Content uploaded by Kristen Shockley
Author content
All content in this area was uploaded by Kristen Shockley on Nov 17, 2017
Content may be subject to copyright.
When flexibility helps: Another look
at the availability of flexible work
arrangements and work–family conflict
Kristen M. Shockley
*
, Tammy D. Allen
The University of South Florida, 4202 E. Fowler Avenue, PCD4118G, Tampa, FL 33620-7200, USA
Received 20 July 2007
Available online 24 August 2007
Abstract
Despite the positive press given to flexible work arrangements (FWA), empirical research inves-
tigating the link between the availability of these policies and work–family conflict is largely equiv-
ocal. The purpose of the present study was to begin to reconcile these mixed results through more
precise measurement and the examination of moderators. Using a sample of employed women,
we found that FWA relate more highly to work interference with family (WIF) than to family inter-
ference with work (FIW) and that temporal flexibility (flextime) has a stronger relationship with
WIF than does spatial flexibility (flexplace). Additionally, we found family responsibility signifi-
cantly moderated these relationships, such that the relationship was stronger for those with greater
family responsibility. Practical and theoretical implications are discussed.
2007 Elsevier Inc. All rights reserved.
Keywords: Flexible work arrangements; Flextime; Flexplace; Telecommuting; Work–family conflict; Family
responsibility; Family-supportive organization perceptions
1. Another look at the relationship between flexible work arrangements availability and
work–family conflict
Due to the changing demographics of the workforce, work–family conflict (WFC) has
become a prevalent issue among employed individuals. In fact, the National Survey of the
0001-8791/$ - see front matter 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jvb.2007.08.006
*
Corresponding author. Fax: +1 813 974 4617.
E-mail address: kshockle@mail.usf.edu (K.M. Shockley).
Available online at www.sciencedirect.com
Journal of Vocational Behavior 71 (2007) 479–493
www.elsevier.com/locate/jvb
Changing Workforce found that 43% of employees with families report ‘‘some’’ or ‘‘a lot’’
of interference between their jobs and family lives (Bond, Thompson, Galinsky, & Prottas,
2002). In an attempt to alleviate such conflicts, organizations have adapted by offering
family–friendly benefits (e.g., Allen, 2001; Grover & Crooker, 1995). In particular, flexible
work arrangements (FWA) have been cited as key in the effort to help employees manage
competing work and family domains (Adams & Jex, 1999; Golden, 2001; Goldenhar,
2003).
Despite the positive press given to FWA, research demonstrating a link between FWA
and less WFC is inconsistent (Allen & Shockley, 2006). The heterogeneity associated with
the research findings regarding FWA is empirically illustrated by two recent meta-analytic
studies. Byron (2005) reported a meta-analytic effect size of .30 between flexibility and
work interference with family (WIF) and of .17 with family interference with work
(FIW). In contrast, Mesmer-Magnus and Viswesvaran (2006) reported nonsignificant
effect sizes of .01 with WIF and .04 with FIW. The inconsistencies found even within
meta-analytic research suggest that there is a great deal of variation associated with the
effectiveness of FWA.
The purpose of the present study is to identify several factors that may help explain the
inconsistent results across existing research. First, few studies have examined flexibility in
terms of both time and place and both directions of WFC within the same study. This is
important in that some forms of flexibility may be more beneficial in terms of reducing or
preventing WFC than others. In addition, few studies have been conducted that examine
the boundary conditions under which FWA are more or less beneficial. Given the disparity
in results regarding FWA and WFC, we investigated the presence of moderator variables.
We posit that some individuals stand to benefit more from FWA than others based on the
extent that they perceive their organization to be family-supportive and their degree of
family responsibility. In order to further target the boundary conditions under which
FWA relate to WFC, our study focuses specifically on working women because research
suggests that women continue to maintain greater responsibility for the majority of
domestic tasks (e.g., Bianchi, Milkie, Sayer, & Robinson, 2000; Bond et al., 2002; Cromp-
ton, Brockmann, & Lyonette, 2005) and are more likely to be part of a dual-earner couple.
Accordingly women may stand to benefit the most from FWA and therefore merit closer
attention within this literature (Byron, 2005).
1.1. Flexible work arrangements and work–family conflict
FWA are formally defined as ‘‘alternative work options that allow work to be accom-
plished outside of the traditional temporal and/or spatial boundaries of a standard work-
day’’ (Rau, 2003). The two most popular forms of FWA are flextime and flexplace
(SHRM Foundation, 2001). Flextime refers to flexibility in the timing of work, whereas
flexplace involves flexibility in the location where work is completed, often referring to
work conducted at home (also known as telework or telecommuting). Flexplace arrange-
ments may also allow flexibility in the timing of work, but this is not always the case.
Although there are other forms of FWA, we focused on flextime and flexplace due to their
widespread availability in organizations.
Many studies have investigated the relationship between FWA and WFC, with little
consensus regarding findings (Allen & Shockley, 2006). For example, some studies have
indicated that flextime and/or flexplace relates to less WFC (e.g., Hammer, Allen, & Gri-
480 K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493
gsby, 1997; Hill, Yang, Hawkins, & Ferris, 2004), whereas others have found no relation-
ship between the two (e.g., Anderson, Coffey, & Byerly, 2002; Galinsky, Bond, & Fried-
man, 1996; Greenhaus, Parasuraman, Granrose, Rabinowitz, & Beutell, 1989; Thomas
& Ganster, 1995). As previously mentioned, the discrepancy in findings is further exempli-
fied by two meta-analyses, in which one found a relationship between flexibility and WIF/
FIW (Byron, 2005), and the other did not (Mesmer-Magnus & Viswesvaran, 2006). The
discrepancy in findings may in part be explained by the different study’s inclusion criteria.
Byron (2005) looked at only schedule flexibility, and Mesmer-Magnus and Viswesvaran
(2006) included studies that looked flexibility in both the timing and location of work.
Byron’s (2005) larger effect sizes suggest that flextime may be more effective than flexplace
in preventing WFC.
Theoretically, there are several reasons why flextime may be more efficacious that flex-
place. Some telecommuting arrangements may be rigid in terms of temporal scheduling.
The telecommuter who must remain chained to his/her computer from 8 to 5 has little
more ‘‘flexibility’’ than someone sitting in a corporate office from 8 to 5. In addition, tele-
commuting may make the maintenance of work and family boundaries more difficult.
Boundary theory asserts that humans create boundaries in order to cope with and under-
stand their environments (Nippert-Eng, 1996). Naturally, individuals often create bound-
aries between their work and family lives; these abstract borders advise individuals when
to fulfill the family role vs. the worker role (Ashforth, Kreiner, & Fugate, 2000). Telework
removes the physical separation between work and family roles, thus making it potentially
more difficult to maintain a boundary between both roles (Eckenrode & Gore, 1990; Lap-
ierre & Allen, 2006).
Both Allen and Shockley’s (2006) qualitative review and Byron’s (2005) quantitative
meta-analyses also suggested that FWA may relate more highly to WIF than to FIW. Fur-
ther, this is consistent with the domain specificity hypothesis, which suggests that the pre-
dictors of WIF reside in the work domain while the predictors of the FIW reside in the
family domain (Frone, 2003). This leads to our first set of hypotheses:
Hypothesis 1a: Flextime is more highly related to WIF than to flexplace.
Hypothesis 1b: Both forms of FWA are more highly related to WIF than to FIW.
The two meta-analytic studies also underscore the possibility of moderators. The cred-
ibility intervals in both studies were large, indicating considerable heterogeneity. In the fol-
lowing sections we propose two potential moderators, one from the family domain and
one from the work domain.
1.2. Family responsibility
A primary function of flexible policies is to ease WFC by facilitating one’s ability to
attend to family matters without missing work obligations (Honeycutt & Rosen, 1997).
For this reason, those with a great deal of responsibility for family and household tasks
stand to gain from FWA. Daytime family-related activities (e.g., doctors appointments,
teacher conferences, extracurricular events) that are made difficult by standard work
arrangements are facilitated by FWA. Additionally, time for domestic tasks or childcare
may increase with flexplace, if employees are able to effectively intertwine them with
K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493 481
work-related duties (Osnowitz, 2005). Elimination/decrease in commute time may also
offer individuals more time to accomplish domestic tasks (Noonan, Estes, & Glass, 2007).
Prior research also provides some preliminary evidence for the importance of family
responsibility. In her meta-analysis, Byron (2005) found that having a higher percentage
of females in a sample was negatively related to the study effect size between schedule flex-
ibility and WIF and FIW. Thus, flexible schedules seem to provide more of a protective
benefit for women than for men, a finding that is likely attributable to the tendency for
women to be more responsible for family matters. In summation, because FWA involve
altering work schedules to meet family demands, women with greater family responsibil-
ities should benefit the most from flexible policies and report less WFC than women with
fewer family responsibilities.
Hypothesis 2: Family responsibility moderates the relationship between flextime access
and (a)WIF/(b)FIW. Specifically, there is a negative relationship between flextime
access and (a)WIF/(b)FIW when family responsibility is high and no relationship
between flextime access and (a)WIF/(b)FIW when family responsibility is low.
Hypothesis 3: Family responsibility moderates the relationship between flexplace access
and (a)WIF/(b)FIW. Specifically, there is a negative relationship between flexplace
access and (a)WIF/(b)FIW when family responsibility is high and no relationship
between flexplace access and (a) WIF/(b)FIW when family responsibility is low.
1.3. Family-supportive organization perceptions
Although family–friendly benefits can serve as valuable tools in helping employees man-
age multiple life roles, research shows that the mere offering of benefits is not always suffi-
cient. Organizations must also adapt their environment, norms, and values in order for
benefits to have the intended impact (Galinsky & Stein, 1990). In line with this notion, Allen
(2001) examined employee’s global perceptions about the extent that their organization is
supportive of family–friendly programs. These family-supportive organizational percep-
tions (FSOP) positively correlated with the number of flexible benefits offered and used
and accounted for unique variance associated with WFC over and above supervisor sup-
port and the availability of policies. This suggests that those with high FSOP may experi-
ence decreased WFC not only through increased use of available policies but also as a direct
result of the organization’s family-related support. On the other hand, low FSOP is asso-
ciated with a variety of factors that may undermine the helpfulness of flexible benefits. For
example, FWA users may face negative career outcomes, disapproving sentiments from
supervisors or coworkers, and a general feeling that use is unacceptable (Thompson,
Beauvais, & Lyness, 1999). Thus, organizations perceived to be family-supportive facilitate
FWA as a means of better balancing competing life demands, whereas those that are seen as
less family-supportive hinder policy effectiveness. Consistent with previous research, we
propose the following hypotheses regarding the moderating effects of FSOP:
Hypothesis 4: FSOP moderates the relationship between access to flextime and (a)WIF/
(b)FIW. Specifically, there is a negative relationship between flextime access and
(a)WIF/(b)FIW when FSOP are high and no relationship between flextime access
and (a)WIF/(b)FIW when FSOP are low.
482 K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493
Hypothesis 5: FSOP moderates the relationship between access to flexplace and
(a)WIF/(b)FIW. Specifically, there is a negative relationship between flexplace access
and (a)WIF/(b)FIW when FSOP are high and no relationship between flexplace access
and (a) WIF/(b)FIW when FSOP are low.
2. Methods
2.1. Participants
Participants in the study were 230 women who worked a minimum of 20 h per week and
who had at least one child living at home or an employed spouse. Their average age was
36.69 (SD = 8.65). Median level of education was a 4-year college degree. The majority of
the participants were Caucasian (91.3%). Participants were from a variety of different
industries and held diverse job titles such as Strategic HR Advisor, Senior Purchasing
Agent, Teacher, Charge RN, and Loan Specialist.
2.2. Procedure
Data were collected as part of a larger work-life study. Several methods were used to
solicit participants. One, personal and professional colleagues were contacted and pro-
vided with a link to an online survey used for data collection. Contacted colleagues were
asked to forward the survey request and link to other contacts and colleagues. Two, the
survey link and an invitation to participate in the study were posted at internet discussion
forums for parents. Finally, the survey link and invitation to participate was sent to an
email distribution list for human resource professionals.
2.3. Measures
2.3.1. Work–family conflict
Both directions (WIF and FIW) of WFC were assessed using the measure developed by
Carlson, Kacmar, and Williams (2000). Nine items targeted WIF (e.g., ‘‘I am so emotion-
ally drained when I get home from work that it prevents me from contributing to my fam-
ily.’’). Nine items measured FIW (e.g., ‘‘The time I spend with my family often causes me
to not spend time in activities at work that could be helpful to my career.’’). Response
options were based on a 5-point scale that ranged from strongly disagree to strongly agree,
with higher scores representing greater WFC. Coefficient alpha for WIF = .89 and for
FIW = .89.
2.3.2. Flexible work arrangement availability
Based on the measure developed by Hyland (2000), access to flextime was assessed
with 2 items representing flexibility in terms of time (e.g., ‘‘I have the freedom to vary
my work schedule’’), and access to flexplace was measured with two items representing
flexibility in terms of place (e.g., ‘‘I have the freedom to work wherever is best for
me—either at home or at work’’). Responses were set on a 5-point scale that ranged
from entirely not true to entirely true. Coefficient alpha for flextime = .84 and for
flexplace = .91.
K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493 483
2.3.3. Family responsibility
Participants were given a list of five family responsibilities: cleaning house, planning
family life, making children understand what is right, taking children to the doctor, and
putting children to bed. For each responsibility, participants were asked to choose one
of the following response options: performed by myself alone, performed by myself with
help from another person, primarily performed by another person. Individuals also had
the option to indicate if the task was not performed in his/her household. Responses were
scored so that higher scores indicated that the individual had greater family responsibility.
Because this is a formative measure internal consistency between items was not expected,
thus coefficient alpha was not computed.
2.3.4. Family-supportive organization perceptions
The 9-item version of Allen’s (2001) measure was used to determine FSOP. Example
items include ‘‘The way to advance is to keep nonwork matters out of the workplace’’
and ‘‘Employees are given ample opportunity to perform both their job and their personal
responsibilities well.’’ Responses were made on a five point scale that ranged from strongly
disagree to strongly agree. Internal consistency was .84.
2.3.5. Control variables
Several variables commonly controlled for in the WFC literature were included. We
assessed whether or not there was at least one child living at home (1, yes; 2, no), partic-
ipant age, average number of hours worked each week, and marital status (1, not married;
2, living with partner or married).
3. Results
Means, standard deviations, and correlations are presented in Table 1. Hypothesis 1a
stated that flextime would be more highly related to WIF than would flexplace. The
correlation between WIF and flextime was r=.23, p< .01 while the correlation
between WIF and flexplace was r=.08, ns.At-test for a significant difference
between dependent rs was computed. The result was significant (t=2.51, p< .05).
Thus, Hypothesis 1a was supported. WIF was more highly associated with flextime
than with flexplace.
Hypothesis 1b stated that both forms of FWA would more highly relate to WIF than
to FIW. The correlation between WIF and flextime was r=.23, p< .01 while the cor-
relation between FIW and flextime was r=.07, ns. The difference in the correlations
was significant (t=3.12, p< .05). The correlation between WIF and flexplace was
r=.08, ns, while the correlation between FIW and flexplace was r= .02, ns. The dif-
ference in the correlation was significant (t=1.97, p< .05). Thus, Hypothesis 1b was
supported in that both forms of flexibility were more highly associated with WIF than
with FIW.
The remaining hypotheses were tested with moderated regression. Hypothesis 2a stated
that family responsibility would moderate the relationship between access to flextime and
WIF while Hypothesis 2b proposed the same for FIW. As shown in Table 2, both inter-
actions were significant (Hypothesis 2a, b=1.33, DR
2
= .05, p< .01; Hypothesis 2b,
b=1.47, DR
2
= .06, p< .001). Fig. 1 illustrates the nature of the interaction between
flextime and family responsibility on WIF. As predicted when family responsibilities were
484 K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493
low, there was no relationship between flextime and WIF. On the other hand, when family
responsibilities were high the relationship between flextime and WIF was negative, indicat-
ing that greater flextime was associated with less WIF. Fig. 2 illustrates the pattern of rela-
tionships for FIW. Again, the results show that when family responsibilities were high,
there was a negative relationship between flextime and FIW. However, the results also
show that when family responsibilities were low, there was a positive relationship between
flextime and FIW.
Table 1
Means, standard deviations, and intercorrelations
Mean SD 12345678910
1. WIF 2.76 0.78 —
2. FIW 2.39 0.71 .71
**
—
3. Flextime 2.91 1.19 .23
**
.07 —
4. Flexplace 2.01 1.12 .08 .02 .60
**
—
5. Family
responsibility
2.33 0.40 .26
**
.15
*
.27
**
.16
*
—
6. FSOP 3.41 0.72 .49
**
.37
**
.20
**
.04 .13 —
7. Age 36.69 8.65 .03 .01 .06 .03 .07 .07 —
8. Marital status 1.88 0.33 .06 .00 .12 .10 .43
**
.01 .02 —
9. Work hours 40.51 9.24 .19
**
.01 .10 .01 .03 .10 .21
**
.08 —
10. Parental
status
1.24 0.43 .00 .10 .03 .02 .10 .02 .20
**
.12 .23
**
—
Marital status: 1, not married; 2, married or living with partner. Parental status: 1, no child living at home; 2,
child living at home.
*
p< .05.
**
p< .01.
Table 2
Moderated regression results for flextime and family responsibility
WIF FIW
Step 1
Age .03 .05
Marital status .10 .11
Work hours .19
**
.04
Parental status .03 .11
R
2
D.04 .01
Step 2
Flextime 1.17
**
1.44
**
Family responsibility .82
***
.81
***
R
2
D.09
***
.02
Step 3
Flextime ·family responsibility 1.33
**
1.47
***
R
2
D.05
**
.06
***
Final F6.30
***
3.12
**
Values represent standardized regression coefficients from the final equation.
**
p< .01.
***
p< .001.
K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493 485
Hypothesis 3a stated that family responsibility would moderate the relationship
between access to flexplace and WIF while Hypothesis 3b suggested the same for FIW.
Results are shown in Table 3. Both interactions were significant (Hypothesis 3a,
b=.95, DR
2
= .03, p< .01; Hypothesis 3b, b=1.07, DR
2
= .04, p< .01), providing
support for both hypotheses. Fig. 3 illustrates the nature of the interaction between flex-
place and family responsibility on WIF. When family responsibility was low, there was no
relationship between flexplace and WIF. On the other hand, when family responsibility
1
2
3
4
5
Low flextime Hi
g
h flextime
WIF
Low fam resp
High fam resp
Fig. 1. Interaction between flextime and family responsibility on WIF.
1
2
3
4
5
Low flextime High flextime
FIW
Low fam resp
High fam resp
Fig. 2. Interaction between flextime and family responsibility on FIW.
486 K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493
was high the relationship between flexplace and WIF was negative, indicating that greater
flextime was associated with less WIF. Fig. 4 illustrates the nature of the interaction for
FIW. When family responsibility was low, greater access to flexplace was associated with
greater FIW. When family responsibility was high, greater access to flexplace was associ-
ated with less FIW.
Table 3
Moderated regression results for flexplace and family responsibility
WIF FIW
Step 1
Age .04 .05
Marital status .10 .11
Work hours .19
**
.02
Parental status .06 .13
R
2
D.04 .01
Step 2
Flexplace .86
*
1.08
**
Family responsibility .60
***
.55
***
R
2
D.06
**
.02
Step 3
Flexplace ·family responsibility .95
**
1.07
**
R
2
D.03
**
.04
**
Final F4.61
***
2.47
*
Values represent standardized regression coefficients from the final equation.
*
p< .05.
**
p< .01.
***
p< .001.
1
2
3
4
5
Low flexplace High flexplace
WIF
Low fam resp
High fam resp
Fig. 3. Interaction between flexplace and family responsibility on WIF.
K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493 487
Hypothesis 4a stated that FSOP would moderate the relationship between access to
flextime and WIF while Hypothesis 4b suggested the same for FIW. As shown in Table
4, neither interaction was significant (Hypothesis 4a, b=.23, DR
2
= .00, ns; Hypothesis
4b, b= .06, DR
2
= .00, ns), providing no support for Hypotheses 4a or 4b. Hypothesis 5a
stated that FSOP would moderate the relationship between access to flexplace and WIF
while Hypothesis 5b suggested the same for FIW. As shown in Table 5 neither interaction
was significant (Hypothesis 5a, b=.02, DR
2
= .00, ns; Hypothesis 5b, b= .43,
1
2
3
4
5
Low flex
p
lace Hi
g
h flex
p
lace
FIW
Low fam resp
High fam resp
Fig. 4. Interaction between flexplace and family responsibility on FIW.
Table 4
Moderated regression results for flextime and FSOP
WIF FIW
Step 1
Age .03 .02
Marital status .02 .01
Work hours .13
*
.01
Parental status .06 .16
*
R
2
D.04 .03
Step 2
Flextime .02 .09
FSOP .36
*
.39
*
R
2
D.26
***
.14
***
Step 3
Flextime ·FSOP .23 .06
R
2
D.00 .00
Final F11.64
***
5.46
***
Values represent standardized regression coefficients from the final equation.
*
p< .05.
***
p< .001.
488 K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493
DR
2
= .01, ns). Thus, there was no evidence that FSOP moderates the relationship
between FWA and either direction of WFC.
4. Discussion
The availability of FWA has been touted as a simple and effective way that organiza-
tions can help prevent or buffer their employees’ WFC. However, closer empirical scrutiny
reveals that FWA may not merit such an efficacious reputation. Research investigating
FWA and WFC has produced mixed results, with inconsistencies present not only across
individual studies but even across meta-analyses. Such differences highlight the need for a
more fine-grained analysis of the conditions under which FWA are more or less related to
WFC. With more precise measurement of both flexibility and WFC and an examination of
moderator variables within a sample of employed women, the results of the present study
represent an initial step toward a better understanding of such conditions.
Our first set of hypotheses involved the direct relationship of flextime and flexplace with
WIF and FIW. As predicted, we found that flextime was more strongly related to WIF
than was flexplace. The results align with boundary theory, which asserts that when work
and family are roles are primed in the same location role conflict may occur (Ashforth
et al., 2000). This may be particularly salient for telecommuters who have children at home
while they work. Thus, it appears that flextime is a more effective policy in reducing WIF
than is flexplace.
We also found support for our hypotheses predicting that both flexplace and flextime
were more strongly related to WIF than to FIW. Congruent with the domain specificity
hypothesis, because FWA involve adjustments to the work role, such policies appear to
be more influential in reducing conflicts that originate in the work domain. Taken
Table 5
Moderated regression results for flexplace and FSOP
WIF FIW
Step 1
Age .02 .03
Marital status .03 .00
Work hours .14
*
.01
Parental status .07 .16
*
R
2
D.04 .03
Step 2
Flexplace .07 .39
FSOP .48
***
.53
***
R
2
D.24
***
.14
***
Step 3
Flexplace ·FSOP .02 .43
R
2
D.00 .01
Final F4.69
***
5.69
***
Values represent standardized regression coefficients from the final equation.
*
p< .05.
***
p< .001.
K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493 489
together, these results provide both practical and theoretical implications. From an
applied standpoint, they suggest that organizations can play a role in helping to decrease
one aspect of WFC, WIF, through the offering of FWA, particularly those that offer tem-
poral flexibility. Theoretically, the differential relationships between the types of flexibility
offered and the two directions of WFC illuminate some of the reasons for the discrepant
results across previous studies. Our findings suggest that examining FWA in relation to
WIF is likely to produce different findings than is examining FIW or overall WFC. Sim-
ilarly, definition of flexibility (focus on temporal versus spatial) makes a difference. Precise
definition and measurement are essential to better understanding these complex
relationships.
We proposed that family responsibility would moderate the relationships between flex-
time and flexplace and WIF and FIW. All four interactions were significant. In general,
when family responsibility was high, the relationship between FWA and WFC was nega-
tive. Thus, FWA appears to be uniformly beneficial for women with greater family obli-
gations. However, different results emerged for women with lower family responsibilities
that depended on the direction of WFC. Specifically, no relationship between FWA and
WIF was found among those with lower family responsibility, meaning FWA neither
helps nor harms under these conditions. However, a different pattern emerged for FIW
indicating that there is a positive relationship between FWA and FIW when family
responsibility is low. Thus, FWA is not beneficial and may even theoretically escalate
FIW among women who have few family obligations. This may be because women with
low family responsibility experience little or no FIW. For these women, FWA can only
open the door to allow some greater degree of FIW to occur. On the other hand, those
with greater family responsibility may not experience an increase in FIW because they
have already met the threshold with which they permit interferences into the workplace.
We also examined the moderating role of FSOP on the relationships between flextime
and flexplace and WIF and FIW. None of the interactions were significant, meaning that
the relationship between FWA and WFC is not impacted differentially by the degree that
the organization is perceived as family-supportive. However, this is not to say that FSOP
is not a meaningful variable. FSOP correlated negatively with both WIF and FIW
(r=.49, .37, p< .01, respectively). This is consistent with previous findings (Allen,
2001), and further reinforces the importance of developing organizational environments
that are family-supportive.
Our results have both theoretical and practical implications. Theoretically, the findings
lend further support for distinguishing between FIW and WIF in modeling the work–fam-
ily interface. Moreover, our results suggest that in order to better understand the relation-
ship between workplace benefits and practices, our theoretical models need to incorporate
moderators. Most research to date has examined the direct relationship between policies
such as FWA and WFC or examined mediators of the relationship (e.g., Allen, 2001; Tho-
mas & Ganster, 1995). Models that recognize the conditions under which FWA are more
or less effective are needed. As suggested in the current study, family environment vari-
ables may be an especially promising area of study.
From a practice perspective, policies from the work domain such as flexibility appear
more helpful in reducing WIF than FIW. However, it is noteworthy that the only significant
bivariate relationship was that between flextime and WIF and that this effect size was small.
These findings are interesting when also considering that several studies have found evidence
of a positive relationship between use of FWA and FIW (Hammer, Neal, Newsome, Brock-
490 K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493
wood, & Colton, 2005; Lapierre & Allen, 2006). Our results support the notion that FWA
should not expected to be universally effective in terms of alleviating all forms of WFC for all
employees. Individuals with high family responsibility have the most to gain from flexible
policies. It may be that FWA are more beneficial in terms of increasing productivity than
decreasing WFC (cf., Hill, Miller, Weiner, & Colihan, 1998). FWA appear to still hold
promise as a tool for helping employees manage their work and nonwork lives. However,
the current state of research is such that we need to know much more about the impact of
FWA use in particular on outcomes associated with individual well-being. Flexibility is
not the cure all for managing work and family responsibilities.
There are several limitations associated with the current study. First, in order to reduce
heterogeneity across participants, our sample was limited to women. Additional research is
needed with male populations to determine the generalizability of results across genders.
Second, although we based our study on the theoretical premise that the availability of
FWA will help prevent the occurrence of WFC, the cross-sectional design precludes any
firm conclusions regarding causal direction. However, while experiencing WFC may influ-
ence one to actually use FWA, it seems unlikely that experiencing WFC would increase the
likelihood that participants would report that FWA are available (Batt & Valcour, 2003).
Third, we relied entirely on self-report data, creating concerns about the influence of com-
mon method on the results. However, the large range in the magnitude of the correlations
between variables suggests that a method factor is not likely to be present in the data. None-
theless, data collection from multiple sources (e.g., spousal reports of family responsibility
or organizational reports of FWA offered) would strengthen the integrity of the results.
Lastly, due to the nature of our data collection, we are unable to determine if there are sub-
stantial differences between those who chose to participate in our study and those who did
not. Although descriptive statistics reveal relatively normal distributions on each variable
and mean levels of WFC consistent with previous research, it is conceivable that respon-
dents were more attune to work and family issues that their nonresponding counterparts.
The results of the present study suggest many avenues for future research. We selected
only two variables, one from the family domain and one from the work domain, to test as
moderators between FWA availability and WFC. However, other work and family condi-
tions may also impact this relationship. Future researchers should consider the moderating
effects of other variables in both domains, such as family responsibility or the organiza-
tion’s face-time orientation, or even more personal individual differences reflective of both
domains such as work–family balance self-efficacy. Similarly, we measured the availability
of only two types of FWA, flexplace and flextime. Although we chose these policies
because of their prevalence, investigation of other FWA such as job sharing is certainly
merited. Furthermore, the present study examined the availability, not the actual use, of
flextime and flexplace. This is an important distinction, as use and availability influence
WFC through different mechanisms (Batt & Valcour, 2003; Christensen & Staines,
1990; Grover & Crooker, 1995). Future researchers should consider measuring both access
and utilization to determine whether results differ.
In conclusion, the present study contributes to our knowledge of the influence of FWA
on WFC. Specifically, we examined the different conditions under which FWA were more
or less likely to relate to WFC. The results suggest that the flextime policies have the stron-
gest relationship with WIF and that both flextime and flexplace are more highly related to
WIF than to FIW. Additionally, employees’ family responsibility is an important determi-
nant of the strength of the relationship between FWA and WFC.
K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493 491
References
Adams, G. A., & Jex, S. M. (1999). Relationships between time management, control, work–family conflict, and
strain. Journal of Occupational Health Psychology, 4, 72–77.
Allen, T. D. (2001). Family-supportive work environments: The role of organizational perceptions. Journal of
Vocational Behavior, 58, 414–435.
Allen, T. D., & Shockley, K. (2006). Flexible work arrangements: Help or hype? Paper presented at the Families
and Health Conference, Provo, Utah.
Anderson, S. E., Coffey, B. S., & Byerly, R. T. (2002). Formal organizational initiatives and informal workplace
practice: Links to work–family conflict and job-related outcomes. Journal of Management, 173, 1–24.
Ashforth, B. E., Kreiner, G. E., & Fugate, M. (2000). All in a day’s work: Boundaries and micro role transitions.
Academy of Management Review, 25, 472–491.
Batt, R., & Valcour, P. M. (2003). Human resources practices as predictors of work–family outcomes and
employee turnover. Industrial Relations, 42, 189–220.
Bianchi, S. M., Milkie, M. A., Sayer, L. C., & Robinson, J. P. (2000). Is anyone doing the housework? Trends in
the gender division of household labor. Social Forces, 79, 191–228.
Bond, J. T., Thompson, C., Galinsky, E., & Prottas, D. (2002). Highlights of the national study of the changing
workforce: executive summary. Families and Work Institute.
Byron, K. (2005). A meta-analytic review of work–family conflict and its antecedents. Journal of Vocational
Behavior, 67, 169–198.
Carlson, D. S., Kacmar, K. M., & Williams, L. J. (2000). Construction and initial validation of a
multidimensional measure of work–family conflict. Journal of Vocational Behavior, 56, 249–276.
Christensen, K. E., & Staines, G. L. (1990). Flextime: A viable solution to work/family conflict? Journal of Family
Issues, 11(4), 455–476.
Crompton, R., Brockmann, M., & Lyonette, C. (2005). Attitudes, women’s employment and the domestic
division of labour: A cross-national analysis in two waves. Work, Employment and Society, 19, 213–233.
Eckenrode, J., & Gore, S. (1990). Stress between work and family. New York: Plenum Press.
Frone, Michael R. (2003). Work–family balance. In J. C. Quick & L. E. Tetrick (Eds.), Handbook of occupational
health psychology (pp. 143–162). Washington, DC: American Psychological Association.
Galinsky, E., Bond, J. T., & Friedman, D. E. (1996). The role of employers in addressing the needs of employed
parents. Journal of Social Issues, 52, 111–136.
Galinsky, E., & Stein, P. J. (1990). The impact of human resource policies on employees: Balancing work/family
life. Journal of Family Issues, 11, 368–383.
Golden, L. (2001). Flexible work schedules: Which workers get them? American Behavioral Scientist, 44,
1157–1178.
Goldenhar, D. (2003). The benefits of flexible work arrangements. New York: Advancing Women Professional
and The Jewish Community.
Greenhaus, J. H., Parasuraman, S., Granrose, C. S., Rabinowitz, S., & Beutell, N. J. (1989). Sources of work
family conflict among two-career couples. Journal of Vocational Behavior, 34, 133–153.
Grover, S. L., & Crooker, K. J. (1995). Who appreciates family-responsive human resource policies: The impact
of family-friendly policies on the organizational attachment of parents and non-parents. Personnel
Psychology, 48(2), 271–288.
Hammer, L. B., Allen, E., & Grigsby, T. D. (1997). Work–family conflict in dual-earner couples: Within-
individual and crossover effects of work and family. Journal of Vocational Behavior, 50, 185–203.
Hammer, L. B., Neal, M. B., Newsome, J. T., Brockwood, K. J., & Colton, C. L. (2005). A longitudinal study of
the effects of dual-earner couples’ utilization of family–friendly workplace supports on work and family
outcomes. Journal of Applied Psychology, 90, 799–810.
Hill, E. J., Miller, B. C., Weiner, S. P., & Colihan, J. (1998). Influences of the virtual office on aspects of work and
work/life balance. Personnel Psychology, 51, 667–683.
Hill, E. J., Yang, C., Hawkins, A. J., & Ferris, M. (2004). A cross-cultural test of the work–family interface in 48
countries. Journal of Marriage and Family, 66, 1300–1316.
Honeycutt, T. L., & Rosen, B. (1997). Family friendly human resource policies, salary levels, and salient identity
as predictors of organizational attraction. Journal of Vocational Behavior, 50, 271–290.
Hyland, M. M. (2000). Flexibility in work arrangements: How availability, preferences and use affect business
outcomes. Paper presented at the Annual Meeting of the Academy of Management, Toronto, ON.
492 K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493
Lapierre, L. M., & Allen, T. D. (2006). Work-supportive family, family-supportive supervision, use of
organizational benefits, and problem-focused coping: Implications for work–family conflict and employee
well-being. Journal of Occupational Health Psychology, 11(2), 169–181.
Mesmer-Magnus, J. R., & Viswesvaran, C. (2006). How family–friendly work environments affect work/family
conflict: A meta-analytic examination. Journal of Labor Research, 4, 555–574.
Nippert-Eng, C. (1996). Calendars and keys: The classification of ‘home’ and ‘work’. Sociological Forum, 11,
563–581.
Noonan, M. C., Estes, S. B., & Glass, J. L. (2007). Do workplace flexibility policies influence time spent in
domestic labor? Journal of Family Issues, 28, 263–288.
Osnowitz, D. (2005). Managing time in domestic space: Home-based contractors and household work. Gender
Society, 19, 83–103.
Rau, B. L. (2003). Flexible work arrangements. Sloan Online Work and Family Encyclopedia. Available from
http://wfnetwork.bc.edu/encyclopedia_entry.php?id=240&area=All.
Society for Human Research Management Foundation (2001). SHRM 2001 benefits survey. Society for Human
Resource Management.
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(1), 6–15.
Thompson, C. A., Beauvais, L. L., & Lyness, K. S. (1999). When work–family benefits are not enough: The
influence of work–family culture on benefit utilization, organizational attachment, and work–family conflict.
Journal of Vocational Behavior, 54(3), 392–415.
K.M. Shockley, T.D. Allen / Journal of Vocational Behavior 71 (2007) 479–493 493
