, C., & Dwyer, J. W. Demographic perspectives on gender
and family caregiving. In J. W. Dwyer & R. T. Coward (Eds), Gender, families and
elder care (pp. 18-33). Thousand Oaks, CA: Sage Publications, Inc.

Crosby, F. (1984). Job satisfaction and domestic life. In J. D. Lee & R. N.
Kanungo (Eds), Management of work and personal life (pp. 41-60). New York: Praeger.

Della-Rocco, P. S., Comperatore, C., Caldwell, L., & Cruz , C. (1999). The
effects of napping on night shift performance. FAA Office of Aviation Medicine Reports,
1-33.

Diener, E. (1994). Assessing subjective well-being: Progress and opportunities.
Social Indicators Research, 31, 103-157.

Diener, E. (1985). The Satisfaction With Life Scale. Journal of Personality
Assessment, 49, 71-75.

Ducharme, L. J ., & Martin, J. K. (2000). Unrewarding work, coworker support,
and job satisfaction. Work and Occupations, 27, 223-243.

Eagle, B. W., Icenogle, M. L., Maes, J. D., & Miles, E. W. (1998). The
importance of employee demographic proﬁles for understanding experiences of work-
family interrole conﬂicts. Journal of Social Psychology, 138, 690-709.

Evans, P., & Bartolome, F. (1980). The relationship between professional and
private life. In C. B. Derr (Ed.), Work, family and career (pp. 281-317). NewYork:

Praeger.

120

Evans, P., & Bartolome, F. (1984). The changing pictures of the relationship
between career and family. Journal of Occupational Behaviour, 5, 9-21.

Frone, M. R., Russell, M., & Cooper, M. L. ( 1992). Antecedents and outcomes of
work-family conﬂict: Testing a model of the work-family interface. Journal of Applied
Psychology, 77, 65-78.

Frone, M. R., Russell, M., & Cooper, M. L. (1995). Job stressors, job involvement
and employee health: A test of identity theory. Journal of Occupational and
Organizational Psychology, 68, 1-11.

Frone, M. R., Yardley, J. K, & Markel, K. S. (1997). Developing and testing an
integrative model of the work-family interface. Journal of Vocational Behavior, 50, 145-
167.

Gignac, M. A. M., Kelloway, E. K., & Gottlieb, B. H. (1996). The impact of
caregiving on employment: A mediational model of work-family conﬂict. Canadian
Journal on Aging, 15, 525-542.

Goff, S. J ., Mount, K., & J arnison, R. L. (1990). Employer supported child care,
work/family conﬂict, and absenteeism: A ﬁeld study. Personnel Psychology, 43, 793-
809.

Greenglass, E. R., & Burke, R. J. (1988). Work and family precursors of burnout
in teachers: Sex differences. Sex Roles, 18, 215-229.

Greenhaus, J. H., & Beutell, N. J. (1985). Sources and conﬂict between work and

family roles. Academy of Management Review, 10, 76-88.

121

Greenhaus, J. H., Parasuraman, S., & Collins, K. M. (2001). Career involvement
and family involvement as moderators of relationships between work-family conﬂict and
withdrawal ﬁ'om a profession. Journal of Occupational Health Psychology, 6, 91-100.

Gutek, B. A., Searle, S., & Klepa, L. (1991). Rational versus gender role
explanations for work-family conﬂict. Journal of Applied Psychology, 76, 560-568.

Hanisch, K. A. & Hulin, C. L. (1991). General attitudes and organizational
withdrawal: An evaluation of causal model. Journal of Vocational Behavior, 39, 110-
128.

Helson, H., & Rholes, F. H. (1959). A quantitative study of reversal of classical
lightness-contrast. American Journal of Psychology, 72, 530-538.

Higgins, C. A., & Duxbury, L. E. (1992). Work-family conﬂict: A comparison of
dual-career and traditional-career men. Journal of Organizational Behavior, 13, 389-411.

Holm, J. E., & Holroyd, K. A. ( 1992). The Daily Hassles Scale (Revised): Does it
measure stress or symptoms? Behavioral Assessment, 14, 465-482.

Hulin, C. L. (1969). Sources of variation in job and life satisfaction: The role of
community and job-related variables. Journal of Applied Psychology, 53, 279-291.

Hulin, C. L., Roznowski, M., & Hachiya, D. (1985). Alternative opportunities and
withdrawal decisions: Empirical and theoretical discrepancies and an integration.
Psychological Bulletin, 97, 233-250.

Hunter, J. E., & Hirsch, H. R. (1987). Applications of meta-analysis. In C. L.
Cooper & I. T. Robertson (Eds), Team performance, assessment and measurement:

Theory, research and applications. Hillsdale, NJ: Erlbaum.

122

 

Ilgen, D. R., & Hamstra, B. W. (1972). Performance satisfaction as a function of
the difference between expected and reported performance at ﬁve levels of reported
performance. Organizational Behavior and Human Decision Processes, 7, 359-3 70.

J arnes, L., & James, L. (1989). Integrating work environment perceptions:
Explorations into the measurement of meaning. Journal of Applied Psychology, 74, 739-
751.

Judge, T. A., Boudreau, J. W., & Bretz, R. D. (1994). Job and life attitudes of
male executives. Journal of Applied Psychology, 79, 767-782.

Judge, T. A., & Watanabe, S. (1993). Another look at the job satisfaction-life
satisfaction relationship. Journal of Applied Psychology, 78, 939-948.

Kinney, J. M., & Stephens, M. P. (1989). Caregiving Hassles Scale: Assessing the
daily hassles of caring for a family member with dementia. Gerontologist, 29, 328-332.

Kossek, E. E., & Ozeki, C. (1998). Work-family conﬂict, policies, and the job-life
satisfaction relationship: A review and directions for organizational behavior-human
resources research. Journal of Applied Psychology, 83, 139-149.

Lambert, S. J. (1990). Processes linking work and family: A critical review and
research agenda. Human Relations, 43, 239-257.

Lee, J. A. (1997). Balancing elder care responsibilities and work: Two empirical
studies. Journal of Occupational Health Psychology, 2, 220-228.

Locke, E. (1976). The nature and causes of job satisfaction. In M. D. Dunnette,
(Ed.), Handbook of industrial and organizational psychology (pp.1297-1349). Chicago:

Rand McNally.

123

London, M., Crandall, R., & Seals, G. W. (1977). The contribution of job and
leisure satisfaction to quality of life. Journal of Applied Psychology, 62, 328-334.

Marks, N. F. (1998). Does it hurt to care? Caregiving, work-family conﬂict, and
midlife well-being. Journal of Marriage and the Family, 60, 951-966.

Marsh, D. T., & Johnson, D. L. (1997). The family experience of mental illness:
Implications for intervention. Professional Psychology: Research & Practice, 28, 229-
237.

Marsh, D. T., Leﬂey, H. P., Evans-Rhodes, D., Ansell, V. 1., & Doerzbacher, B.
M. (1996). The family experience of mental illness: Evidence for resilience. Psychiatric
Rehabilitation Journal, 20, 3-12.

McCann, B. S., Russo, J., & Benjamin, G. A. H. (1997). Hostility, social support,
and perceptions of work. Journal of Occupational Health Psychology, 2, 175-185.

Mobley, W. (1977). Intermediate linkages in the relationship between job
satisfaction and employee turnover. Journal of Applied Psychology, 62, 23 7-240.

Murphy, K. R., & Myors, B. (1998). Statistical Power Analysis. Mahwah, NJ:
Lawrence Earlbaum Associates.

Netemeyer, R. G., Boles, J. S., & McMurrian, R. (1996). Development and
validation of work-family conﬂict and family-work conﬂict scales. Journal of Applied
Psychology, 81, 400-410.

O’Driscoll, M. P., Ilgen, D. R., & Hildreth, K. (1992). Time devoted to job and
off-job activities, interrole conﬂict, and affective experiences. Journal of Applied

Psychology, 77, 272-279.

124

Parasuraman, S. Purohit, Y. S., Godshalk, V. M.,& Beutell, N. J. (1996). Work
and family variables, entrepreneurial career success, and psychological well-being.
Journal of Vocational Behaviour, 48, 275-300.

Perlesz, A., Kinsella, G., & Crowe, S. (1999). Impact of traumatic brain injury of
the family: A critical review. Rehabilitation Psychology, 44, 6-35.

Perrewe, P. L., Hochwarter, W. A., & Kiewitz, C. (1999). Value attainment: An
explanation for the negative effects of work-family conﬂict on job and life satisfaction.
Journal of Occupational Health Psychology, 4, 318-326.

Pond, S. B., & Green, S. B. (1983). The relationship between job and marriage
satisfaction within and between spouses. Journal of Occupational Behaviour, 4, 145-155.

Quittner, A. L., Espelage, D. L., Opipari, L. C., Carter, B., Eid, N., & Eigen, H.
(1998). Role strain in couples with and without a child with a chronic illness:
Associations with marital satisfaction, intimacy, and daily mood. Heath Psychology, 1 7,
112-124.

Richardsen, A. M., Burke, R. J ., & Mikkelsen, A. (1999). Job pressures,
organizational support, and health among Norwegian women managers. International
Journal of Stress Management, 6, 167-177.

Roznowski, M., & Hanisch, K. A. (1990). Building systematic heterogeneity into
work attitudes and behavior measures. J Wf Vocatioﬁrml Behavior. 36 361-3 75.

Scharlach, A. E., Sobel, E., L. & Roberts, R. E. (1993). Employment and
caregiver strain: An integrative model. Gerontologist, 31, 778-787.

Staines, G. L., & Pleck, J. H. (1984). Nonstandard work schedules and family life.

Journal of Applied Psychology, 69, 515-523.

125

Stephens, M. A. P., Franks, M. M., & Atienza, A., A. (1997). Where two roles
intersect: Spillover between parent care and employment. Psychology and Aging, 12, 30-
37.

Stephens, M. A. P., Franks, M. M., & Townsend, A. L. (1994). Stress and
rewards in women’s multiple roles: The case of women in the middle. Psychology and
Aging, 9, 45-52.

Stephens, M. A. P., & Townsend, A. L. (1997). Stress of parent care: Positive
and negative effects of women's other roles. Psychology and Aging, 12, 376-386.

Stewart, W., & Barling, J. (1996). Fathers’ work experiences affect children’s
behaviors via job-related affect and parenting behaviors. Journal of Organizational
Behavior, 17, 221-232.

Tartarkiewicz, W. (1976). Analysis of Happiness. The Hague, The Netherlands:
Matinus Nihjoff.

Thomas, L. T., & Ganster, D. C. (1995). Impact of family-supportive work
variables on work-family conﬂict and strain: A control perspective. Journal of Applied
Psychology, 80, 6-15.

Thompson, C. A., & Blau, G. (1993). Moving beyond traditional predictors of
job involvement: Exploring the impact of work-family conﬂict and overload. Journal of
Social Behavior and Personality, 8, 635-646.

Wagrrild, G. M., & Young, H. M. (1993). Development and psychometric
evaluation of the resilience scale. Journal of Nursing Measurement, 1, 165-178.

Wallace, J. E. (1997). It's about time: A study of hours worked and work spillover

among law ﬁrm lawyers. Journal of Vocatinal Behavior, 50, 227-248.

126

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation or
brief measures of positive and negative affect: The PANAS scales. Journal of Personality
and Social Psychology, 54, 1063-1070.

Zedeck, 8. (Ed.). (1992). Work. families, and organizations. San Francisco:
Jossey-Bass.

Zedeck, S., & Mosier, K. L. (1990). Work in the family and employing

organization. American Psychologist, 45, 240-251.

127