Vous êtes sur la page 1sur 8

Children and Youth Services Review 35 (2013) 14471454

Contents lists available at SciVerse ScienceDirect

Children and Youth Services Review

journal homepage: www.elsevier.com/locate/childyouth

Relationships between worklife balance and job-related factors among

child welfare workers
Lei Wu a,, Binahayati Rusyidi b, Nancy Claiborne c, Mary L. McCarthy c
Department of Social Work, School of Sociology and Population Studies, Renmin University of China, Beijing, China
Department of Social Welfare, Faculty of Social and Political Sciences, University of Padjadjaran, West Java, Indonesia
Social Work Education Consortium, University at Albany, Albany, NY, USA

a r t i c l e i n f o a b s t r a c t

Article history: This study examined the relationships between worklife balance and several job-related factors among 573
Received 29 June 2012 public child welfare workers in a northeastern state in the United States of America. It explored job-related
Received in revised form 28 May 2013 correlates of worklife balance and the possible mediating role of worklife balance between these factors
Accepted 28 May 2013
and job satisfaction. Multiple regression analysis demonstrated that the job-related factors organizational
Available online 7 June 2013
support, job value, work time, and income were signicantly associated with worklife balance among child
welfare workers. It also conrmed that worklife balance partially mediated the associations of organizational
Worklife balance support and job value with job satisfaction. Implications for child welfare research, policy, and practice are
Child welfare workers proposed based on the empirical ndings.
Job-related factors 2013 Elsevier Ltd. All rights reserved.

1. Introduction possible job-related correlates of worklife balance in child welfare.

On the other hand, working conditions are likely to result in low job
In the context of personal life and family demands in contemporary satisfaction, which has powerful and far reaching consequences for
society (Carlson, Grzywacz, & Zivnuska, 2009; Hobson, Delunas, & Kesic, social workers and organizations (Lambert, Pasupuleti, Cluse-Tolar,
2001; Lewis, 2003; Moen & Yu, 2000), the capacity of balancing Jennings, & Baker, 2006). Given the evidence-based inuence of work
employees' work and personal lives is essential for the well-being of life balance in shaping job satisfaction (Lambert et al., 2006; Losoncz &
workforce. A national survey (Hobson et al., 2001) disclosed that the Bortolotto, 2009), the role of worklife balance as a mediator of the
10 most stressful life events for US workers revolved around major relationship between working conditions and job satisfaction merits
life and family issues. As working couples have become the norm in close attention.
society (Moen & Yu, 2000), rising household nancial and care needs Worklife balance has been intensively studied in business, manage-
(Hobson et al., 2001), together with increasing expectations for personal ment, and other disciplines (e.g., Carlson et al., 2009; Connell, 2005;
satisfaction and enjoyment (Lewis, 2003), have a high propensity to Dean, 2007; Greenhaus, Collins, & Shaw, 2003; Hobson et al., 2001;
cause conicts between employees' work responsibilities and their Jang, 2009; Kossek, Lewis, & Hammer, 2010; Lewis & Campbell, 2008;
personal life demands (Carlson et al., 2009). Workers, including in Losoncz & Bortolotto, 2009; Mescher, Benschop, & Doorewaard,
child welfare, are obliged to balance their work and life roles and 2010; Ollier-Malaterre, 2010; Virick, DaSiva, & Arrington, 2010). How-
domains simultaneously. It is therefore necessary to understand their ever, it is still a novel research topic in social work, let alone child
worklife balance by administrators and researchers. welfare research. Some studies explored the association between work
Child welfare is regarded as a demanding and stressful eld with life balance and job retention in child welfare (Smith, 2005; Strolin-
challenging working conditions. It entails large caseloads, long hours, Goltzman, Auerbach, McGowan, & McCarthy, 2007), but various job-
on-call responsibilities, inadequate compensation, insufcient supervi- related correlates of worklife balance have been under-examined so
sion and training, lack of adequate resources to serve children and far. Although prior research asserted the impacts of worklife balance
their families, stringent state and federal policy requirements, etc. on social work and human service worker job satisfaction (Lambert
(Westbrook, Ellis, & Ellett, 2006). On one hand, working conditions are et al., 2006), the mediating effect of worklife balance on job satisfaction
likely to worsen the existing conicts between work and life among is still an untouched issue among child welfare workforce.
child welfare workers. This suggests a strong need to understand the This study seeks to examine the relationships between worklife
balance and job-related factors supervisor support, organizational sup-
port, job value, work time, income, supervisory status, tenure, and job
Corresponding author at: Department of Social Work, School of Sociology and
Population Studies, Renmin University of China, 59 Zhongguancun Street, Haidian District,
satisfaction among 573 public child welfare workers in a northeastern
Beijing 100872, China. state in the United States of America. It explores job-related correlates
E-mail address: leiwucn@gmail.com (L. Wu). of worklife balance and the possible mediating role of worklife

0190-7409/$ see front matter 2013 Elsevier Ltd. All rights reserved.
1448 L. Wu et al. / Children and Youth Services Review 35 (2013) 14471454

balance between these factors and job satisfaction. Implications for Bretz, 1992). If appreciating the value of child welfare work can alter
child welfare research, practice, and policy are proposed based on the employees' job facet perceptions and decision, it may facilitate the
empirical ndings. workers' perception of worklife balance. However, the relationship
between job value and worklife balance is still unknown.
2. Literature review
2.1.4. Worklife balance and other job-related factors
2.1. Worklife balance and its correlates The job-related factors work time, supervisory status, income, and
tenure may be correlates to worklife balance among child welfare
2.1.1. Worklife balance workers. In their study of working mothers, Losoncz and Bortolotto
Research on the balance between individuals' workplace, household, (2009) conrmed that negative worklife balance was associated
and other aspects of life has developed over time (Dean, 2007; Jang, with long working hours and work overload. If an employee allots
2009; Lewis, 2003). Initially the studies emphasized on workfamily more time for work, he or she will lack time and energy for personal
balance and later on moved to worklife balance (Lewis & Campbell, and family life thus resulting in conicts between work and life domains.
2008). The term balance was both distinct from and interrelated Similarly, supervisors, undertaking much more job responsibility, pres-
with conict and enrichment in work and family domains (Carlson sure, and intensive work content, hence usually with higher income
et al., 2009). In general, worklife balance encompassed broader activi- than their subordinates, may have degraded worklife balance. In addi-
ties than workfamily conict and was keen on gender equality. While tion, an employee with longer tenure in the position is highly likely to be
workfamily conict focused on care for family dependents and tensions experienced in the job and have adapted himself to it, thus having a
resulting from incompatible demands of work intruding on family better worklife condition. However, work time, supervisory status,
responsibilities, worklife balance addressed non-earning activity and income, and tenure have rarely been examined in worklife literature
servicing work and personal life needs for all employees (Carlson et al., of child welfare workers.
2009; Fenner & Renn, 2010; Lewis, 2003; Lewis & Campbell, 2008).
The major concern of the research in worklife area was how to inte-
grate work and personal life of the workforce (Lambert et al., 2006). 2.1.5. Worklife balance and socio-demographic factors
Extant research from various elds has revealed the negative out- Socio-demographic variables gender, age, race, and locality have
comes of worklife imbalance and the benecial consequences of been treated as control variables in relevant research (Lambert et al.,
worklife balance for individuals and organizations. Worklife imbalance 2006; McGowan, Auerbach, & Strolin-Goltzman, 2009). Gender has
negatively affects productivity, absenteeism, organizational commitment, been strongly emphasized in worklife discussions (Connell, 2005;
turnover, parenting problems, domestic violence, stress, illness, life Lewis, 2003; Lewis & Campbell, 2008; Mescher et al., 2010; Moen & Yu,
satisfaction, etc. (Greenhaus et al., 2003; Hobson et al., 2001) and is 2000). The gendered portrayal of worklife balance practices (Lewis,
positively related to job satisfaction, organizational commitment, family 2003; Mescher et al., 2010) asks for examination of worklife balance
functioning, and well-being of employees (Carlson et al., 2009; Jang, between gender groups. It is obvious that older workers have fewer fam-
2009; Ollier-Malaterre, 2010). ily responsibilities and personal development demands than young and
The goal of this study is to explore correlates of worklife balance. It middle-age workers. Therefore, it may be valid to assume that older
concerns such job-related factors as supervisor support, organizational workers may face lessened worklife imbalance than their younger
support, job value, work time, income, supervisory status, and tenure. counterparts. Race has rarely been examined in previous worklife
Socio-demographic factors are also included as controls. research, but its relationship with crucial workforce constructs such as
organizational commitment in social work (Lambert et al., 2006)
2.1.2. Worklife balance and supervisor and organizational support reminds us of the possible ethnic differences in worklife balance.
In work environment, supervisor support was social support for Although the rural/urban differences in job satisfaction, intention to
employees granted by supervisors. Employees also received support leave, and retention were examined in detail among child welfare
from co-workers and organizations where they perceived that they workers (Barth, Lloyd, Christ, Chapman, & Dickinson, 2008; Strolin-
were valued and cared about in the organizations (Stamper & Johlke, Goltzman et al., 2007; Westbrook et al., 2006), few studies have looked
2003). Hopkins (2005) stated that supports from supervisors and orga- at worklife balance across localities. Only McGowan et al. (2009)
nizations were required in successful integration of employees' work reported that urban workers were lower in worklife balance than
and lives. In organizations without ofcially sanctioned worklife poli- rural workers in their research on intention to leave. Hence, socio-
cies or programs, supervisor support was crucial in helping workers demographic factors gender, age, race, and locality are included as
strike a worklife balance. This was conrmed by Jang (2009) who stated control variables in the present study.
that both perceived supervisor support and workplace support were
related to worklife balance for working parents. Since support could 2.2. Mediating effect of worklife balance on job satisfaction
mitigate conict and stress caused by work and life responsibilities, it
is expected that child welfare workers will experience better work Job satisfaction represented employees' evaluative feelings about
life conditions if they receive more supervisor and organizational the job. This received considerable attention over the past several
support. decades (Sharma et al., 1997; Spector, 1985). Several studies have
examined characteristics and predictors of job satisfaction in child
2.1.3. Worklife balance and job value welfare settings (Auerbach, McGowan, Ausberger, Strolin-Goltzman, &
Job value, i.e., appreciation of the value of child welfare work (Smith, Schudrich, 2010; Barth et al., 2008; Sharma et al., 1997; Vinokur-
2005) and its relation with worklife balance are still under-specied in Kaplan, 1991). Empirical research indicated that age (Spector, 1985),
child welfare. Only one study in child welfare involved job value and it salary (Spector, 1985), locality (Barth et al., 2008), supervisor support
indicated that job value was not associated with child welfare workers' (Rauktis & Koeske, 1994), supervision (Barth et al., 2008), and tenure
job retention (Smith, 2005). However, a study of nurses in Australia (Lambert et al., 2006) had their roles in explaining job satisfaction for
appeared to challenge this nding because the study found that intrinsic social workers or child welfare workers. For nurses and other work-
and extrinsic work value did impact upon intention to leave employ- forces, perceived organizational support (Bradley & Cartwright, 2002;
ment and job satisfaction (Hegney, Plank, & Parker, 2006). Studies con- Lee & Cummings, 2008; Stamper & Johlke, 2003), job values (Blood,
rmed that work value altered individuals' subjective nature of work 1969; Bokemeier & Lacy, 1987; Hegney et al., 2006; Kwak, Chuang,
experience (Miller, 1980) and affected job choice decision (Judge & Xu, & Eun-Jung, 2010; Meglino, Ravlin, & Adkins, 1989; Miller, 1980;
L. Wu et al. / Children and Youth Services Review 35 (2013) 14471454 1449

Stamper & Johlke, 2003), work time (Miller, 1980), and position Retention Survey which was funded by the U.S. Children's Bureau and
(Bokemeier & Lacy, 1987; Miller, 1980) were related to job satisfaction. collected from 2002 to 2005. Participants of the survey were sampled
Previous research has established the association of worklife bal- non-randomly and from 25 public child welfare agencies across the
ance and job satisfaction, within or outside of social work. Strain and state (for details of the survey, see Strolin-Goltzman et al., 2007). Only
conict in worklife relations were found to have negative effects on baseline data of 782 respondents were involved in this study. To avoid
job satisfaction for social and human services workers (Lambert et al., the negative impacts of missing values on the analyses, 206 respon-
2006) and working mothers (Losoncz & Bortolotto, 2009). In addition, dents were excluded from the study using listwise deletion technique
measures facilitating worklife balance such as exible working due to their incomplete answers for any of the analyzed variables
arrangements and telecommuting were reported to relate with changes (Allison, 1999, p. 79, 2002, p. 6). In addition, frequencies and histograms
in job satisfaction (Kelliher & Anderson, 2010; Virick et al., 2010). of all variables were checked to assess the minimum and maximum
As discussed earlier, job-related and socio-demographic factors values, outliers, and distribution. This led to the exclusion of additional
associated with worklife balance in prior research also impacted job three respondents with extreme answers of more than 70 hours or less
satisfaction. However, the mechanism that can explain the relationships than one hour weekly working time. These 209 respondents excluded
among these factors, worklife balance, and job satisfaction warrants were primarily female (84.1%), whites (92.8%), and urban staff
further investigation. It is possible that, as an important stressor or (75.6%), with a mean age of 43 years. As a result, 573 valid samples
contributor in shaping job satisfaction (Lambert et al., 2006), work were included in this study and the sample size was large enough to
life balance also mediates the relationship between the previously do multivariate analyses.
mentioned factors and job satisfaction. Child welfare workers may be
experiencing worse worklife balance when they labor under unpleasant 3.2. Measures and variables
working conditions, which leads to lower evaluation of their job.
Improvement of working conditions may positively change workers' 3.2.1. Worklife balance
perception toward their worklife relationships and transform into pos- Worklife balance was measured by a scale consisted of eight items
itive perceptions of job satisfaction. In short, empirical and logic-based (shown in Table A) to assess child welfare workers' perception of how
evidence implies the mediating role of worklife balance in the rela- balanced their job was with their personal lives (Strolin-Goltzman
tionship between job-related factors and job satisfaction. This is to be et al., 2007). The items were adopted based on thorough literature review
explored in the present study. (Carlson et al., 2009; Jang, 2009; Smith, 2005; Strolin-Goltzman et al.,
2007). Items were measured with a ve-point Likert scale from 1 =
2.3. Conceptual model, research questions, and hypotheses strongly disagree to 5 = strongly agree. For each respondent, the
total score of the items was divided by the number of items. The mean
The conceptual model tested in the study is illustrated in Fig. 1. score of the items indicated respondents' perception of worklife balance
The present study intends to address two research questions and with higher scores representing better worklife balance and vice versa.
relevant hypotheses: What job-related factors are closely associated The Cronbach's alpha of the scale in this study was 0.76.
with worklife balance among child welfare workers? It is hypothesized
that supervisor support, organizational support, job value, and tenure 3.2.2. Job satisfaction
are positively associated with child welfare workers' perception of A single item asking respondents All in all I am satised with my
worklife balance, while work time, income, and supervisory status are job was used to assess their overall job satisfaction. Using a global
negatively associated with worklife balance. Does worklife balance measure to investigate the overall job satisfaction rather than assessing
partially mediate the inuence of job-related factors on job satisfaction? dimensions of job satisfaction was acceptable in relevant research
It is hypothesized that worklife balance partially mediates the relation- (Bradley & Cartwright, 2002; Lambert et al., 2006). In the original
ships between several job-related factors and job satisfaction. dataset, this was the sole item to assess overall job satisfaction. The
item was measured with a ve-point Likert scale from 1 = strongly
3. Method disagree to 5 = strongly agree.

3.1. Data and sample 3.2.3. Supervisor support and organizational support
Supervisor support was measured by a thirteen-item scale and
This study used a secondary dataset of public child welfare case- organizational support was measured by an eleven-item scale (Table A).
workers and front line supervisors in a northeastern state of the United Both scales were in a ve-point Likert scale format ranging from 1 for
States of America. The original data were part of the Workforce strongly disagree to 5 for strongly agree. The mean score of each
scale represented the variable with higher scores indicating more
perceived support. The scales were validated in previous research on
Job-related factors: job retention (Smith, 2005). In this study, the Cronbach's alphas for
Work time each instrument were 0.95 and 0.71 respectively.
Income Work-life
Supervisory status balance 3.2.4. Job value
Tenure Job value was measured by a nine-item scale (Table A), which was
Supervisor support found to be reliable and valid in a previous study (Smith, 2005). It was a
Organizational support ve-point Likert scale with 1 for strongly disagree and 5 for strongly
Job value Job agree and the mean score indicated the variable. In this study it had a
Socio-Demographic satisfaction Cronbach's alpha of 0.75.
Gender 3.2.5. Other job-related factors
Race Work time was measured by asking respondents On average how
Locality many hours per week do you work in your county child welfare job?
Age The self-reported answers were recorded as a continuous variable in
hours. Income was measured by asking respondents for their annual
Fig. 1. Conceptual model. salary. It was an ordinal scale (1 = less than $25,000, 2 = $25,001
1450 L. Wu et al. / Children and Youth Services Review 35 (2013) 14471454

30,000, 3 = $30,00135,000, 4 = $35,00145,000, and 5 = more signicant relationships with worklife balance. When the appreciation
than $45,000). Supervisory status was coded 0 if the respondent for the child welfare work value increased so did their perception of
was a caseworker and 1 if he/she was a supervisor. Tenure was worklife balance. Organizational support also assisted in explaining
operationalized as years of time in a position which was calculated worklife balance because the more organizational support perceived,
by subtracting the start date from the survey date (Smith, 2005). It the higher worklife balance reported. As work time and income
was a continuous variable in years. decreased, the workers reported more balance between their work
and personal lives. However, supervisor support and other factors had
3.2.6. Socio-demographic variables no signicant impacts on worklife balance in the regression model.
The socio-demographic information consisted of age (self-reported In terms of magnitude of effects as reported in the column (the stan-
in years), gender (0 = male, 1 = female), race (0 = nonwhite, 1 = dardized regression coefcient) of Table 2, job value had the largest
white), and locality (0 = rural county, 1 = urban county). The rural impact which was nearly 1.5 times the magnitude of impact than orga-
urban division followed the 2003 RuralUrban Continuum Codes (U.S. nizational support, followed by income and work time. Furthermore,
Department of Agriculture, 2003). the overall model explained 48% (F = 47.84, p b 0.001) of variance in
worklife balance. In addition, the values of Variance Ination Factor
4. Results (VIF) for all the entered variables were below 2.00, indicating that
multicollinearity was not a problem in the study (Allison, 1999, pp.
4.1. Descriptive statistics 140142). The normal PP plot of regression standardized residuals
indicated that the residuals were normally distributed and the scatter
Statistical Package for the Social Sciences (SPSS) 16.0 for Windows plot of the standardized residuals on the standardized predicted values
was used to analyze the data. The sample consisted of 573 participants reected that heteroskedasticity was not a problem in the regression
in total. Table 1 shows that the sample was made up of primarily analysis (Allison, 1999, pp. 125128).
women, whites, and urban staff, with a mean age of 40 years. The
participants reported that they worked 39 hours per week in their 4.3. Mediating effect of worklife balance on job satisfaction
child welfare job and one-fth of them were supervisors. Nearly
one-third (32.3%) of the employees had an annual income less than Four steps performed with three regression equations (Baron &
$30,000. Forty-ve percent of the participants indicated that they Kenny, 1986; Frazier, Tix, & Barron, 2004) were used to establish the
were very or somewhat satised with their job. Minimum and maximum mediating role of worklife balance on job satisfaction. Step 1 involved
values of the variables are reported in Table 1. Descriptive data for the determining the signicant relationship between job-related variables
measures of perceived supervisor support, organizational support, job and job satisfaction without the presence of worklife balance. Step 2
value, and worklife balance are also presented in Table 1. The mean of was to show that these job-related factors were related to the mediating
worklife balance was 3.32, with a standard deviation of 0.62. variable of worklife balance. Step 3 required a signicant association
between worklife balance and job satisfaction, controlling for other
4.2. Predicting worklife balance variables. Finally, in step 4, the strength of the relation between these
job-related variables and job satisfaction was remarkably reduced after
Multiple linear regression analysis was performed to examine the controlling for the mediating variable of worklife balance. Table 3
relationships between worklife balance and the preceding variables shows the results of the analyses.
and the results are summarized in Table 2. The results show that work Results show that worklife balance partially mediated the rela-
time, income, organizational support, and job value had statistically tionship between two job-related variables of organizational support

Table 1
A brief summary of descriptive statistics of the variables (N = 573).

Variables Frequency (%) Mean (SD)/median Minimum Maximum Cronbach's

Male 115 (20.07)
Female 458 (79.93)
Nonwhite 36 (6.28)
White 537 (93.72)
Rural 185 (32.29)
Urban 388 (67.71)
Age 40.07 (10.04) 22 62
Work time 39.07 (5.72) 7 70
Income 30,00135,000
b25,000 41 (7.16)
25,00130,000 144 (25.13)
30,00135,000 112 (19.55)
35,00145,000 169 (29.49)
>45,000 107 (18.67)
Supervisory status
Worker 458 (79.93)
Supervisor 115 (20.07)
Tenure 4.07 (4.26) 0.1 24.4
Supervisor support 3.68 (0.78) 1.08 5.00 0.95
Organizational support 3.07 (0.53) 1.27 4.64 0.71
Job value 3.35 (0.54) 1.67 4.67 0.75
Worklife balance 3.32 (0.62) 1.12 5.00 0.76
Job satisfaction 3.17 (1.05) 1.00 5.00

Notes: SD: standard deviation.

L. Wu et al. / Children and Youth Services Review 35 (2013) 14471454 1451

Table 2 reected that heteroskedasticity was not a problem in any of the regres-
Regression of worklife balance on the job-related variables (N = 573). sion analyses (Allison, 1999, pp. 125128).
Independent variables B (SE) The Sobel's (1982) test indicated that the indirect effect of organiza-
tional support over worklife balance to job satisfaction was signicant
Gender 0.01 (0.05) 0.01
Race 0.11 (0.08) 0.04 (z = 5.36, p b 0.001). Moreover, the indirect effect of job value over
Locality 0.00 (0.05) 0.00 worklife balance to job satisfaction was signicant (z = 6.45, p b
Age 0.00 (0.00) 0.01 0.001). According to the calculation approach proposed by Sobel
Work time 0.01 (0.00) 0.08
(1982), the amount of mediating effect was 0.18 for organizational
Income 0.07 (0.02) 0.13
Supervisory status 0.04 (0.05) 0.02 support and 0.26 for job value in this research. In short, these ndings
Tenure 0.01 (0.01) 0.04 suggested that worklife balance partially mediated the associations
Supervisor support 0.02 (0.03) 0.02 of organizational support and job value with job satisfaction.
Organizational support 0.35 (0.05) 0.30
Job value 0.51 (0.05) 0.44
R2 0.48 5. Discussion and conclusion
Adjusted R2 0.47
F (signicance) 47.84 (0.000)
df 11/561
5.1. Summary

Notes: df: degrees of freedom.

This study examined job-related correlates and the mediating role
Signicance levels.
p b 0.01. of worklife balance among 573 public child welfare workers. The
p b 0.001. regression results demonstrated that organizational support, job value,
work time, and income were signicantly associated with worklife
balance, thus partly supporting the rst hypothesis that job-related fac-
and job value and job satisfaction. Organizational support, shown in
tors were associated with child welfare workers' perception of worklife
Table 3, was signicantly related to job-satisfaction in step 1 ( =
balance. Worklife balance was conrmed to partially mediate the asso-
0.33, p b 0.001), controlling for other variables. In step 2, it was asso-
ciations of organizational support and job value with job satisfaction. So
ciated signicantly with worklife balance ( = 0.30, p b 0.001). In
far, the second hypothesis was partly supported.
step 3 after controlling for worklife balance and other variables, it
was remarkably related to job satisfaction ( = 0.24, p b 0.001). In
step 4, there was a decrease in value from 0.33 in step 1 to 0.24 5.2. Contribution
for the association between organizational support and job satisfac-
tion with the inclusion in the regression model of worklife balance. This study contributes to worklife balance research by identifying
In the same vein, job value was signicantly associated with job organizational support, job value, work time, and income as job-
satisfaction in step 1 ( = 0.43, p b 0.001), worklife balance in step 2 related correlates of worklife balance. In an integrative framework,
( = 0.44, p b 0.001), and job satisfaction in step 3 ( = 0.30, p b the combined and relative impacts of these correlates were examined.
0.001). There was also a decrease in value from 0.43 in step 1 to 0.30 It also contributes by proposing the mediating role of worklife balance
for its association with job satisfaction with the inclusion in the regres- in the relationships between two job-related factors of organizational
sion model of worklife balance. The values of VIF for the independent support and job value and job satisfaction. The inuencing mechanisms
variables examined in all regression analyses were below 2.28, indicat- of these factors on job satisfaction were claried, therefore enriching
ing that there was no problem of multicollinearity (Allison, 1999, the literature on worklife balance and job satisfaction. In addition, it
pp. 140142). The normal PP plot of regression standardized residuals has merits in specifying characteristics and correlates of worklife
indicated that the residuals were normally distributed and the scatter balance in a child welfare setting thus expanding knowledge of child
plot of the standardized residuals on the standardized predicted values welfare workforce.

Table 3
Mediational analyses (N = 573).

Independent variables Step 1 Step 2 Steps 3 & 4

B (SE) B (SE) B (SE)

Gender 0.03 (0.08) 0.01 0.01 (0.05) 0.01 0.03 (0.07) 0.01
Race 0.17 (0.13) 0.04 0.11 (0.08) 0.04 0.11 (0.12) 0.03
Locality 0.05 (0.09) 0.02 0.00 (0.05) 0.00 0.05 (0.08) 0.02
Age 0.00 (0.00) 0.00 0.00 (0.00) 0.01 0.00 (0.00) 0.00
Work time 0.01 (0.01) 0.05 0.01 (0.00) 0.08 0.00 (0.01) 0.02
Income 0.06 (0.04) 0.07 0.07 (0.02) 0.13 0.10 (0.03) 0.11
Supervisory status 0.05 (0.08) 0.02 0.04 (0.05) 0.02 0.04 (0.08) 0.01
Tenure 0.01 (0.01) 0.05 0.01 (0.01) 0.04 0.02 (0.01) 0.07
Supervisor support 0.09 (0.04) 0.07 0.02 (0.03) 0.02 0.08 (0.04) 0.06
Organizational support 0.65 (0.08) 0.33 0.35 (0.05) 0.30 0.47 (0.08) 0.24
Job value 0.85 (0.08) 0.43 0.51 (0.05) 0.44 0.59 (0.08) 0.30
Worklife balance 0.50 (0.06) 0.30
R2 0.54 0.48 0.58
Adjusted R2 0.53 0.47 0.57
F (signicance) 59.03 (0.000) 47.84 (0.000) 64.98 (0.000)
df 11/561 11/561 12/560

Notes: in step 2, the dependent variable is worklife balance. In steps 1, 3, & 4, the dependent variable is job satisfaction. df: degrees of freedom.
Signicance levels.
p b 0.05.
p b 0.01.
p b 0.001.
1452 L. Wu et al. / Children and Youth Services Review 35 (2013) 14471454

In this study, appreciation for child welfare work value was a purpose of workforce retention administration and research (Smith,
signicant predictor and had the greatest magnitude of impact on 2005; Strolin-Goltzman et al., 2007). They were not intended for a sec-
perception of worklife balance. It also exerted indirect inuence on ondary study and may need modication for research focusing
job satisfaction through worklife balance. This is the rst time re- on worklife balance and its correlates. Even though the scales
search conrms the impacts of job value on and through employee were found to be reliable and valid in previous and present
worklife balance. It was possible that the mission to help children studies, they have not been standardized. Third, job satisfaction
and families in need facilitated the balance between child welfare was measured with one item in the secondary dataset rather
workers' job and personal lives, keeping them positively perceive than a scale of all possible dimensions (see Spector, 1985). Even
their job even in unpleasant working conditions. As a relatively con- though measuring the concept globally was accepted (Bradley
stant variable, job value formed a powerful motivator of employees & Cartwright, 2002; Bokemeier & Lacy, 1987; Lambert et al.,
toward appreciating their work and personal lives, possibly offsetting 2006; Rauktis & Koeske, 1994; Sharma et al., 1997), assessing it
the stressful and demanding working conditions child welfare workers with dimensions would provide much information. More rigorous
encountered in their work environments. As a result, strengthening job measurement of job satisfaction is needed in future research on
value may be helpful for organizations if working conditions are hard to worklife balance.
improve. Fourth, this study employed a cross-sectional design, which
The association between organizational support and worklife prohibited establishing a causal model. Although logic and plausibil-
balance in the framework conrmed the importance of organizational ity can help in making a conclusion in the present study, future
support in successful integration of employees' work and life (Hopkins, studies with a longitudinal design may be able to establish causal
2005; Jang, 2009). As no prior research investigated the impact of orga- relationships. Fifth, the survey applied a nonprobability sampling
nizational support through worklife balance on workforce satisfaction, and was conned to 25 counties of one state. Study ndings cannot
this nding is a pioneering one. Obviously, more organizational support be generalized to an overall picture. Sixth, even if the four conditions
might mitigate work pressure and reduce conicts between work and of a meditational model were met in the study, there are likely to be
life, thus enhancing satisfaction in professional life. In the vein of the alternative models that t the data (Frazier et al., 2004). Because the
mediating effect of worklife balance, organizational support was linked predictors, mediator, and outcome were measured at the same point
with job satisfaction by dual ways a direct way and an indirect way. in time, the direction of the relationships among these variables was
This means that, for administrations concerned worker performance unclear. For instance, a reverse causal order could exist. Further,
and retention in the job, worklife issues should not be neglected. Still, omitted variables may impact both the mediator and the outcome
the quality and quantity of support in work environments are essential variable.
for workers to weather excessive personal barriers and demands, as
well as unpleasant working conditions.
Unexpectedly, supervisor support did not predict worklife bal- 5.4. Implications
ance as organizational support did. When organizational support
positively impacted the workers' worklife balance through mea- As no prior research has probed into correlates and medicating
sures and climate in the overall work environment, it was plausible role of worklife balance, this study is, by nature, exploratory. Its
that supervisors had no control over issues related to worklife tentative ndings offer insights to future research on worklife balance
balance with the current regulatory driven practice focus. Thus, a su- in child welfare. More empirical research is needed to verify the
pervisor might be as helpless as his/her workers to change working ndings of the present study. For instance, the lack of supervisor
conditions regarding work and life. Moreover, because the em- support impacting worklife balance warrants justication with future
ployees were somewhat satised with the supervisor support they studies. The exploration of mediating effects of worklife balance
received as shown in the descriptive statistics, it was evident that among job-related factors needs to be replicated with other samples
the supervisor support received was not impacting their worklife and research designs. Further efforts also need to rule out alternative
conditions. models and might use other methods such as Structural Equation
While child welfare workers generally enjoyed their work and Modeling (SEM) to test the meditating role of worklife balance.
personal life relation as shown in the descriptive results, the ndings re- In addition, more variables need to be included to assess their
mind us of the detrimental inuence of overtime work on the workers' inuence.
worklife balance which was consistent with Losoncz and Bortolotto's The ndings of the study suggest several practice and policy impli-
(2009) research. The negative impact of income on worklife balance cations. Given the critical role of job value in work environment, the
may be explained using the relationships between income and other way to develop and strengthen job value in social work education
conditions like position and job responsibilities. Higher income usually and organization practices merits strong consideration. The study
is related with higher position and greater responsibilities which could also suggests the adoption and evaluation of measures to increase
impact the worklife relation. This nding complements the lack of the quality and quantity of organizational support in child welfare.
evidence regarding the relationship between income and worklife The lack of supervisor support implies that a worklife orientation
balance. Besides work time and income, other job-related and socio- should be stressed in the supervisors' providing support to their subor-
demographic factors (tenure, supervisory status, locality, age, race, and dinates in child welfare practice. Moreover, all kinds of ways should be
gender) did not predict worklife balance which warrants examination taken to control overwork and its detrimental effects on employees'
in future research. work and personal lives. In short, structural and cultural transitions of
organizations such as implementing exible job designs and favorable
5.3. Limitations human resource policies and advocating for supportive climate
(Hobson et al., 2001; Kelliher & Anderson, 2010; Kossek et al., 2010;
Several limitations should be considered when interpreting the Lewis, 2003; Lewis & Campbell, 2008; Peters & Heusinkveld, 2010;
ndings. First, this study used a secondary dataset which lacked Virick et al., 2010) are possible ways for child welfare organizations to
many important variables such as marital status, physical and mental facilitate their workers' worklife balance. In closing, this study asks
health, and work overload. These variables may have important impacts administrators and policy-makers to address worklife balance of
on worklife balance but, since they were missing in the dataset, they child welfare workers in an era of challenging working conditions and
cannot be tested in the model. Second, the scales used in the study excessive work demands, as well as imperative family responsibilities
were developed by child welfare administrators and researchers for the and personal requirements.
L. Wu et al. / Children and Youth Services Review 35 (2013) 14471454 1453

Appendix A Connell, R. W. (2005). A really good husband: Work/life balance, gender equity and
social change. Australian Journal of Social Issues, 40, 369383.
Dean, H. (2007). Tipping the balance: The problematic nature of worklife balance in a
low-income neighborhood. Journal of Social Policy, 36, 519537.
Fenner, G. H., & Renn, R. W. (2010). Technology-assisted supplemental work and work-
to-family conict: The role of instrumentality beliefs, organizational expectations
Table A
and time management. Human Relations, 63, 6382.
Construction of the scales.
Frazier, P. A., Tix, A. P., & Barron, K. E. (2004). Testing moderator and mediator effects in
Scales Items counseling psychology. Journal of Counseling Psychology, 51, 115134.
Greenhaus, J. H., Collins, K. M., & Shaw, J. D. (2003). The relation between workfamily
Worklife There is a good t between my personal life and work life. balance and quality of life. Journal of Vocational Behavior, 63, 510531.
balance There is a good t between my family life and work life. Hegney, D., Plank, A., & Parker, V. (2006). Extrinsic and intrinsic work values: Their
I receive support and recognition from family members. impact on job satisfaction in nursing. Journal of Nursing Management, 14, 271281.
There is a good t between my job and my personal health. Hobson, C. J., Delunas, L., & Kesic, D. (2001). Compelling evidence of the need for
I am able to do my job and not burnout. corporate work/life balance initiatives: Results from a national survey of stressful
This job enables me to continue living where I live now. life-events. Journal of Employment Counseling, 38, 3844.
Hopkins, K. (2005). Supervisor support and worklife integration: A social identity
I have sufcient emotional energy for the job.
perspective. In Ellen Ernst Kossek, & Susan J. Lambert (Eds.), Work and life integration:
My work offers schedule exibility.
Organizational, cultural, and individual perspectives. Lawrence Erlbaum Associates,
Supervisor My supervisor genuinely cares about me. Publishers.
support My supervisor gives me help when I need it. Jang, S. J. (2009). The relationships of exible work schedules, workplace support,
My supervisor shows approval when I succeed. supervisory support, worklife balance, and the well-being of working parents.
My supervisor provides the help I need to complete my required tasks. Journal of Social Service Research, 35, 93104.
My supervisor provides the help I need to complete required Judge, T. A., & Bretz, R. D. (1992). Effects of work values on job choice decisions. Journal
paperwork. of Applied Psychology, 77, 261271.
My supervisor supports me in balancing the demands of my job Kelliher, C., & Anderson, D. (2010). Doing more with less? Flexible working practices
with my personal life. and the intensication of work. Human Relations, 63, 83106.
My supervisor supports me in difcult case situations. Kossek, E. E., Lewis, S., & Hammer, L. B. (2010). Worklife initiatives and organizational
My supervisor helps me learn and improve. change: Overcoming mixed messages to move from the margin to the mainstream.
My supervisor values and seriously considers my opinion in case Human Relations, 63, 319.
Kwak, C., Chuang, B. Y., Xu, Y., & Eun-Jung, C. (2010). Relationship of job satisfaction
decision making.
with perceived organizational support and quality of care among South Korean
My supervisor can accept an alternative perspective.
nurses: A questionnaire survey. International Journal of Nursing Study, 47, 12921298.
My supervisor is supportive of any on-the-job-training I attend.
Lambert, E. G., Pasupuleti, S., Cluse-Tolar, T., Jennings, M., & Baker, D. (2006). The
My supervisor is available to me when I ask for help. impact of workfamily conict on social work and human service worker job satisfac-
My supervisor helps me prevent and address burn-out. tion and organizational commitment: An exploratory study. Administration in Social
Organizational My agency is committed to my personal safety in the ofce. Work, 30(3), 5574.
support My agency is committed to my personal safety in the eld. Lee, H., & Cummings, G. G. (2008). Factors inuencing job satisfaction of front line
The support staff in the agency is adequate. nurse managers: A systematic review. Journal of Nursing Management, 16, 768783.
Physical work space is satisfactory. Lewis, S. (2003). The integration of paid work and the rest of life: Is post-industrial
My professional opinions are respected in this agency. work the new leisure? Leisure Studies, 22, 343355.
When my co-workers are successful, I feel successful. Lewis, J., & Campbell, M. (2008). What's in a name? Work and family or work and life
There is a can do attitude among co-workers. balance policies in the UK since 1997 and the implications for the pursuit of gender
I receive support and recognition from co-workers. equality. Social Policy & Administration, 42, 524541.
Losoncz, I., & Bortolotto, N. (2009). Worklife balance: The experiences of Australian
The work I do is recognized by other professionals in the community.
working mothers. Journal of Family Studies, 15, 122138.
The agency is respected in the community.
McGowan, B. G., Auerbach, C., & Strolin-Goltzman, J. S. (2009). Turnover in the
I would recommend my agency to others seeking employment in
child welfare workforce: A different perspective. Journal of Social Service Research, 35,
child welfare. 228235.
Job value In my work, I have a feeling of success and accomplishment. Meglino, B. M., Ravlin, E. C., & Adkins, C. L. (1989). A work values approach to corporate
Clients regularly succeed in reaching goals. culture: A eld test of the value congruence process and its relationship to individual
My work offers opportunities to make a difference. outcomes. Journal of Applied Psychology, 74, 424432.
My work offers opportunities for improving knowledge and skills. Mescher, S., Benschop, Y., & Doorewaard, H. (2010). Representations of worklife
My work offers opportunities to ensure the safety and well-being balance support. Human Relations, 63, 2139.
of clients. Miller, J. (1980). Individual and occupational determinants of job satisfaction: A focus
I am prepared for my job because of my prior training and education. on gender differences. Work and Occupations, 7(3), 337366.
The job has the right level of challenge. Moen, P., & Yu, Y. (2000). Effective work/life strategies: Working couples, work condi-
The agency helps me to implement best practices. tions, gender, and life quality. Social Problems, 47, 291326.
This job ts with my career goals. Ollier-Malaterre, A. (2010). Contributions of worklife and resilience initiatives to the
individual/organization relationship. Human Relations, 63, 4162.
Peters, P., & Heusinkveld, S. (2010). Institutional explanations for managers' attitudes
towards telehomeworking. Human Relations, 63, 107135.
Rauktis, M. E., & Koeske, G. F. (1994). Maintaining social worker morale: When supportive
References supervision is not enough. Administration in Social Work, 18, 3960.
Sharma, J., McKelvey, J., Hardy, R., Epstein, M. H., Lomax, R. G., & Hruby, P. J. (1997). Job
Allison, P. D. (1999). Multiple regression: A primer. Thousand Oaks, CA: Pine Forge Press. satisfaction of child welfare workers in an urban setting: Status and predictors.
Allison, P. D. (2002). Missing data. Thousand Oaks, CA: Sage. Journal of Child and Family Studies, 6, 209219.
Auerbach, C., McGowan, B. G., Ausberger, A., Strolin-Goltzman, J., & Schudrich, W. Smith, B. (2005). Job retention in child welfare: Effects of perceived organizational
(2010). Differential factors inuencing public and voluntary child welfare workers' support, supervisor support, and intrinsic job value. Children and Youth Services
intention to leave. Children and Youth Services Review, 32, 13961402. Review, 27, 153169.
Baron, R. M., & Kenny, D. A. (1986). The moderatormediator variable distinction in Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations
social psychological research: Conceptual, strategic, and statistical considerations. models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp. 290312). San Francisco,
Journal of Personality and Social Psychology, 51(6), 11731182. CA: Jossey-Bass.
Barth, R. P., Lloyd, C. E., Christ, S. L., Chapman, M. V., & Dickinson, N. S. (2008). Child Spector, P. E. (1985). Measurement and human service staff satisfaction: Development of
welfare worker characteristics and job satisfaction: A national study. Social Work, the job satisfaction survey. American Journal of Community Psychology, 13, 693713.
53, 199209. Stamper, C. L., & Johlke, M. C. (2003). The impact of perceived organizational support
Blood, M. R. (1969). Work values and job satisfaction. Journal of Applied Psychology, 53, on the relationship between boundary spanner role stress and work outcomes.
456459. Journal of Management, 29, 569588.
Bokemeier, J. L., & Lacy, W. B. (1987). Job values, rewards, and work conditions as Strolin-Goltzman, J., Auerbach, C., McGowan, B. G., & McCarthy, M. (2007). The rela-
factors in job satisfaction among men and women. The Sociological Quarterly, 28, tionship between organizational characteristics and workforce turnover among
189204. rural, urban, and suburban public child welfare systems. Administration in Social
Bradley, J. R., & Cartwright, S. (2002). Social support, job stress, health, and job satisfaction Work, 32, 7791.
among nurses in the United Kingdom. International Journal of Stress Management, 9, U.S. Department of Agriculture (2003). Ruralurban continuum codes. Retrieved from.
163182. http://www.ers.usda.gov/data-products/rural-urban-continuum-codes.aspx
Carlson, D. S., Grzywacz, J. G., & Zivnuska, S. (2009). Is workfamily balance more than Vinokur-Kaplan, D. (1991). Job satisfaction among social workers in public and volun-
conict and enrichment? Human Relations, 62, 14591486. tary child welfare agencies. Child Welfare, 70, 8191.
1454 L. Wu et al. / Children and Youth Services Review 35 (2013) 14471454

Virick, M., DaSiva, N., & Arrington, K. (2010). Moderators of the curvilinear relation Westbrook, T. M., Ellis, J., & Ellett, A. J. (2006). Improving retention among public child
between extent of telecommuting and job and life satisfaction: The role of performance welfare workers: What can we learn from the insights and experiences of committed
outcome orientation and worker type. Human Relations, 63, 137154. survivors? Administration in Social Work, 30(4), 3762.