Work Barriers in the Context of Pathways to the Employment of
Shawna J. Lee Æ Æ Amiram D. Vinokur
Published online: 2 October 2007
? Springer Science+Business Media, LLC 2007
the welfare rolls and stay in the labor force is often limited
by the work barriers they face. Using a sample of 1,404
female welfare-to-work clients we first examined the
structure of work barriers and then tested their contribution
to current work status in the context of a structural equation
employment. Whereas work barriers included diverse
factors ranging from lack of transportation to low quality
jobs, they were shown to constitute a uni-dimensional
construct. Furthermore, work barriers had a net adverse
effect on employment outcomes, controlling for job search
self-efficacy and employment intention. We conclude
with discussion of implications for the development of
The ability of welfare-to-work clients to leave
welfare-to-work programs and interventions that target
Work barriers ? Path model
Welfare ? Work ? Low-income women ?
The 1996 Personal Responsibility and Work Opportunity
Reconciliation Act (PRWORA) welfare reform legislation
replaced the federal entitlement program with a block grant
to individual states, and established a 5-year time limit on
receipt of welfare benefits. Prior to 1996, welfare had
provided a social safety net to recipients who were mostly
single mothers with children. Current federal law requires
30 h of work per week, or 20 h for women with children
under age 6, and 10 h of education, other training, volun-
teer or community service programs may count toward the
work requirement (Haskins and Offner 2003). A dramatic
decline in welfare caseloads, from approximately 4 million
to 2 million families in the first 5 years following welfare
reform, corresponded with a large increase in workforce
participation among current and former welfare recipients
(Blank 2002; Lichter and Jayakody 2002).
Yet, despite the increase in employment among low-
income women following welfare reform, many current
and former welfare recipients face difficulty in transition-
ing successfully to the workforce and even those who
obtain jobs often return to welfare. Job instability is com-
mon (Johnson and Corcoran 2003; Lee 2004), and of those
who left welfare between 1997 and 1999, 22% returned to
the welfare rolls by 1999 (Loprest 2002a). Barriers that
hinder the transition from welfare to work are one reason
that some women have difficulty obtaining and maintaining
employment, and return to the welfare rolls (Danziger et al.
2000; Taylor 2001; Taylor and Barusch 2004). For
This paper is based on research supported by the National Institute of
Mental Health grant number P30-MH38330 to the Michigan
Prevention Research Center. The first author was supported in part by
National Institute of Mental Health Michigan grant number
T32-MH63057. The first author presented portions of this paper at the
2005 annual meeting of the International Society for Quality-of-Life
Studies (ISQOLS), Philadelphia, PA, and at the 2004 annual meeting
of the Society for Social Work and Research (SSWR), New Orleans,
S. J. Lee
Wayne State University, Detroit, MI, USA
S. J. Lee (&)
School of Social Work, Wayne State University
, 4756 Cass Avenue, Detroit, MI 48201, USA
A. D. Vinokur
Institute for Social Research, University of Michigan,
426 Thompson Street, Ann Arbor, MI 48106, USA
Am J Community Psychol (2007) 40:301–312
example, the low quality of jobs in terms of the pay and
benefits are often a barrier to employment (Johnson and
Corcoran 2003; Lee 2004). Low wage jobs that do not offer
health care benefits may contribute to job instability and
increase the likelihood that women will return to welfare
(Johnson and Corcoran 2003; Lee 2004; Nam 2005). Even
with years of work experience, research suggests that many
former welfare recipients do not attain jobs with wages
above the official U.S. poverty level (Danziger and Johnson
2005; Loprest and Zedlewski 2006).
Beyond the quality of jobs in terms of pay and benefits,
inadequate human capital characteristics comprise another
set of barriers to work. In one panel study using a repre-
sentative sample of current and former welfare recipients,
31.4% reported not having a high school diploma, 15.4%
had limited work experience, and 21.2% had few job skills
(Danziger et al. 2000). Mental and physical health prob-
lems pose additional challenges. At some time during the
first 4.5 years the study was conducted over 70% of the
respondents reported limited physical functioning and 60%
met the criteria for a mental health disabling condition such
as generalized anxiety disorder or severe depression
(Corcoran et al. 2003). Additionally, 31% reported lack of
adequate and affordable childcare (Danziger et al. 2004)
and 47.1% said that lack of transportation was a barrier to
work (Danziger et al. 2000).
Work barriers are strongly associated with employment.
Women with more barriers work less over time and have
more difficulty sustaining employment (Corcoran et al.
2003; Danziger et al. 2000, Loprest and Zedlewski 2006).
In one study using data from the National Survey of
America’s Families, 14.1% of the women had two or more
barriers to work; and they were much less likely to be
working while receiving welfare compared to those who
reported no barriers to work, approximately half of the
sample (Zedlewski 2003). Low education, limited work
experience, few job skills, and poor mental and physical
health decrease job retention and increase the likelihood
that a former recipient will return to welfare (Corcoran
et al. 2000; Loprest 2002b; Nam 2005). Those with a work
history in which they have been on welfare (or have not
worked) for an extended period of time are more likely to
return to welfare (Loprest 2002b).
In addition to the work barriers mentioned above that
reduce the likelihood and extent of employment among
low-income women, some studies have examined addi-
tional work barriers such as learning disabilities, perceived
discrimination, drug dependence, child health problems,
substance abuse problems, and domestic violence (e.g.,
Danziger et al. 2000; Dooley and Prause 2002; Gutman
et al. 2003; Nam 2005; Taylor 2001; Taylor and Barusch
2004; Pollack et al. 2002; Zedlewski 2003). To date,
measurement of work barriers has not been conceptualized
in a coherent manner and researchers have not asked if
there is a cohesive structure underlying factors related to
work barriers. While one study noted the high internal
consistency of a work barriers measure (Taylor 2001), most
often work barriers are treated as distinct variables or
summed to create a composite score. Thus, there remains a
question of whether these barriers constitute discrete and
unique factors or whether they are indicators of a global
uni-dimensional construct that impedes employment.
To address this gap in the literature regarding the con-
ceptualization of work barriers, the first goal of the current
study is to examine the underlying structure of diverse
work barriers that in past literature have been examined
separately as predictors of employment. We focus on those
barriers that are most common and have demonstrated a
relationship to employment outcomes, including: inade-
quate human capital (education and job experience),
inadequate job quality in terms of pay and health benefits,
childcare, lack of transportation, and depression. Estab-
lishing that a diverse set of work barriers form a uni-
dimensional construct is an important step in demonstrat-
ing the predictive validity of the construct particularly if, as
we expect, work barriers predict future low level of
employment or unemployment.
The second goal of this study is to examine the extent to
which the work barrier construct predicts future employ-
ment engagement among welfare-to-work clients. Support
for negative consequences of the work barrier construct on
employment outcomes would substantiate the theoretical
validity of the barrier construct particularly if it is obtained
within a broader network of constructs that are tied to
We also seek to expand knowledge related to work
barriers by measuring barriers in a manner that diverges
from previous research. In past research work barrier
variables have been constructed based on respondent’s
demographic and other characteristics; for example, the
absence of a high school diploma as an indicator of an
education work barrier. Instead, this study focuses on the
perception of work barriers by asking participants to
indicate to what extent factors such as not having enough
education and work experience prevented them from
looking for or getting a job. In this way we extend previous
research by examining if the belief that one has work
barriers influences the intention to participate in the job
search and gain employment in the same way that job
search self-efficacy is found to influence intention, though
in the opposite direction (e.g., van Ryn and Vinokur 1992).
The objective aspects of work barriers may cause them to
be highly interrelated. For example, low education may
hinder the ability to obtain an adequately paying job, which
in turn creates financial problems linked to childcare and
transportation. At the same time, the perception that these
302Am J Community Psychol (2007) 40:301–312
factors act as barriers to work creates a psychological
motivational force akin to, but in the opposite direction of,
job search self-efficacy or mastery by inhibiting either the
intention to gain employment or effective behavior that
could lead to positive work outcomes.
Structural Model of Pathways to Employment
In order to assess the extent to which the work barrier
construct predicts employment we constructed a model to
predict employment experience. This model is based on
earlier studies with unemployed job seekers (Vinokur and
Schul 1997, 2002; Vinokur et al. 2000) that have demon-
strated direct and mediating relationships among mastery,
financial strain, job search self-efficacy, and employment
intention to employment. In studies using both community
samples and intervention participants, higher levels of
mastery were positively related to job search self-efficacy,
reduced financial strain, and lowered depressive symptoms;
these mediators in turn increased job search intention and
are related to more positive employment outcomes over
time (Vinokur and Schul 1997, 2002; Vuori and Vinokur
2005). In the current study, it is additionally hypothesized
that the work barrier construct hinders employment. Spe-
cific hypotheses are examined in more detail below.
Following from the model of human agency proposed
by Bandura (1989), motivation and intention to act is
enhanced when individuals have a belief in their ability to
exercise control over their environment. High levels of
mastery can promote positive, proactive coping strategies
in the face of difficult life circumstances, thereby reducing
vulnerability to depression and other negative psychologi-
cal outcomes (Elder and Russell 2000; Conger et al. 2000;
Conger et al. 1999; Vinokur et al. 2000). When those with
lower levels of mastery encounter difficult life circum-
stances their sense of mastery is further eroded, increasing
emotional distress and symptoms of depression (Conger
et al. 1999).
Mastery is believed to be central to, but distinct from,
self-efficacy. Self-efficacy is a feeling of competency in a
particular domain that is derived from a sense of control
over one’s environment in that domain, and is believed to
emanate in part from a broader sense of mastery. There-
fore, to isolate the effect of self-efficacy that is specific to
the job domain we conceptualized mastery as a more
general variable influencing self-efficacy as well as finan-
cial strain, employment intention, and work barriers. It is
only when known predictors of employment are controlled
for that the net effects of a work barrier construct can be
demonstrated with appropriate rigor.
Consistent with the notion of mastery as a general var-
iable, and as shown in previous research (Vinokur and
Schul 1997, 2002), the effect of mastery on employment
experience is not a direct one but is mediated by more
proximal variables related to employment. In our model we
include financial strain as a mediator. Previous research
with middle-class samples has shown that mastery reduces
financial strain (Vinokur and Schul 2002). Financial strain
may also increase the intention to seek a job due to the
need for income, while work barriers may discourage and
dampen the intention to become employed. Assessing the
impact of financial strain in the context of work barriers is
important because although employment may be beneficial
to low-income mothers (Raver 2003), it does not neces-
sarily reduce financial strain due to the low quality of
employment and other factors (Danziger and Kalil 2002;
Gyamfi et al. 2001; Jackson et al. 2000). Furthermore, in a
low-income sample, financial strain is indicative of short-
age of financial resources that could have otherwise been
deployed to diminish or remove work barriers (Vinokur
and Schul 1997, 2002).
The model includes three direct predictors of employ-
ment experience: work barriers, job search self-efficacy,
and employment intention. Mastery according to the model
is hypothesized to have a direct impact on job search self-
efficacy, employment intention (Vinokur and Schul 1997,
2002) as well as on work barriers, which in turn influence
As noted above, previous research with low-income
women and welfare recipients has also established that
work barriers are impediments to obtaining and maintain-
ing employment (Danziger et al. 2000; Danziger et al.
2004; Johnson and Corcoran 2003; Taylor and Barusch
2004; Zedlewski 2003), whereas positive psychological
mechanisms, including self-efficacy, are related to lowered
reliance on welfare, positive employment outcomes, and
well-being (Danziger et al. 2001; Kalil et al. 2001; Kunz
and Kalil 1999). In several cross-sectional studies of low-
income mothers who were working or on welfare, self-
efficacy was positively correlated with employment
(Jackson 2000; Jackson and Scheines 2005). Following
from this research it is hypothesized that barriers to work
form another mediator in the relationship between mastery
and women’s employment outcomes, and that mastery
decreases perceived work barriers.
Job search self-efficacy is also hypothesized to have a
direct effect on employment and mediate the relationship
between mastery and employment outcomes. Multiple
studies suggest that domain specific self-efficacy increases
the likelihood that one will successfully initiate behavior in
a particular domain (Bandura and Cervone 1983; Bandura
1989; Conger et al. 1999; Elder and Russell 2000). When
individuals feel efficacious and competent they are able to
take on challenges and are more likely to persist even in the
face of failure (Bandura 1989). Job search self-efficacy is
Am J Community Psychol (2007) 40:301–312303
therefore likely to increase intention to seek and find a job
as well as directly improve employment outcomes (van
Ryn and Vinokur, 1992).
According to the theory of planned behavior, intention is
the direct predictor of future behavior (Ajzen 2002) and
behavior is in turn a predictor for relevant outcomes, such
as in our case, employment. In the absence of a measure of
job search behavior we hypothesize that intention is a
direct predictor of employment experience. We also
hypothesize that employment intention is increased by job
search self-efficacy and financial strain, and is decreased by
work barriers. Employment experience is increased by job
search self-efficacy and employment intention, in addition
to being negatively influenced by work barriers as dis-
To investigate the question of whether work barriers
constitute a valid construct, and to test our model based on
the hypotheses stated above, which also incorporated work
barriers, we used data collected from female welfare-to-
work clients in a job training program. Our analyses will
first examine the structure of work barriers and then pro-
ceed to test a structural model predicting employment
Study participants were 1,404 female welfare applicants in
an urban southeastern county who took part in the Winning
New Jobs (WNJ) job preparation program to qualify for
welfare benefits.1Participation in the study was completely
voluntary. The mean age of participants was 29.27 (SD =
7.91) years old. In terms of education, 37% have not
completed high-school education, 39.4% were high-school
graduates, 21% had attended some college, and only 2%
reporting either a college degree or graduate degree. Most
of the participants (62%) were African-Americans; 30%
were Caucasian/whites; and 6.6% were of bi-racial, Asians,
Native Americansor others
The majority of participants indicated their marital sta-
tus as never married (65%). Only 6.1% were married,
15.6% were separated, 10.5% were divorced, and 1% of the
participants were widowed. Participants had an average of
1.96 (SD = 1.15, median = 2) children. Over 96% of the
participants had children; 37% had one child, 32% had two
children, 16.6% had three children, and almost 10% had
four or more children. Furthermore, many had young
children under age 5 at home: 40% had one young child at
home, 17.5% had two young children at home, and 4.7%
had three or more young children at home.
From September 2000 to September 2002 all individuals
applying for welfare benefits and who participated in WNJ
were recruited to the study. They were asked to complete a
self-administered questionnaire at the beginning of the
WNJ workshop. Of the 1,597 women who applied to the
welfare cash assistance program and thus were required to
participate in WNJ in order to receive benefits, 1,404
showed up to the workshop and provided Time 1 (T1)
baseline data (193 women never showed up to participate
in the program). Four months after completing the WNJ
workshop participants were re-contacted (Time 2, or T2)
and asked to participate in a telephone interview. Of the
1,404 who participated at T1, 951 (68%) women provided
T2 follow-up data. The participants were paid $5 for
completing the T1 baseline questionnaire and $10 for
completing the T2 follow-up telephone interview.
Sense of Mastery (Perlin and Schooler 1978) measure
(a = .72, M = 3.01, SD = .52) consisted of seven questions
that asked respondents to rate their sense of personal
control over aspects of their life, with response options
ranging from 1 (strongly disagree) to 4 (strongly agree).
Questions included: ‘‘There is really no way I can solve
some of the problems I have’’; ‘‘Sometimes I feel I am
being pushed around in life’’; ‘‘I have little control over the
things that happen to me’’; ‘‘I can do just about anything I
really set my mind to do’’; ‘‘I often feel helpless in dealing
with the problems of life’’; ‘‘What happens in the future
mostly depends on me,’’ and ‘‘There is little I can do to
change many of the important things in my life.’’ In our
model this construct was indicated with two parcels formed
by the means of the randomized sets of three or four items
in each parcel. Parceling items to form indicators is a
common procedure in structural modeling analysis because
it results in parcel indicators with more appropriate distri-
butional properties in terms of multivariate normality (see,
Bandalos 2002; Kline 2005).
Job Search Self-Efficacy was assessed with a measure
used in earlier investigations on job search and employ-
ment (e.g., Vinokur and Schul 2002). The measure
(a = .93, M = 3.90, SD = 1.02) consisted of six questions
1The original sample included additional 139 male applicants.
Because of their small number and the fact that females face unique
challenges in the transition from welfare-to-work, our analyses
included only the female participants.
304Am J Community Psychol (2007) 40:301–312
asking respondents to rate on a five-point scale from 1 (not
at all confident) to 5 (a great deal confident) their degree of
confidence in being able to successfully perform six job
search activities: making a good list of all the skills that
you have and can be used to find a job; talking to friends
and other contacts to find out about potential employers
who need your skills; talking to friends and other contacts
to discover promising job openings that are suitable for
you; completing a good job application and resume; con-
tacting and persuading potential employers to consider you
for a job; and making the best impression and getting your
points across in a job interview. In our modeling analysis
the indicators for this construct were formed by two parcels
using the means of the randomized sets of three items in
Financial Strain measure (a = .76, M = 3.57, SD =
1.02) was constructed from answers to three questions
used in earlier investigations on job search and employ-
ment (e.g., Vinokur and Caplan 1987). Respondents
indicated how difficult it was for them to live on their total
household income using a scale from 1 (not at all difficult)
to 5 (extremely difficult or impossible). Respondents then
rated the likelihood of experiencing financial hardships and
of having to live with only bare necessities in the next 2
months on a scale from 1 (not at all) to 5 (a great deal). The
three items served as indicators for the financial strain
Employment Intention measure (a = .65, M = 4.25,
SD = 1.02) was also used in earlier studies on reem-
ployment (Vinokur and Caplan 1987; Vinokur and Schul
2002). On a scale from 1 (not at all hard/very unlikely) to
5 (extremely hard/very likely) respondents were asked
two questions: ‘‘In the next 2 months, how hard do you
intend to try to find a job where you’d work 20 h or more
per week?’’ and ‘‘In the next 2 months, how likely is it
that you will try hard to get a job where you’d work 20 h
or more per week?’’ These two items served as indicators
for the employment intention construct in the modeling
Work Barriers were assessed using five indicator mea-
sures: depressive symptoms, inadequate human capital,
inadequate job quality, childcare problems, and transpor-
tation problems. Below is a description of each of these
Hopkins Depression Symptom Checklist (HSCL; a
measure based on Derogatis et al. 1974; a = .92, M = 2.66,
SD = 1.09) was used to assess symptoms of depression. On
a response scale ranging from 1 (not at all) to 5 (extre-
mely), respondents rated the extent to which they were
bothered or distressed by each of eight symptoms in the last
2 weeks, including: poor appetite, feeling low in energy or
slowed down, feeling hopeless about the future, crying
easily, blaming yourself for things, feeling lonely, feeling
no interest in things, and feeling blue. This scale was used
in previous studies of job search and employment (Caplan
et al. 1989; Vinokur and Schul 2002) and has been found
effective in identifying individuals who are at risk of
meeting clinical criteria for depression (Sandanger et al.
1998). The depressive symptoms construct was indicated
by three parcels that were constructed based on the mean
scores of randomly chosen subsets of two or three items
from the HSCL symptom checklist.
The other four work barrier domains in this study were
based on barriers identified in WES (Danziger et al. 2000),
a longitudinal panel study of current and former welfare
recipients in urban Michigan that examined work barriers
and employment outcomes following welfare reform.
However, wording of the work barrier questions used in
this study differed from WES. In WES, barriers were
constructed based on objective indicators such as self-
reported educational level. Our measures asked for
respondents’ perception of the extent to which various
factors acted as a barrier to looking for or getting a job.
Participants rated on a five-point scale ranging from 1 (not
at all) to 5 (a great deal) ‘‘How much does each of the
following things prevent you from looking for a job or
from getting a job?’’ Inadequate human capital was
assessed by two ratings in response to ‘‘not having enough
education’’ and ‘‘not having enough job experience’’.
Inadequate job quality was assessed by the two ratings in
response to ‘‘jobs with too little pay’’ and ‘‘jobs without
health benefits’’. Childcare related barrier was assessed by
two ratings in response to ‘‘cost/availability of childcare’’
and ‘‘wanting/needing to stay home with your children’’.
Finally, difficulties with transportation were assessed by
the rating in response to ‘‘having transportation problems’’.
For our initial confirmatory factor analysis of work barri-
ers, each of the items served as an indicator of the
respective latent construct. In the final model to predict
employment experience, the means of the items for each
type of barrier were indicators for the respective barrier,
with a total of five indicators for the latent factor of work
Work Experience (Employment) was measured at T2, 4
months after T1 baseline data collection. Respondents
reported whether they are working for pay or not, and, if
working, the number of hours per week they worked for
pay. The analyses below focus on the following dependent
variables of employment (1) dichotomous work variable
coded 0 for reports of not working at all and 1 for reports of
working any number of hours (Fig. 2), (2) dichotomous
work variable coded 0 for reports of not working or
working up to 19 h per week and 1 for reports of working
20 or more hours per week, and (3) work variable based on
the reported number of hours working for pay per week
(including 0 for not working).
Am J Community Psychol (2007) 40:301–312 305
Although approximately 450 of the original WNJ
participants either declined to participate or could not be
located at the T2 4-month follow-up, analyses revealed
that T2 participants and non-participants did not differ on
most variables, including race, education attainment,
marital status, number of children, children under age 5 at
home, employment intention, job search self-efficacy,
mastery, and the work barriers (depression, quality of
jobs, human capital, kids transportation). However, par-
ticipants who failed to complete the T2 follow-up were
slightly younger (M = 28.36, SD = 7.47) than those that
did complete the follow-up (M = 29.70, SD = 8.09) (t
(1,401) = –2.98, p\.01). This difference is very small
and may have little practical significance. T2 participants
also differed with respect to financial strain: those who
did not complete the follow-up survey had lower levels of
financial strain than those that did complete the follow-up
(M = 3.35, 3.48, respectively, t (1,395) = –2.54, p\.05),
and again, while the difference is statistically significant,
it is small.
Table 1 displays the correlation matrix, means, and
standards deviations of demographic variables and vari-
ables used in both models displayed in Figs. 1 and 2. For
the most part, as expected, desirable attributes (e.g.,
mastery, job search self-efficacy, and employment inten-
tion) were negatively correlated with financial strain and
work barriers. The various work barriers were positively
Overview of the Analytic Procedures
Models were tested by a confirmatory latent-variable
structural analysis using EQS version 6.1 (Bentler 2005).
Analyses were based on maximum likelihood procedure
applied to listwise covariance matrices. In accordance with
guidelines for reporting structural equation modeling
results (Boomsma 2000; Raykov et al. 1991) we report the
following goodness-of-fit measures: normed fit index
(NFI), nonnormed fit index (NNFI), comparative fit index
(CFI), and the misfit measure known as the root-mean
square error of approximation (RMSEA). NFI, NNFI, and
CFI fit indices that exceed .90 indicate that the data provide
acceptable fit for the model (Raykov et al. 1991). Good-
ness of fit indices that meet or exceed .95 and a RMSEA
index at or below .06 are indicative of good fit (Hu and
Model 1. Do Work Barriers Constitute a Uni-
We tested a confirmatory factor model of work barriers
with one second-order factor of work barriers indicated by
five first-order factors that represented the five types of
work barriers (i.e., human capital, quality of jobs, child-
care, transportation, and depression). The results of the
final model are presented in Fig. 1. The results for the
estimation of the work barrier model demonstrate accept-
able fit with v2(df = 32, N = 1,250) = 63.67; NFI = .986;
NNFI = .990; CFI = .993; RMSEA = .028. These results
highlight a coherent underlying structure to work barriers
even as these factors represent unique types of work
Model 2. Pathways to Employment: Does the Work
Barrier Construct Predict Negative Employment
Next we tested a structural model for predicting employ-
ment experience at T2 4 months following participation in
WNJ. Employment experience was analyzed using a
dichotomous dependent variable indicating if respondents
were not working at all (coded 0) or working 1 or more
hours per month (coded 1); 43% of respondents (N = 539)
were not working at all, and 57% were working between 1
and 75 h per week (M = 33.64, SD = 10.72). Because of
the inclusion of this dichotomous variable the model was
estimated using a maximum likelihood with robust proce-
dure in order to obtain the Sattora–Bentler corrected v2(S–
Bv2) as well as the corrected standard errors for the
parameters (see, Bentler 2005). The model included one
observed (i.e., measured) variable, work experience, and
five latent factors with their respective indicators (see,
Section on Measures), with the relevant paths of influence
explained by the study hypotheses. The work barrier factor
was developed based on Model 1. Each work barrier was
indicated by the mean score of the relevant items, resulting
in the five work barriers indicators described in the mea-
surement section (depressive symptoms, inadequate human
capital, inadequate job quality, childcare problems, and
transportation problems). In Model 2 we reduced the
number of individual indicators for each barrier to attain
greater parsimony while maintaining the theoretical rele-
vance of the work barriers construct. The model and results
are presented in Fig. 2.
Prior to testing the structural model we estimated its
underlying measurement model. The results for the mea-
surement model demonstrated acceptable fit with v2
(df = 76,
N = 846) = 241.84;
CFI = .95; RMSEA = .050. The estimation results for the
NFI = .93;NNFI = .93;
306 Am J Community Psychol (2007) 40:301–312
structural model also provided acceptable fit with S–Bv2
(df = 80,
N = 846) = 281.42;
CFI = .94; RMSEA = .055. Because of the potential for
bias due to considerable attrition at T2 posttest, we con-
ducted an analysis using EM imputation procedure
available in EQS software and estimated the model in
Fig. 2 with a fully imputed data set. The results demon-
strated that the data provided slightly better fit to the model
N = 1,404) = 384.02 with NFI and NNFI = .93, CFI = .94,
and RMSEA = .050 [.046 .057]. Most importantly, the size
of the beta coefficients of the paths in the model remained
virtually the same as well as the coefficients’ level of sta-
tistical significance (or non-significance for the path from
job search efficacy to work).
Although it is more parsimonious to report results using
either ‘‘working’’ or ‘‘not working’’ as the outcome vari-
able, we tested two additional models with different work
experience outcomes. Given that many low-income women
are employed in jobs that do not have steady, consistent
weekly hours, it was important to use several different
employment outcome variables to demonstrate that the
model results were not sensitive to small changes in the
outcome variable. Alternative models were tested using: a
dichotomized variable with those not working or working
NFI = .92;NNFI = .92;
fewer than 20 h per week (coded 0) versus working at least
20 h per week (coded 1); then once more by the reported
number of hours working per week. The 20-h rate is con-
sistent with the federal welfare work requirements policies
that require recipients to work, do community service, or
participate in education for 20 h a week. For the dichoto-
mous variable, 61% of respondents were not working or
working fewer than 20 h per week. For the continuous
variable reflecting number of hours worked per week,
participants worked an average of 33.64 h (SD = 10.72).
The results for the estimation of the two models demon-
strated equally good fit and nearly the same path
coefficients. The results for the former model were S–Bv2
(df = 80,
N = 846) = 278.61;
CFI = .94; RMSEA = .054. The results for the latter model
(df = 80, N = 846) = 311.15; NFI = .93;
NNFI = .93; CFI = .94; RMSEA = .058.
In addition to the overall good fit to the data, the model
lends support to most of our specific hypotheses (Fig. 2)
with statistically significant paths. Only one path, from job
search self-efficacy to employment experience, did not
attain statistical significance. Another path, from work
barriers to employment intention, suggests that work bar-
riers increased the intention to work rather than to decrease
it as was originally hypothesized. It seems that the
NFI = .92; NNFI = .92;
Table 1 Correlation matrix
Study variable1.2.3. 126.96.36.199. 8.9. 10. 11.12.13. 14. 15.
4. Mastery–.04.13 .20–
5. Job search self-efficacy.07 .19.16.38–
6. Financial strain.20.18 –.13–.18 .09–
7. Employment intention .01.15 .02 .15.32 .18–
8. Mean score of all work barriers–.00–.11–.18 –.35–.11.31.05–
9. Human capital work barrier–.08–.34–.09–.19–.17 .07–.02.57 –
10. Job quality work barrier.08.04–.00–.08 .08.16.09.60.19–
11. Childcare work barrier –.23–.04–.13–.13–.07 .08.03.57.17.20–
12. Transportation work barrier.12–.00 –.12–.26–.07.20.00.59.11.13.09–
13. Depression work barrier.08.04–.20–.40–.13.44.05.66 .20.19.19 .48–
14. Employment at T2 –.01.12.06.11.14–.01 .12 –.04–.02 .08–.04–.06–.09–
15. Hours worked per week–.02.16 .05.13.14.01.14–.05–.06.06–.05–.05–.07.92–
29.27 1.881.76 3.023.913.444.25 2.632.583.192.96 1.772.66.4314.51
SD7.91.80 .56.52 1.03.941.02.731.241.301.321.181.09.5018.09
Note: All the correlations that are equal or larger than .07 are statistically significant at the .05 level. Higher scores correspond with higher
variable levels on all variables
aEducation: 1 = less than high school, 2 = high school graduate, 3 = some college, 4 = college or post-college graduate
bRace: 1 = Caucasian/white, 2 = African-American and other minorities
cEmployment at T2: 0 = not working, 1 = working
Am J Community Psychol (2007) 40:301–312307
realization of the adversity in the form of barriers serves to
increase the intention to seek work for welfare-to-work
clients who are mandated to do so while still having a net
adverse effect on employment outcomes. Their situation
may be analogous to people who need to lose weight and
face obstacles or barriers to engage in weight loss behavior.
It would be reasonable to hypothesize that they would form
stronger intentions exactly because of their need to lose
weight than their counterparts who do not have to lose
weight as urgently.
Both mastery and financial strain were shown to have a
strong impact on work barriers. While sense of mastery
decreased perceived barriers (b = –.49), financial strain
increased perceived barriers (b = .46). Despite the positive
effect of work barriers on intention (b = .21), work barriers
had a negative impact on work experience (b = –.11).
Whereas work barriers bring good intentions, they produce
poor employment outcomes as also shown in past research
(Corcoran et al. 2003; Zedlewski 2003).
The remaining pathways to work experience were quite
similar to those shown in research using large and diver-
sified community samples of unemployed job seekers (e.g.,
Vinokur and Schul 2002). Sense of mastery increased
decreased financial strain (bs= .31, .45 and .–19, respec-
tively). Job search self-efficacy and financial strain
search self-efficacy, and
While it is known that barriers to work are negatively
associated with women’s employment (Corcoran et al.
2000; Danziger et al. 2000; Nam 2005; Zedlewski 2003),
little is known about the pathways by which barriers hinder
employment of women on welfare include personal resil-
(exceptions include Danziger et al. 2001; Jackson 2000;
Jackson and Scheines 2005; Kalil et al. 2001). Studies
investigating the effects of barriers on employment rarely
control for personal resource factors such as financial
strain, mastery, and self-efficacy (e.g., Nam 2005; Taylor
and Barusch 2004), nor do most studies examine respon-
dents’ perceived work barriers. This study addresses these
gaps in the literature and makes a unique contribution by
examining pathways to employment, including resiliency
factors, among low-income women.
mastery and self-efficacy
Fig. 1 Uni-dimensional
N = 1,250) = 63.67; NFI = .986, NNFI = .990, CFI = .993, and
RMSEA = .028
(df = 32,
R² = .20
R² = .53
R² = .03
R² = .22
0 = no, 1 = yes
R² = .06
Fig. 2 Structural model of
pathways to employment among
low-income women. S–Bv2
(df = 80, N = 846) = 281.42,
NFI = .92, NNFI = .92,
CFI = .94, and RMSEA = .055.
Paths are shown with the
standardized coefficients (bs),
all significant at p\.05, with
the exception of the indicated
pathway from job search self-
efficacy to work
308Am J Community Psychol (2007) 40:301–312
We first examined the underlying structure of work
barriers and then tested their contribution to work experi-
ence in the context of a structural equation model that
incorporated other central pathways to employment.
Results demonstrated that a diverse set of impediments to
getting or holding a job, including inadequate human
capital, inadequate job quality, childcare problems, trans-
portation problems, and depression, were indicators of an
underlying uni-dimensional construct that can be con-
ceived of as work barriers. Furthermore, the construct was
shown to have a net adverse effect on employment out-
comes, controlling for job search self-efficacy and
The model offers numerous implications for interven-
tions to reduce the adverse effects of work barriers. Not all
women are ready and able to respond to the welfare sys-
tem’s ‘‘work-first’’ mandate—particularly with regards to
maintaining employment over time—and welfare clients
themselves indicate that they are in need of services to
address a broad range of work barriers (Danziger and
Seefeldt 2002). This study suggests the need for interven-
tion at multiple levels: many women may benefit from
individual supports that promote mastery and self-efficacy
in the transition to work and programs that address mental
health problems; furthermore, there is clearly the necessity
for systems-level programs and policies that direct concrete
resources, such as subsidized transportation and childcare,
to low-income women. Building from this framework, we
discuss two approaches that may be useful in addressing
work barriers, particularly when used in conjunction with
one another. First, short-term interventions focusing on
‘‘soft skills’’ to promote women’s well-being beyond the
work-first focus on employment skills (Holzer et al. 2004);
and second, policies responsible for the transfer of
resources to low-income women.
As it stands now, the welfare-to-work model used in
many locations often requires welfare applicants to attend
job training or work-first classes that are focused on
immediate re-attachment to the labor market. Content often
includes traditional job search skills, such as re ´sume ´
writing or making job contacts. The current study suggests
that traditional job training programs that also enhance
mastery and job search self-efficacy may improve welfare
clients’ job readiness by increasing their intention to search
for a job, become employed, and stay in the job market.
Such programs have been successfully used in numerous
settings with unemployed working- and middle-class
individuals (e.g., Caplan et al. 1989; Vuori et al. 2002;
Vuori and Vinokur 2005). For example, the Jobs Program
(Caplan et al. 1989; Vinokur et al. 1995) was designed as a
job-search skill enhancement workshop to promote reem-
ployment among the recently unemployed. However,
program activities were designed to increase self-efficacy
in the job search and also to inoculate participants against
setbacks in the search for new employment (Vinokur and
Schul 1997; Vuori and Vinokur 2005).
In a low-income context, this model should be extended to
address common work barriers through structured activities
intended to reduce some of the adverse effects of work bar-
riers. For example, the learning context could include
activities that help women to anticipate those barriers that
they are likely to experience while employed, and then
activities to facilitate the development of a plan to address the
barriers. These activities are likely to bolster confidence in
one’s ability to deal with work barriers while being employed
and contribute to inoculation against setbacks that promote
effective behavior change in stressful situations (Meichen-
with specific goal-directed responses as well as engaging in
inoculation against setbacks are effective means to promote
goal-directed motivation and implementation of behavioral
change in numerous settings (Gollwitzer and Brandstatter
1997; Gollwitzer and Kinney 1989; Gollwitzer 1999),
including among the unemployed seeking jobs (Vuori and
A second level of intervention is improved policies and
programs that promote work, which have been discussed in
detail elsewhere (e.g., Hamilton 2002; Michalopoulos et al.
2000a; Pavetti et al. 1996; Scarpace et al. 2005). Policies
that ‘‘make work pay,’’ such as income disregards and
expanded eligibility for transitional Medicaid, increase
women’s ability to obtain and sustain employment (e.g.,
Bos et al. 1999; Michalopoulos et al. 2000b; Miller et al.
2000). Increasing the minimum wage and more job training
and educational programs are needed so that working a
full-time, minimum wage job pushes poor families out of
poverty (Lino 1994).
Perhaps most relevant to the barriers discussed in this
study, there is now a substantial literature pointing to the
need for expanded subsidized childcare options, such Head
Start and public pre-kindergarten programs, and childcare
that responds to the employment constraints of low-income
earners (e.g., accommodate nonstandard work schedules)
(Coley et al. 2001; Lee 2004).
Whereas this is the first study to examine the structure of a
work barrier construct and pathways connecting work
barriers and psychological constructs to employment, it has
a number of limitations that should be addressed in future
studies. First, we focus on those barriers that are most
common among low-income women and thus do not
include substance dependence, criminal record, lack of
access to services, and physical health problems (e.g.,
Am J Community Psychol (2007) 40:301–312309
Dooley and Prause 2002; Gutman et al. 2003; Taylor
2001). The second limitation of this study results from the
fact that the respondents participated in a mandated job
preparation program, which may have helped them over-
come barriers thus reducing the impact of work barriers on
employment in our sample. Finally, the longitudinal aspect
of our model only pertains to the measurement of
employment outcomes. All other variables were assessed at
Time 1, and as such constitute a cross-sectional design
where the direction of influence may be the reverse of what
appears in our model. For example, work barriers may have
a causal influence on financial strain rather than be its
outcome as suggested in our model. A more definitive
study should include a wider array of measured work
barriers and a design with more follow-up periods of all
measures using a sample of respondents whose experience
is not affected by participation in a program.
Summary of Results
In this study we expand current knowledge related to low-
income women and work by establishing that perceived
work barriers form a uni-dimensional construct and then
examining the independent adverse effect of work barriers
on employment within multiple other pathways to
employment, including personal resiliency factors. Results
suggest that welfare programs and policies would better
serve clients if they addressed the problems that are com-
monly experienced by women transitioning from welfare to
work. This could be accomplished through development of
innovative programs that identify a broad range of barriers
to work and serve to improve well-being, as well as com-
prehensive policies that provide concrete resources to low-
Ajzen, I. (2002). Perceived behavioral control, self-efficacy, locus of
control, and the theory of planned behavior. Journal of Applied
Social Psychology, 32, 665–683.
Bandalos, D. L. (2002). The effects of item parceling on Goodness-of-
fit and parameter estimate bias in structural equation modeling.
Structural Equation Modeling, 9, 78–102.
Bandura, A. (1989). Human agency in social cognitive theory.
American Psychologist, 44, 1175–1184.
Bandura, A., & Cervone, D. (1983). Self-evaluative and self-efficacy
mechanisms governing the motivational effects of goal systems.
Journal of Personality and Social Psychology, 45, 1017–1028.
Bentler, P. M. (2005). EQS version 6.1 structural equation program
manual. Los Angeles: BMDP Statistical Software.
Blank, R. (2002). Evaluating welfare reform in the United States.
Journal of Economic Literature, XL, 1105–1166.
Boomsma, A. (2000). Reporting analyses of covariance structures.
Structural Equation Modeling, 7, 461–482.
Bos, J. M., Huston, A. C., Granger, R. C., et al. (1999). New hope for
people with low incomes: Two-year results of a program to
reduce poverty and reform welfare. New York: MDRC.
Caplan, R. D., Vinokur, A. D., Price, R. H., & van Ryn, M. (1989).
Job-seeking, reemployment and mental health: A randomized
field experiment in coping with job loss. Journal of Applied
Psychology, 74, 759-769.
Coley, R. L., Chase-Lansdale, P. L., & Li-Grining, C. P. (2001). Child
care in the era of welfare reform: Quality, choices and
preferences (welfare, children and families: A three-city study,
policy brief 01–4). Baltimore: Johns Hopkins University.
Corcoran, M., Danziger, S. K., Kalil, A., & Seefeldt, K. S. (2000).
How welfare reform is affecting women’s work. Annual Review
of Sociology, 26, 241–269.
Corcoran, M., Danziger, S. K., & Tolman, R. (2003). Long term
employment of African American and White welfare recipients
and the role of persistent health and mental health problems.
Women & Health, 39, 21–32.
Conger, R. D., Conger, K. J., Matthews, L. S., & Elder, G. H. (1999).
Pathways of economic influence on adolescent adjustment.
American Journal of Community Psychology, 27, 519–541.
Conger, K. J., Rueter, M. A., & Conger, R. D. (2000). The role of
economic pressure in the lives of parents, their adolescents: The
family stress model. In L. J. Crockett & R. K. Silbereisen (Eds.),
Negotiating adolescence in times of social change (pp. 201–
223). Cambridge: Cambridge University Press.
Danziger, S., & Johnson, R.C. (2005, winter). Welfare reform: The
morning after (pp. 9–15). The Milken Institute Review.
Danziger, S. K., Ananat, E. O., & Browning, K. G. (2004). Childcare
subsidies and the transition from welfare to work. Family
Relations, 53, 219–228.
Danziger, S. K., Carlson, M. J., & Henly, J. R. (2001). Post-welfare
employment and psychological well-being. Women & Health,
Danziger, S. K., Corcoran, M., Danziger, S., Heflin, C., Kalil, A.,
Levine, J., Rosen, D., Seefeldt, K., Siefert, K., & Tolman, R.
(2000). Barriers to employment of welfare recipients. In R.
Cherry, & W. M. Rodgers III (Eds.), Prosperity for all? The
economic boom and African Americans. New York: Russell Sage
Danziger, S., & Kalil, A. (2002). Welfare reform: Effects of TANF on
family well-being. In R. M. Lerner, F. Jacobs, & D. Wertlieb
(Eds.), Handbook of applied developmental science: Promoting
positive child, adolescent, and family development (pp. 395–
420). Thousand Oaks: Sage.
Danziger, S. K., & Seefeldt, K. S. (2002). Barriers to employment and
the ‘‘hard to serve’’: Implications for services, sanctions, and
time limits. Focus, 22, 76–81.
Derogatis, L. R., Lipmann, R. S., Rickels, K., Uhlenhuth, E. H., &
Covi, L. (1974). The Hopkins Symptom Checklist (HSCL): A
measure of primary symptom dimensions. In P. Pichot (Ed.),
Psychological measurements in psychopharmacology: Vol. 7.
Modern problems in pharmacopsychiatry (pp. 79–110). Basel:
Dooley, D., & Prause, J. (2002). Mental health and welfare
transitions: Depression and alcohol abuse in AFDC women.
American Journal of Community Psychology, 30, 787–813.
Elder, G. H., Jr., & Russell, S. T. (2000). Surmounting life’s
disadvantage. In L. J. Crockett & R. K. Silbereisen (Eds.),
Negotiating adolescence in times of social change (pp. 17–35).
Cambridge: Cambridge University Press.
Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of
simple plans. American Psychologist, 54, 493–503.
Gollwitzer, P. M., & Brandstatter, V. (1997). Implementation
intentions and effective goal pursuit. Journal of Personality
and Social Psychology, 73, 186–199.
310Am J Community Psychol (2007) 40:301–312
Gollwitzer, P. M., & Kinney, R. F. (1989). Effects of deliberative and
implemental mind-sets on illusion of control. Journal of
Personality and Social Psychology, 56, 531–542.
Gutman, M. A., McKay, J., Ketterlinus, R. D., & McLellan, A. T.
(2003). Potential barriers to work for substance-abusing women
on welfare: Findings from the CASAWORKS for families pilot
demonstration. Evaluation Review, 27, 681–694.
Gyamfi, P., Brooks-Gunn, J., & Jackson, A. (2001). Associations
between employment and financial and parental stress. Women
and Health, 32, 119–135.
Hamilton, G. (2002). Moving people from welfare to work: Lessons
from the National Evaluation of Welfare-to-Work Strategies.
New York: MDRC.
Haskins, R., & Offner, P. (2003). May, Achieving compromise on
welfare reform reauthorization (Brookings Institution policy
brief welfare reform and beyond #25). Washington: The
Holzer, H. J., Stoll, M. A., & Wissoker, D. (2004). Job performance
and retention among welfare recipients. Social Service Review,
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in
covariance structure analysis: Conventional criteria versus new
alternatives. Structural Equation Modeling, 6, 1–55.
Jackson, A. P. (2000). Maternal self-efficacy and children’s influence
on stress and parenting among single black mothers in poverty.
Journal of Family Issues, 21, 3–16.
Jackson, A. P., Brooks-Gunn, J., Huang, C., & Glassman, M. (2000).
Single mothers in low-wage jobs: Financial strain, parenting, and
preschooler’s outcomes. Child Development, 71, 1409–1423.
Jackson, A. P., & Scheines, R. (2005). Single mothers’ self-efficacy,
parenting in the home environment, and children’s development
in a two-wave study. Social Work Research, 29, 7–20.
Johnson, R.C., & Corcoran, M.E. (2003). The road to economic self-
sufficiency: Job quality and job transition patterns after welfare
reform. Journal of Policy Analysis & Management, 615–639.
Kalil, A., Schweingruber, H. A., & Seefeldt, K. S. (2001). Correlates
of employment among welfare recipients: Do psychological
characteristics and attitudes matter? American Journal of
Community Psychology, 29, 701–723.
Kline, R. B. (2005). Principles and practice of structural equation
modeling (2nd edn.). New York: The Guilford Press.
Kunz, J., & Kalil, A. (1999). Self-esteem, self-efficacy, and welfare
use. Social Work Research, 23, 119–126.
Lee, S. (2004). Women’s work supports, job retention, and job
mobility: Child care and employer-provided health insurance
help women stay on jobs (Research-in-Brief #C359). Washing-
ton: The Institute for Women’s Policy Research.
Lichter, D. T., & Jayakody, R. (2002). Welfare reform: How do we
measure success? Annual Review of Sociology, 28, 117–141.
Lino, M. (1994). Income and spending patterns of single-mother
families. Monthly Labor Review, 117(5), 29–37
Loprest, P. (2002a). New federalism national survey of America’s
families: Who returns to welfare? Washington: The Urban
Loprest, P. (2002b). Snapshots of America’s families III, No. 3:
Disconnected welfare leavers face serious risks. Washington:
The Urban Institute.
Loprest, P., & Zedlewski, S. (2006). The changing role of welfare in
the lives of low-income families with children (occasional paper
#73). Washington: The Urban Institute.
Meichenbaum, D. (1985). Stress inoculation training: A clinical
guidebook. New York: Pergamon.
Michalopoulos, C., Card, D., & Gennetian, L. A. (2000a). The self-
sufficiency program at months: Effects of a financial work
incentive on employment and income. Ottawa: Social Research
and Demonstration Corp.
Michalopoulos, C., Schwartz, C., & Adams-Ciardullo, D. (2000b).
What works best for whom: Impacts of 20 welfare-to-work
programs by subgroup. New York: MDRC.
Miller, C., Knox, V., Gennetian, L. A. et al. (2000). Final report on
the Minnesota family investment program. Vol. 1, effects on
adults. New York: Manpower Demonstration Research Corp.
Nam, Y. (2005). The roles of employment barriers in welfare exits
and reentries after welfare reform: Event history analyses. The
Social Service Review, 79, 268–293.
Pavetti, L., Olson, K. K., Pindus, N. M., Pernas, M., & Isaacs, J.
(1996). Designing welfare-to-work program for families facing
personal or family challenges. Washington: Urban Institute.
Perlin, L., & Schooler, C. (1978). The structure of coping. Journal of
Health and Social Behavior, 19, 2–21.
Pollack, H. A., Danziger, S., Seefeldt, K. S., & Jayakody, R. (2002).
Substance use among welfare recipients: Trends and policy
responses. The Social Service Review, 76, 256–276.
Reporting structural equation modeling results in Psychology
and Aging: Some proposed guidelines. Psychology and Aging, 6,
Raver, C. C. (2003). Does work pay psychologically as well as
economically? The role of employment in predicting depressive
symptoms and parenting among low-income families. Child
Development, 74, 1720–1736.
Sandanger, I., Moum, T., Ingebrigtsen, G., Dalgard, O. S., Sorensen,
T., & Gruusgaard, D. (1998). Concordance between symptom
screening and diagnostic procedure: The Hopkins Symptom
Checklist-25 and the composite international diagnostic inter-
view. Social Psychiatry and Psychiatric Epidemiology, 33(7),
Scarpace, R., Crichton, L. I., & Meyer, V.A. (2005). Minnesota
Family Investment Program longitudinal study: Four years after
baseline (10th report in a series). St. Paul, MN: Minnesota
Department of Human Services. Available online at: http://www.
Taylor, L. C. (2001). Work attitudes, employment barriers, and
mental health symptoms in a sample of rural welfare recipients.
American Journal of Community Psychology, 29, 443–463.
Taylor, M. J., & Barusch, A. S. (2004). Personal, family, and multiple
barriers of long-term welfare recipients. Social Work, 49,
van Ryn, M., & Vinokur, A. D. (1992). How did it work? An
examination of the mechanisms through which a community
intervention influenced job-search behavior among an unem-
ployed sample. American Journal of Community Psychology, 5,
Vinokur, A., & Caplan, R. D. (1987). Attitudes and social support:
Determinants of job seeking behavior and well-being among the
unemployed. Journal of Applied Social Psychology, 17,
Vinokur, A. D., Price, R. H., & Schul, Y. (1995). Impact of the
JOBS intervention on unemployed workers varying in risk for
depression. American Journal of Community Psychology, 23,
Vinokur, A. D., & Schul, Y. (1997). Mastery and inoculation against
setbacks as active ingredients in the JOBS intervention for the
unemployed. Journal of Consulting and Clinical Psychology, 65,
Vinokur, A. D., & Schul, Y. (2002). The web of coping resources and
pathways to reemployment following a job loss. Journal of
Occupational Health Psychology, 7, 68–83.
Vinokur, A. D., Schul, Y., Vuori, J., & Price, R. H. (2000). Two years
after a job loss: Long term impact of the JOBS program on
reemployment and mental health. Journal of Occupational
Health Psychology, 5(1), 32–47.
Am J Community Psychol (2007) 40:301–312 311
Vuori, J., Silvonen, J., Vinokur, A. D., & Price, R. H. (2002). The Download full-text
Tyo ¨ho ¨n Job Search Program in Finland: Benefits for the
unemployed with risk of depression or discouragement. Journal
of Occupational Health Psychology, 7, 5–19.
Vuori, J., & Vinokur, A. D. (2005). Job-search preparedness as a
mediator of the effects of the Tyo ¨ho ¨n Job Search Intervention on
re-employment and mental health. Journal of Organizational
Behavior, 26, 275–291.
Zedlewski, S. R. (2003). Snapshots of America’s families III, No. 3:
Work and barriers to work among welfare recipients in 2002.
Washington: The Urban Institute.
312 Am J Community Psychol (2007) 40:301–312