New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Work-Family Conflict and Job Outcomes: The Moderating Effects
of Flexitime Use in a New Zealand Organization
JARROD HAAR*
Abstract
The influence of work-family practices on the outcomes of work-family conflict has received
little attention in New Zealand. The present study of 100 New Zealand employees found workfamily conflict held a significant negative relationship with job satisfaction and organizational
commitment, while the moderating effects of current and future use of flexitime were mixed.
Current flexitime use moderated work-family conflict’s relationship with both job satisfaction
and organizational commitment with the negative influence of work-family conflict on these
outcomes increasing when conflict was high. However, future flexitime use held a significant
buffering effect towards work-family conflict and job satisfaction, with respondents reporting
future flexitime use reporting increased job satisfaction when work-family conflict was high.
The implications for research are discussed.
Keywords: work-family conflict, job outcomes, flexitime, New Zealand
Introduction
Work and family issues have become increasingly important for organizations to consider
(Tenbrunsel, Brett, Maoz, Stroh, & Reilly, 1995). There are a number of factors that have lead to
increased conflict and attention to matters relating to work and family. These include increased
participation rates of working women and working mothers, rising numbers of dual-career
couples and single-parent families, and the enlargement in the elderly population (Milliken,
Martins and Morgan, 1998; Goodstein, 1994, 1995; Morgan and Milliken, 1992). These
demands have lead to escalated demands on individual’s time (Kinnunen, Geurts, & Mauno,
2004; Aryee, Srinivas, & Tan, 2005). These changes have been echoed around the world
including the United States, United Kingdom, Europe, Australasia and New Zealand. For
example, the OECD (2007) reported that average hours actually worked by United States (US)
workers and New Zealander (NZ) workers were very comparable (1797 hours a year in the US
versus 1787 hours a year in NZ). In response to these increased pressures, organizations have
offered a number of human resource practices to aid in better work-family balance. Specifically,
work-family practices have been recognised as allowing employees greater balance of their
work and family roles (Goodstein, 1994; Judge, Boudreau and Bretz, 1994; Osterman, 1995).
Exploring this area is important, because managing the conflict between work and family
responsibilities has become recognised as a critical challenge for organizations (Kossek &
Ozeki, 1998).
*
Dr Jarrod Haar is an Associate Professor in the Department of Strategy & Human Resource Management,
University of Waikato, Private Bag 3105, Hamilton. Email: [email protected]
38
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Work-Family Conflict
Greenhaus and Beutell (1985) defined work-family conflict (WFC) as
“a form of interrole conflict in which the role pressures from the work and family
domains are mutually incompatible in some respect. That is, participation in the
work (family) role is made more difficult by virtue of participation in the family
(work) role” (p. 77).
WFC has been suggested as a source of job stress (Mauno & Kinnunen, 1999), which
consequently leads to negative outcomes. Further, Frone (2000) stated that “work-family
conflict represents a source of stress at the interface of work and family life” (p. 888).
Consequently, the interaction between an employee’s work and family roles can create stress
that ultimately leads to detrimental outcomes. Frone (2000) utilised identity theory to understand
the associations between WFC and outcomes. Frone (2000) stated because both work and family
roles represent core components of adult identity, impediments to both work- and family-related
identity formation and maintenance are likely to be experienced as stressful” (p. 888). This
theory contends that employees reinforce their role identity as workers, parents, partners etc.
through the way they do things. For example, working hard and being productive might
stimulate the worker identity, while engaging in activities with children might stimulate the
parent identity. However, a worker who engages with work too much and too often while
reinforcing their work identity might likewise be unable to engage in children’s activities (by
being at work too often), which would lead to a reduction in other identities such as a parent. As
such, the problems associated with WFC might lead employees to resent their job and
organization.
It has been widely established that WFC has negative aspects, including diminished life
satisfaction (Rice, Frone, & McFarlin, 1992), and poor physical health (Thomas & Ganster,
1995). Importantly, WFC has been found across all work environments (Boles, Johnston, &
Hair, 1997). This paper examines WFC as a predictor of job satisfaction and organizational
commitment. While these relationships have been established in the literature, the influence that
work-family practices might have on these outcomes from has not been well explored.
Consequently, the potential moderating effects of a work-family practice (flexitime) is explored.
This is because flexitime is often seen as a popular and important work-family practice. For
example, Ezra and Deckman (1996) found public sector employees using flexitime reported
more work-family balance. Further, the adoption rates of flexitime are much higher in the public
sector (e.g. 77%, Durst, 1999), than in the private sector (e.g. 41%, Wood, 1999), encouraging
the study of this practice in a public sector organization. The following section now hypotheses
the direct effects of WFC on job outcomes and the potential interacting effects of flexitime on
these direct relationships.
Hypotheses
Job Satisfaction
The literature contains many examples of a negative relationship between WFC and job
satisfaction (Boles et al., 1997; Good, Page, & Young, 1996; Netemeyer, Boles & McMurrian,
1996). However, it must be noted that most of these studies are not able to determine causality.
Thus, increased conflict at work might lead an employee to see the job as less satisfying.
39
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Alternatively, a less satisfying job (e.g. missing the latest promotion), might lead the employee
to bring work issues into the home, creating increased WFC. A meta-analysis by Kossek and
Ozeki (1998) reported significant and consistent negative relationships between job satisfaction
and WFC. Clearly, there is strong theoretical and empirical support for WFC being negatively
related to job satisfaction, and the same is expected in this sample of New Zealand employees.
This leads to the first hypothesis.
Hypothesis 1: WFC will be negatively related to job satisfaction.
The links between WFC and organizational commitment have been much less explored than job
satisfaction. For example, a meta-analysis of WFC and outcome studies (Allen, Herst, Bruck &
Sutton, 2000) noted there are over six times as many studies on job satisfaction as organizational
commitment (38 versus 6 studies respectively). The study maintained that while the correlation
between WFC and organization commitment was moderately strong, the individual study results
were highly inconsistent. While some studies have found significant links between WFC and
organizational commitment (Netemeyer et al., 1996; Lyness & Thompson, 1997; Good, Sisler,
& Gentry, 1988; Good et al., 1996), other studies have been less supportive of this link
(Anderson-Kulman & Paludi, 1986, Wiley, 1987; O’Driscoll, Ilgen & Hildreth, 1992).
Organizational commitment has been defined as an employee’s obligation to the organization
(Taylor, Audia, & Gupta, 1996) and an employee’s willingness to expend personal resources on
behalf of the organization (Mowday, Porter, & Steers, 1982). Further, Shaffer, Harrison, Gilley,
and Luk (2001) asserted that organizational commitment means an employee gives their
commitment in return for a range of benefits and support. As such, increased WFC might lead to
an employee perceiving the organization as failing to provide adequate support, and thus react
with reduced commitment. Alternatively, an employee who feels less obligated to the
organization might be susceptible to taking workplace issues home with them (e.g. higher
WFC). Finally, Wiley (1991) warned that commitment to a particular domain (e.g. work) might
worsen with the level of conflict originating from that domain (e.g. workplace). While findings
have been mixed, there is still strong theoretical support for testing WFC as being negatively
related to organizational commitment. This leads to the second hypothesis.
Hypothesis 2: WFC will be negatively related to organizational commitment.
Moderating Effects of Flexitime
While a number of WFC studies (Frone & Yardley, 1996; Goff, Mount, & Jamison, 1990;
Martins, Eddleston, & Veiga, 2002; Thomas & Ganster, 1995) have explored support as a
moderator (e.g. supervisor, manager, and co-worker support), there have been few studies that
considered work-family practices as a moderator between WFC and job outcomes. This is
despite some studies indicating that flexibility in scheduling work has positive effects on job
outcomes. For example, Thomas and Ganster (1995) found nurse use of flexible schedules was
related to lower WFC. Similarly, Saltzstein, Ting and Saltzstein (2001) found work-family
practices such as flexible and compressed schedules predicted job satisfaction. Marchese,
Bassham and Ryan (2002) asserted that flexitime might reduce WFC, through greater flexibility
in scheduling. They also noted that while it is linked positively with job outcomes, a downside is
that it may not be applicable to some jobs and can cause scheduling problems. While Major,
Klein and Ehrhart (2002) also explored schedule flexibility, it was not found to moderate WFC.
In this paper, the focus is upon flexitime, given that it is among the most common work-family
practices offered in the public sector (Durst, 1999), and is similarly common in the New Zealand
public sector.
40
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Flexitime is a work-family practice that typically allows employees to have flexibility in their
start and finish times, while still remaining within their total work hours (e.g. 40 hour week).
Thus, an employee might start work late to help get children off to school, and make this up by
working late, or having a shorter break for lunch. It has been widely suggested that users of
flexitime, through the ability to provide greater work and family role balance, will enjoy fewer
reductions in job-related outcomes when conflict levels increase. In effect, the ability to better
plan and control one’s working schedule, around both the home and workplace demands, should
enable employees to feel less resentment and dissatisfaction when WFC levels rise. As such,
they may be able to maintain their work identity and non-work identities which might lead to
reductions to negative outcomes when conflict increases.
A recent study expanded the measure of work-family practice use to include future or
anticipated use (Rothausen, Gonzalez, Clarke, & O’Dell, 1998). As such, use of work-family
practices in different time contexts may have distinct effects. To our knowledge, no study has
explored actual use and anticipated use of a work-family practice as a moderator. Employees
may find that flexitime makes it easier for them to do their work and reduce the way work can
interfere with family responsibilities. In addition, employees not currently using flexitime, but
who are experiencing greater conflict might anticipate using flexitime as a means of reducing
the felt conflict. In effect, the ability to tap into flexitime use in the future might have a ‘safety
net’ or buffering effect on conflict outcomes. For example, an upcoming birth of a child may
clearly indicate to an employee that work-family issues will become more complex. As such,
focusing on future flexitime use as a means to reduce this complexity might be highly desirable,
and positive in their current situation. The worker might realise that their WFC is currently high
due to rising workplace issues interfering with the home, such as taking leave for doctor’s visits
etc. However, knowing that they will be able to access flexitime in the future might alleviate the
stressor of WFC and thus lead to lower reductions in job outcomes that those not anticipating
flexitime use in the future. As flexitime conceivably allows employers to better manage work
and family roles, the potential moderating effects of flexitime use are hypothesised as buffering
the negative relationships between WFC and job satisfaction and organizational commitment.
Hypothesis 3: The negative relationship between WFC and job satisfaction will be weakened by
flexitime use. When WFC is high, those using (a) current flexitime use or (b) future flexitime use
will report higher job satisfaction than those not using flexitime.
Hypothesis 4: The negative relationship between WFC and organizational commitment will be
weakened by flexitime use. When WFC is high, those using (a) current flexitime use or (b) future
flexitime use will report higher organizational commitment than those not using flexitime.
Method
Organizational Profile and the Work-Family Context
The organization at the centre of this study is a local government agency, which deals with city
council issues in a large rural community in New Zealand. At the centre of the region is a town
with a population of approximately 60,000 people, and the organization is a major employer in
the region. Hours of work in the organization are 8am to 5:30pm, allowing a wide time frame for
public access, and allowing employees some latitude in their use of flexitime. Employee’s job
roles can be characterised within six major areas: engineering, planning, parks, cultural, clerical
and financial. Conflict arising in the workplace may relate to workload issues, dealing with the
public on controversial or sensitive issues, working long hours due to public meeting
41
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
commitments, or issues regarding monthly deadlines inherent in the public sector. Similar to the
international context, in New Zealand there is a stronger focus on work-family practices in the
public sector than the private sector. While this organization offered six work-family practices,
the present study focuses only upon flexitime due to its universal nature and allowance for
greater flexibility which should allow must influence on work-family conflict (as opposed to
say, paid parental leave).
Sample and Procedures
Data were collected from 206 employees spread over seven sites in a major rural city. Surveys
were sent through the organizations Intranet, which everyone had access to. In an attempt to
reduce the possibility of common method variance where responses to WFC items might
influence responses about the job and organization, two matched surveys were administered
containing different aspects of the study and separated by a one-week time lag. Survey one
contained predictor and demographic variables, while survey two contained the criterion and
moderator variables. Surveys were linked by a private ID code that was used by employees to
identify their two surveys. A total of 100 responses to both surveys were received for a 48.5%
response rate. On average, the participants were 41.7 years old, white (92%), married (77%),
female (69%), and parents (67%). On average, tenure was nine years, respondents earned
between $35,000 and $45,000, and 73% were blue collar and 27% were white collar workers.
Full-time employees made up the majority of respondents (91%). According to the organizations
HR department, the data on respondents to the survey were not significantly different from the
entire organization population.
As used in other work-family studies (e.g. Major et al., 2002), Harman’s one factor test was
conducted as a simple test of common method variance. This requires a global factor analysis
(unrotated), and if a resulting single component is produced, this is likely due to common
method variance (Podsakoff & Organ, 1986). For the present study, this resulted in 13
components, with the largest accounting for 30% of the variance. Similar to Major et al. (2002),
there is no evidence of common method variance.
Measures
WFC was measured by 6-items from Greenhaus, Callanan, and Godshalk (2000), coded
1=strongly disagree, 5=strongly agree. Sample items are “My work schedule often conflicts with
my family life” and “Because my work is demanding, at times I am irritable at home”. Of the six
items, only one item focused on work interference beyond family, that being “On the job, I have
so much work to do that it takes away from my personal interests”. Overall, the measure has
been well validated (Haar, 2006) and shown to be a useful predictor of work-family conflict.
This measure had a Cronbach’s alpha of .89.
Job Satisfaction was measured using Lounsbury and Hoopes’ (1986) 7-item scale, coded
1=extremely dissatisfied and 7=extremely satisfied. Respondents were asked to indicate how
satisfied or unsatisfied they were with different features of their present job, including coworkers, the work itself, pay and fringe benefits, physical surroundings at worksite, immediate
supervisor, promotional opportunities, and their job overall. This measure had a Cronbach’s
alpha of .84.
Organizational Commitment was measured using 15-items from Mowday et al. (1982), coded
1=strongly disagree, 7=strongly agree. Sample items are “I talk up this organisation to my
42
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
friends as a great organisation to work for” and “I feel very little loyalty to this organisation”
(reverse coded). This measure had a Cronbach’s alpha of .88.
Flexitime use was a dichotomous variable, which have been used in work-family conflict
moderation studies (e.g. Frone, 2000). Employees were asked whether they were current
flexitime users (coded 1), anticipated using flexitime (coded 2), or were non-users (coded 0),
similar to the approach used by Rothausen et al. (1998). Employees could choose only one
option. Two dummy variables were then created: current flexitime use (1) and non-users (0), and
future flexitime use (1) and non-user (0).
As per typical work-family conflict studies, gender (1=female, 0=male), marital status
(1=married/de facto, 0=single), family size (total number of children), and total hours worked
(total hours per week) were controlled for (Fu & Shaffer, 2001; Herman & Gyllstrom, 1977;
Major et al., 2002). In addition, the effect of perceived organizational support (Eisenberger,
Huntington, Hutchison, & Sowa, 1986) was controlled so employee perceptions of support were
isolated. This was measured with 15-items, coded 1=strongly disagree, 7=strongly agree.
Questions included “The organisation is willing to help me when I need a special favour” and
“The organisation strongly considers my goals and values”. This measure had a Cronbach’s
alpha of .92. Controlling for the influence of perceived organizational support is important as a
meta-analysis found POS to be a major predictor of job outcomes including job satisfaction and
organizational commitment (Rhoades & Eisenberger, 2002).
Analysis
To examine the direct effects of WFC on job satisfaction and organizational commitment
(Hypotheses 1 and 2), and the potential moderating effects of flexitime use on these
relationships (Hypotheses 3 and 4), separate hierarchical regression analyses were computed.
Control variables (gender, marital status, number of children, total hours worked, and perceived
organizational support) were entered in Step 1. The predictor variable (WFC) was entered at
Step 2. The potential moderator variables (current flexitime use and anticipated flexitime use)
were entered separately in Step 3. Lastly, the interaction variables (predictor multiplied by
moderators) were entered separately in Step 4. Aiken and West’s (1991) centering procedure
was followed, where interaction variables were z-scored. In accordance with Cohen and Cohen
(1983), regression coefficients for the control effects were obtained from Step 1, predictor
effects from Step 2, moderator effects from Step 3, and interaction effects from Step 4.
Results
Descriptive statistics for all variables are shown in Table 1. The mean score for flexitime usage
shows that 27% of respondents currently use flexitime and 17% were anticipating using
flexitime. WFC is significantly correlated with job satisfaction (r= -.42, p< .01), organizational
commitment (r= -.37, p< .01), and perceived organizational support (r= -.40, p< .01) and salary
(r= .20, p< .05). Job satisfaction is significantly correlated with organizational commitment (r=
.60, p< .01), while current flexitime use is significantly correlated with future flexitime use (r= .28, p< .01). Perceived organizational support is significantly correlated with future flexitime
use (r= .23, p< .05), job satisfaction (r= .64, p< .01), and organizational commitment (r= .69, p<
.01).
43
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Table 1. Correlations and Descriptive Statistics of the Study Variables
Variables
M SD
1
2
3
4
5
1. Hours Worked
40. 5.3
-3
2.
Perceived 4.4 .98 .16
-Organizational Support
3. WFC
2.7 .95 .20*
-.40*
*
4. Current Flexitime Use
.27 .45 -.05 -.05
.07
-5. Future Flexitime Use
.17 .38 -.09 .23* -.12
-.28*
*
6. Job Satisfaction
4.6 .92 .16 .64*
-.06
.11
*
.42*
*
7.
Organizational 4.6 .83 .05 .69*
-.06
.14
Commitment
*
.37*
*
N=100. *p<.05, **p<.01.
6
7
--
.60*
*
--
Results of the regressions for Hypotheses 1 to 4 are shown in Tables 2 to 5.
Table 2. Hierarchical Regression Analysis for Moderator Effects of Current Flexitime Use
on WFC towards Job Satisfaction
Job Satisfaction
Variables
Step 1
Step 2
Step 3
Step 4
Controls
Gender
-.27*
-.29*
-.30*
-.30**
Marital Status
.07
.08
.07
.07
Family Size
-.05
-.07
-.08
-.09
Total Hours Worked
.01
.10
.09
.08
Perceived Organizational Support
.59***
.45***
.44***
.45***
Predictor
WFC
Coping Moderators
Current Flexitime Use
Interactions
WFC x Current Flexitime Use
-.33**
-.32**
-.28**
-.09
-.08
-.23**
.05*
R2 change
.39***
.08**
.01
2
Total R
.39
.47
.48
.53
Adjusted R2
.34
.42
.41
.46
F Statistic
7.398***
8.383***
7.270***
7.696***
† p< .1, * p< .05, ** p< .01, *** p< .001, Standardized regression coefficients. All
significance tests were single-tailed.
44
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Table 3. Hierarchical Regression Analysis for Moderator Effects of Current Flexitime
Use on WFC towards Organizational Commitment
Organizational Commitment
Variables
Step 1
Step 2
Step 3
Step 4
Controls
Gender
.07
.05
.03
.03
Marital Status
.17
.18
.17
.16
Family Size
.02
.01
-.01
-.01
Total Hours Worked
-.02
.04
.04
.03
Perceived Organizational Support
.72***
.62***
.60***
.61***
Predictor
WFC
-.24*
Coping Moderators
Current Flexitime Use
-.22*
-.19†
-.17†
-.16†
Interactions
WFC x Current Flexitime Use
-.16*
.52***
.04*
.03†
.03†
R2 change
2
Total R
.52
.56
.59
.61
Adjusted R2
.48
.52
.54
.56
F Statistic
12.365*** 11.967*** 11.199*** 10.728***
† p< .1, * p< .05, ** p< .01, *** p< .001, Standardized regression coefficients. All significance
tests were single-tailed.
Table 4. Hierarchical Regression Analysis for Moderator Effects of Future Flexitime Use
on WFC towards Job Satisfaction
Job Satisfaction
Variables
Step 1
Step 2
Step 3
Step 4
Controls
Gender
-.27*
-.29*
-.29*
-.28*
Marital Status
.07
.08
.08
.10
Family Size
-.05
-.07
-.06
-.07
Total Hours Worked
.01
.10
.10
.06
Perceived Organizational Support
.59***
.45***
.44***
.43***
Predictor
WFC
-.33**
Coping Moderators
Future Flexitime Use
-.33**
-.36**
.02
.06
Interactions
WFC x Future Flexitime Use
R2 change
Total R2
.22*
.39***
.39
.08**
.47
.00
.48
.04*
.53
45
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Adjusted R2
.34
.42
.41
.46
F Statistic
7.398*** 8.383*** 7.070*** 7.126***
† p< .1, * p< .05, ** p< .01, *** p< .001, Standardized regression coefficients. All
significance tests were single-tailed.
Table 5. Hierarchical Regression Analysis for Moderator Effects of Future
Flexitime Use on WFC towards Organizational Commitment
Organizational Commitment
Variables
Step 1
Step 2
Step 3
Step 4
Controls
Gender
.07
.05
.05
.06
Marital Status
.17
.18
.18
.19
Family Size
.02
.01
.01
.01
Total Hours Worked
-.02
.04
.04
.03
Perceived Organizational
.72***
.62***
.62***
.62***
Support
Predictor
WFC
Coping Moderators
Future Flexitime Use
Interactions
WFC x Future Flexitime Use
-.24*
-.23*
-.25*
-.02
-.01
.08
R2 change
Total R2
Adjusted R2
F Statistic
.01
.52***
.04*
.00
.52
.56
.56
.57
.48
.52
.51
.50
12.365** 11.967** 10.092** 8.884***
*
*
*
† p< .1, * p< .05, ** p< .01, *** p< .001, Standardized regression coefficients. All
significance tests were single-tailed.
WFC was significantly related to job satisfaction (β= -.33, p< .01) and organizational
commitment (β= -.24, p< .05), which supports Hypotheses 1 and 2. From Step 2 it can be seen
that WFC accounted for 8% (p< .01) of variance for job satisfaction, while only 4% (p< .05) of
the variance for organizational commitment. Current flexitime use had a significant interaction
effect on WFC and job satisfaction (ß= -.23, p< .01), accounting for an additional 5% (p< .05) of
the variance and on WFC and organizational commitment (ß= -.16, p< .05), accounting for an
additional 3% (p< .1) of the variance. However, given these effects are both negative and not
positive (buffering) as hypothesized, this offers no support for Hypotheses 3 (a) or 3(b). While
future flexitime use did not have a significant interaction effect on WFC and organizational
commitment (ß= .08, non significant), future flexitime use did have a significant interaction
effect on WFC and job satisfaction (ß= .22, p< .05), accounting for an additional 4% (p< .05) of
the variance. This positive effect supports Hypothesis 4(a) but not Hypothesis 4(b).
46
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
To provide greater understanding of the moderating effects of current and future flexitime use,
and to facilitate interpretation of the significant effects on job outcomes, plots of the interactions
are presented in Figure 1 (WFC-job satisfaction-current flexitime use), Figure 2 (WFCorganizational commitment-current flexitime use), and Figure 3 (WFC-job satisfaction-future
flexitime use). On these Figures, WFC high and low represent points above and below the mean
(M=2.7), and this is the same for the graphed lines for current flexitime use (M=.27) and future
flexitime use (M=.17).
Figure 1. Interaction Effects of Current Flexitime Use towards WFC and Job Satisfaction.
Plotting the interaction terms (Figure 1) illustrates that when WFC is low, there is little
difference in job satisfaction between current flexitime users and non-users. However, at high
levels of WFC, current flexitime users report greater reductions in job satisfaction than current
flexitime non-users.
47
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Figure 2. Interaction Effects of Current Flexitime Use towards WFC and Organizational
Commitment.
Plotting the interaction terms (Figure 2) illustrates that when WFC is low, there is little
difference in organizational commitment between current flexitime users and non-users.
However, as with job satisfaction (Figure 1), at high levels of WFC, current flexitime users
report greater reductionss than current flexitime non-users.
Figure 3. Interaction Effects of Future Flexitime Use towards WFC and Job Satisfaction.
Plotting the interaction terms (Figure 3) illustrates that when WFC is low, there is a significant
difference, with future flexitime users reporting lower job satisfaction than future flexitime nonusers. However, at high levels of WFC, future flexitime users report a significant increase in job
satisfaction, while anticipated flexitime non-users report a significant decrease in job
satisfaction.
Overall, the regression models for predicting job satisfaction and organizational commitment
were all significant and accounted for large amounts of variance: WFC-job satisfaction-current
flexitime use (R2 = .53, F = 7.696, p< .001), WFC-organizational commitment-current flexitime
use (R2 = .61, F = 10.728, p< .001), WFC-job satisfaction-future flexitime use (R2 = .53, F =
7.126, p< .001), and WFC-organizational commitment-future flexitime use (R2 = .57, F = 8.884,
p< .001). I also tested the variance inflation factors (VIF) for evidence of multicollinearity,
which occurs at values of 10.0 or higher (Neter, Kutner, Nachtsheim, & Wasserman, 1996). For
the present study all the scores were below 1.5 in both regression models, indicating
multicollinearity did not unduly influence the regression estimates.
48
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Discussion
The major focus of this paper was to explore the potential moderating effects of flexitime use.
The findings indicate that for employees in this organization, current use of flexitime does not
reduce the negative relationships with job satisfaction and organizational commitment, but
instead increases them. When WFC is high, current flexitime users have significantly lower job
satisfaction and organizational commitment than current flexitime non-users. This finding is
important because it suggests use of flexitime may not be inherently positive. One reason for
this negative effect may be that employees in this organization are using flexitime as a reaction
to workplace conflict rather than as a way to control workplace issues. While this finding does
not provide the evidence work-family critics have called for regarding confirmation of the
beneficial nature of work-family practices (Frone and Yardley, 1996; Gonyea & Googins, 1992;
Kingston, 1990), it does offer greater understanding of employee motives for using flexitime. It
appears that rather than facilitating balance, the negative links with satisfaction and commitment
might indicate employees blame their workplace as the cause of their conflict, and while they
might try to manage this conflict better through flexitime use, the organization ultimately pays
through reduced employee commitment to the organization, and satisfaction with the job. Of
course, further replication in other organizations is needed to determine whether this negative
effect is generalizable.
Despite these negative findings, anticipated flexitime use was found to have positive effects,
with future users reporting increased job satisfaction, while respondents not planning on using
flexitime in the future reporting decreased job satisfaction. This suggests that those planning on
using flexitime might perceive benefits from using the work-family practice, and this actually
enhances their satisfaction with their job when workplace conflict increases. Further studies into
what benefits employees believe they might receive from using flexitime (e.g. reduced stress)
may be an appropriate factor in predicting flexitime usage. Additionally, in the context of the
findings of the present study regarding current flexitime use, whether this perception is
maintained when anticipated users become actual users would also be a useful option for future
study.
The present study argued that WFC would negatively influence the job outcomes explored
because under identity theory, workers who focused too much effort and time in their workplace
would likely lead to issues in their family, which would ultimately lead to lower satisfaction
with their job and commitment towards their organization. Importantly, the interaction effects of
current flexitime use showed that those who use flexitime had more detrimental outcomes when
WFC increased. Applied to identity theory, this indicates that flexitime might add to the
detrimental relationships between WFC and job outcomes because employee identities might be
seen as being further eroded because they need to rely on a work-based practice like flexitime.
In effect, the work identity is creating detrimental effects on the family, perhaps through
excessive work (i.e. higher WFC) which leads to lower job outcomes. But, by relying on
flexitime, employees might be further alienating other identities due to seeking solutions based
on their work identity, as flexitime is an employer offered practice. The difference with future
flexitime use is this is an anticipated rather than actual utilization, and perhaps it’s not until
employees use flexitime that they find it is unable to reduce detrimental outcomes compared to
how they thought it might help. Hence, the differences we find here between actual and future
flexitime use might be aligned to reality versus expectation.
49
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Overall, the present study provides empirical evidence that flexitime can have a significant
moderating effect on work-family conflict outcomes. However, whether this outcome is positive
or negative may depend on when the flexitime is used. Replication of these effects is needed
before we can consider consigning flexitime use to being a reaction to conflict. Nevertheless, for
the present organization, it appears that employees may use flexitime as a reaction to increased
conflict, rather than as a mechanism for reducing conflict. The additional analysis on users of
flexitime found that employees who are married and/or parents are more likely to be current
flexitime users. It might also be that these individuals are exacerbating their conflict occurring in
the workplace by using flexitime for reasons that do not help them balance their workplace
issues. For example, employees might use flexitime to relieve issues with their partner or
dependants, rather than workplace issues. Exploring what flexitime is used for was beyond the
scope of the present study, but does offer an area for future research.
Finally, it is worth noting that while the overall models for predicting job satisfaction and
organizational commitment were high with Total R2 values above .50, it can be shown from Step
1 that these models are significantly influenced by the measure of perceived organizational
support. Given the strong correlations in the present study and the meta-analysis findings linking
these variables (Rhoades & Eisenberger, 2002), that is not unexpected. However, it does show
that WFC can influence job satisfaction and organizational commitment over and above the
influence of perceived organizational support, and likewise, that the interactions with flexitime
enhance the models for these outcomes, highlighting the importance of the effects tested in the
present study.
Limitations
As with all research, there are limitations that encourage caution when interpreting these results.
This paper is based upon a single organization and using a single work-family practice as a
moderator does limit the generalisability of the findings. Further studies, particularly with other
work-family practices would be desirable. In addition, the use of self reported data is
problematic, although Major et al. (2002) has noted that self-reported data use is typical in workfamily conflict research. As noted in the methods section, from Harman’s one factor test there
appears no evidence to support common method variance being an issue. Another issue relates
to the outcomes analysed here, with job satisfaction and organizational commitment being
highly correlation here and this is typical of these measures. As such, other, unrelated outcomes
should be explored, to gain a greater understanding of how flexitime affects outcomes. For
example, will these effects hold towards turnover intentions or psychological distress? In
addition, the small sample size (n=100) is a factor of New Zealand based research, where only
0.5 percent of firms employee over 100 employees (New Zealand Statistics, 2006). This makes
the organization used in the present study here ‘large’ by New Zealand standards, if not
internationally. Finally, the sole focus on WFC might be viewed as a limitation, although other
moderating studies have also focused solely upon WFC (e.g. Martins et al., 2002).
Conclusion
This paper has some unique aspects; including being one of only a few work-family conflict
studies not set in the United States and Europe, and indicates that the negative outcomes from
work-family conflict may be generalizable in a variety of international settings. In addition, this
study finds organizational commitment to be negatively related to WFC, which has been an
under explored outcome in the work-family conflict literature. Further, the use of flexitime does
50
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
offer interesting findings with current use increasing the negative outcomes for job satisfaction
and organizational commitment, and future use decreasing the job satisfaction decline. In the
context of this single organization, it suggests employee use of flexitime can have a significant
effect, although the effects are not necessarily be a way of reducing negative outcomes. The
findings encourage further research into employee motivations behind flexitime use, especially
with regards to what benefits are expected and what the reality may be for users. Overall, the
focus here on flexitime begins to meet the call of Frone and Yardley (1996), and provides useful
directions for future research.
References
Aiken, L.S and West, S.G. (1991). Multiple Regression: Testing and Interpreting Interactions.
Newbury Park: Sage
Allen, T.D., Herst, D.E.L., Bruck, C.S. and Sutton, M. (2000). Consequences associated with
Work-to-Family Conflict: A Review and Agenda for Future Research. Journal of Occupational
Health Psychology. 5(2): 278-308
Anderson-Kulman, R.E. and Paludi, M.A. (1986). Working Mothers and the Family Context:
Predicting Positive Coping. Journal of Vocational Behavior. 28: 241-253
Aryee, S., Srinvias, E.S., and Tan, H.H. (2005). Rhythms of Life: Antecedents and Outcomes of
Work-Family Balance in Employed Parents. Journal of Applied Psychology. 90(1): 132-146
Boles, J.S., Johnston, M.W., and Hair Jr., J.F. (1997). Roles Stress, Work-Family Conflict and
Emotional Exhaustion: Inter-relationships and Effects on some Work-related Consequences. The
Journal of Personal Selling and Sales Management. 17(1): 17-28
Cohen, J. and Cohen, P. (1983). Applied Multiple Regression/Correlation Analysis for the
Behavioural (2nd ed). Hillsdale: Erlbaum
Durst, S.L. (1999). Assessing the Effects of the Family Friendly Programmes on Public
Organizations. Review of Public Personnel Administartion. Summer: 19-33
Eisenberger, r. Huntington, R., Hutchison, S. and Sowa, D. (1986). Perceived Organizational
Support. Journal of Applied Psychology. 71(1): 500-507
Ezra, M., and Deckman, M. (1996). Balancing Work and Family Responsibilities: Flextime and
Childcare in the Federal Government. Public Administration Review. 56(2): 174-179
Frone, M.R. (2000). Work-family Conflict and Employee Psychiatric Disorders: The National
Comorbidity Survey. Journal of Applied Psychology. 85(6): 888-895
Frone, M.R. and Yardley, J.K. (1996). Workplace Family-Supportive Programmes: Predictors of
Employed Parents’ Importance Ratings. Journal of Occupational and Organizational
Psychology. 69(4): 351-366
Fu, C.K. and Shaffer, M.A. (2000). The Tug of Work and Family: Direct and Indirect DomainSpecific Determinants of Work-Family Conflict. Personnel Review. 30(5): 502-522
51
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Goff, S.J., Mount, M.K. and Jamison, R.L. (1990). Employer Supported Child Care,
Work/Family Conflict and Absenteeism: A Field Study. Personnel Psychology. 43: 793-809
Gonyea, J.G. and Googins, B.K. (1992). Linking the Worlds of Work and Family: Beyond the
Productivity Gap. Human Resource Management. 31(3): 209-226
Good, L.K., Page, T.J., and Young, C.E. (1996). Assessing Hierarchical Differences in JobRelated Attitudes and Turnover among Retail Managers. Academy of Marketing Science. 24(2):
148-156
Good, L.K., Sisler, G.F., and Gentry, J.W., (1998). Antecedents of Turnover Intentions among
Retail Management. Journal of Retailing. 64(3): 295-314
Goodstein, J.D. (1994). Institutional Pressures and Strategic Responsiveness: Employer
Involvement in Work-Family Issues. Academy of Management Journal. 37(2): 350-382
Goodstein, J.D. (1995). Employer Involvement in Eldercare: An Organizational Adaptation
Perspective. Academy of Management Journal. 38(6): 1657-1671
Greenhaus, J.H., Callanan, G.A. and Godshalk, V.M. (2000). Career Management (3rd 3d).
Orlando: The Dryden Press.
Greenhaus, J.H. and Beutell, N.J. (1985). Sources of Conflict Between Work and Family Roles.
Academy of Management Review. 10(1): 76-88
Haar, J.M. (2006). The Downside of Coping: Work-Family Conflict, Employee Burnout and the
Moderating Effects of Coping Strategies. Journal of Management and Organization. 12(2): 146159
Herman, J.B. and Gyllstrom, K.K. (1977). Working Men and Women: Inter and Intra-role
Conflict. Psychology of Women Quarterly. 1: 319-333
Judge, T.A., Boudreau, J.W. and Bretz Jr. R.D. (1994). Job and Life Attitudes of Male
Executives. Journal of Applied Psychology. 79(5): 767-782
Kingston, P.W. (1990). Illusions and Ignorance about the Family-Responsive Workplace.
Journal of Family Issues. 11: 438-454
Kinnunen, U., Geurts, S. and Mauno, S. (2004). Work-to-Family Conflict and its Relationship
with Satisfaction and Well-Being: A One-Year Longitudinal Study on Gender Differences.
Work and Stress. 18: 1-22
Kossek, E.E. and Ozeki, C. (1998). Work-Family Conflict, Policies and the Job-Life Satisfaction
and Relationship: A Review and Directions for Organizational Behavior-Human Resources
Research. Journal of Applied Psychology. 83(2): 139-149
Lounsbury, J.W., and Hoopes, L.L. (1986). A Vacation from Work: Changes in Work and
Nonwork Outcomes. Journal of Applied Psychology. 71(3): 395-401
52
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Lyness, K.S. and Thompson, D.E. (1997). Above the Glass Ceiling? A Comparison of Matched
Samples of Female and Male Executives. Journal of Applied Psychology. 82: 359-375
Major, V.S., Klein, K.J., and Ehrhart, M.G. (2002). Work time, Work interference with Family
and Psychological Distress. Journal of Applied Psychology. 87(3): 427-436
Major, V.S., Klein, K.J., and Ehrhart, M.G. (2002). Work time, Work Interference with Family
and Psychological Distress. Journal of Applied Psychology. 87(3): 427-436
Marchese, M.C., Bassham, G. and Ryan, J. (2002). Work-Family Conflict: A Virtue Ethics
Analysis. Journal of Business Ethics. 40(2): 145-154
Martins, L.L., Eddleson, K.A. and Veiga, J.F. (2002). Moderators of the Relationship Between
Work-Family Conflict and Career Satisfaction. Academy of Management Journal. 45(2): 399409
Mauno, S. and Kinnunen, U. (1999). The Effects of Job Stressor on Marital Satisfaction in
Finnish Dual-Earner Couples. Journal of Organizational Behavior. 20(6): 879-895
Milliken, F.J., Martins, L.L., and Morgan, H. (1998). Explaining Organizational Responsiveness
to Work-Family Issues: The Role of Human Resource Executives as Issue Interpreters. Academy
of Management Journal. 41(5): 580-592
Morgan, H., and Milliken, F.J. (1992). Key to Action: Understanding the Differences in
Organizations’ Responsiveness to Work-and-Family Issues. Human Resource Management.
31(3): 227-248
Mowday, R.T., Porter, L.W. and Steers, R.M. (1982). Employee-Organization Linkages. New
York: Academic Press.
Netemeyer, R.G., Boles, J.S. and McMurrian, r. (1996). Development and Validation of Workfamily Conflict and Family-Work Conflict Scales. Journal of Applied Psychology. 81(4): 400410
Neter, J., Kutner, M.H., Nachtsheim, C.J. and Wasserman, W. (1996). Applied Linear Statistical
Models. Chicago: Irwin
O’Driscoll, M.P., Llgen, D.R. and Hildreth, K. (1992). Time Devoted to Job and Off-job
Activities, Interrole Conflict and Affective Experiences. Journal of Applied Psychology. 77:
272-279
OECD. (2007). OECD Employment Outlook. OEDC. Paris
Osterman, P. (1995). Work/Family Programmes and the Employment Relationship.
Administrative Science Quarterly. 40(4): 681-700
Podaskoff, P.M. and Organ, D.W. (1996). Self-reports in Organizational Research: Problems
and Prospects. Journal of Management. 12(4): 531-544
53
New Zealand Journal of Employment Relations 2008, 33(1): 38-54
Rhoades, L. and Eisenberger, R. (2002). Perceived Organizational Support: A Review of the
Literature. Journal of Applied Psychology. 87(4): 698-714
Rice, R.W., Frone, M.R., and McFarlin, D.B. (1992). Work-Nonwork Conflict and the
Perceived Quality of Life. Journal of Organizational Behavior. 13: 155-168
Rothausen, T.J., Gonzalez, J.A., Clarke, N.E., and O’Dell, L.L. (1998). Work-Family Backlash
– Fact or Fiction? The Case of Organizations’ On-Site Child Care Centres. Personnel
Psychology. 51: 685-706
Saltztine, A.L., Ting, Y. and Saltzstine, G.H. (2001). Work-Family Balance and Job
Satisfaction: The Relationship of Family-friend Policies on Attitudes of Federal Government
Employees. Public Administration Review. 61(4); 452-467
Shaffer, M.A., Harrison, D.A., Gilley, K.M., and Luk D.M. (2001). Struggling for Balance amid
Turbulence on International Assignments: Work-Family Conflict, Support and Commitment.
Journal of Management. 27: 99-121
Statistics New Zealand. (2006). Business Demographic Statistics Review Report Based on
Statistics New Zealand’s Longitudinal Business Frame: May 2006. Wellington: New Zealand
Statistics.
Taylor, M.S., Audia, G. and Gupta, A.K. (1996). The Effect of Lengthening Job Tenure on
Managers’ Organizational Commitment and Turnover. Organizational Science. 7: 632-648
Tenbrunsel, A.E., Brett, J.M., Maoz, E., Stroh, L.K. and Reilly, A.H. (1995). Dynamic and
Static Work-Family Relationship. Organizational Behavior and Human Decision Processes.
63(3): 233-246
Thomas, L. and Ganster, D. (1995). Impact of Family-Supportive Work Variable on a WorkFamily Conflict and Strain: A Control Perspective. Journal of Applied Psychology. 80(1): 6-15
Wiley, D.L. (1987). The Relationship Between Work/Nonwork Role Conflict and job Related
Outcomes: Some Unanticipated Findings. Journal of Management. 13: 467-472
Wiley, M.G. (1991). Gender, Work and Stress: The Potential Impact of Role Identify Salience
and Commitment. The Sociological Quarterly. 32: 495-510
Wood, S. (1999). Family-Friendly Management: Testing the Various Perspectives. National
Institute Economic Review. April: 99-116
54