Journal of Vocational Behavior 75 (2009) 212–223

Contents lists available at ScienceDirect

Journal of Vocational Behavior

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

Test of a social cognitive model of work satisfaction in teachers

Ryan D. Duffy a,*, Robert W. Lent b
a
University of Florida, Department of Psychology, PO Box 112250, Gainesville, FL 32611, USA
b
University of Maryland, Department of Counseling and Personnel Services, 3214 Benjamin Building, College Park, MD 20742, USA

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

Article history: Lent and Brown [Lent, R. W., & Brown, S. D. (2006). Integrating person and situation perspec
Received 2 June 2009 tives on work satisfaction: A social-cognitive view. Journal of Vocational Behavior, 69, 236–
Available online 11 June 2009 247] recently proposed an integrative model of work satisfaction linked to social cognitive
career theory. The model posits that work satisfaction is predicted by five classes of vari-
ables: work conditions, goal progress, self-efficacy, goal and efficacy relevant supports,
Keywords: and personality traits (e.g., positive affect). We tested this model in a sample of 366 teach-
Work satisfaction
ers, finding good overall model-data fit. Of the five predictor classes, work conditions, self-
Work conditions
Social cognitive career theory
efficacy, and positive affect were each found to explain unique predictive variance. This
Self-efficacy suggests that teachers who are most satisfied with their jobs see their work environment
Support as supportive, are confident in their abilities to complete work-related tasks and goals, and
Positive affect report high levels of trait positive affect. Findings also offered support for the contention
Teachers that measures of subjective person–environment fit may not be empirically distinct from
work satisfaction. Implications of these findings for future research and applied efforts
are considered.
Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction

Work (or job) satisfaction can be defined as the extent to which people like or dislike their jobs, including the work con-
ditions within which their jobs are embedded (Lent, 2008; Spector, 1997). In both counseling psychology and industrial/
organizational psychology, a vast amount of research has been aimed at understanding the factors that predict work satis-
faction, and both of these psychological specialties have spawned theoretical models used to explain why certain individuals
are more or less satisfied with their jobs. The key components of these models include person-environment fit (Dawis & Lof-
quist, 1984; Kristof-Brown, Zimmerman, & Johnson, 2005), goal setting (Locke & Latham, 1990), personality and affect (Brief
& Weiss, 2002; Judge, Heller, & Mount, 2002), work characteristics (Hackman & Oldham, 1976), expectations (Porter &
Lawler, 1968; Vroom, 1964), and a variety of self-related cognitions, such as self-efficacy, self-esteem, and locus of control
(Judge, Erez, Bono, & Thoresen, 2002).
Lent and Brown (2006) recently proposed an integrative model of work satisfaction. This model combines many of the
traditionally studied components of job satisfaction (e.g., work related fit, trait affect) into a unified, empirically-testable
framework that is conceptually linked to social cognitive career theory (Lent, Brown, & Hackett, 1994). It was also designed
to feature several variables that may be relatively modifiable (e.g., self-efficacy, goal participation) and, thus, may help to
inform counseling interventions aimed at work adjustment concerns. Although this model has been found to be useful in
predicting educational and social domain satisfaction in previous studies (e.g., Lent, Singley, Sheu, Schmidt, & Schmidt,
2007; Lent et al., 2005), it has yet to be tested specifically in samples of working adults.

* Corresponding author. Fax: +1 302 831 2148.

E-mail address: rduf@ufl.edu (R.D. Duffy).

0001-8791/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jvb.2009.06.001
 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223 213

The primary purpose of this study was to test the Lent and Brown (2006) model within a sample of school teachers.
Teachers were chosen for two main reasons. First, there is much concern with the satisfaction and retention of teachers
in the US, especially given high rates of teacher dropout (Perie & Baker, 1997). Identification of modifiable precursors to their
work satisfaction could, therefore, be socially useful. Second, using individuals from a single occupation provides a common
referent for assessing participants’ work tasks, goals, and conditions, thereby aiding efforts to generalize the findings. More-
over, it has been suggested that it is useful to employ relatively homogeneous samples when testing new models (Weston &
Gore, 2006).

1.1. Central predictors of work satisfaction

Along with the outcome variable of work satisfaction, the Lent and Brown (2006) model contains five classes of predictor
variables (see Fig. 1). These include (a) personality/affective traits (path 1), (b) participation in/progress at goal-directed
activities (path 2), (c) work-related self-efficacy (path 3), (d) work conditions (path 4), and (e) goal and efficacy relevant envi-
ronmental supports or obstacles (path 5). Each of these five predictor classes has been related to satisfaction within work or
educational domains in previous research; we briefly review relevant findings in this section.
Lent and Brown (2006) posit that several types of personality or affective traits are linked to work satisfaction and most
prior tests of their model in non-work domains have focused on trait positive affect, which refers to an individual’s tendency
to experience a positive state of emotion (Watson, Clark, & Tellegen, 1988). There has been much research on the relation of
positive affect to job satisfaction. In a meta-analysis of 27 studies, Connolly and Viswesvaran (2000) reported a .49 correla-
tion between positive affect and job satisfaction. In another meta-analysis of 79 studies, Thoresen, Kaplan, Barsky, Warren,
and de Chermont (2003) found a correlation of .34. These studies indicate that there is a moderate to strong relation between
positive affect and job satisfaction, suggesting that individuals who generally experience positive emotion are more likely to
be satisfied at work.
A second component of the Lent and Brown (2006) model is goal participation. Having important goals, and making pro-
gress towards these goals, has been empirically supported as a key predictor of general well-being (Lent, 2004; Ryan & Deci,
2001). Locke and Latham (1990) extended this principle to the work domain and suggested that work-related goals would
serve a motivating function and thus promote work performance and satisfaction. The Lent and Brown (2006) model specif-
ically posits that work satisfaction is partly a result of making progress toward one’s work-related goals. Work-related goal
progress has been found to be moderately to strongly related to work satisfaction in samples of workers both in the US.
(Wiese & Freund, 2005) and Germany (Maier and Brunstein (2001). Goal progress has also been found to relate strongly
to college students’ academic satisfaction (Lent et al., 2005; Lent et al., 2007).
A third component of the work satisfaction model is self-efficacy, referring to the belief that one is capable of performing
specific behaviors, including those necessary to achieve particular goals. A number of recent studies with working adults
(e.g., Judge et al., 2002; Judge, Erez, Bono, & Thoresen, 2003; Judge, Bono, Erez, & Locke, 2005) have found that generalized
self-efficacy relates to job satisfaction; Judge and Bono’s (2001) meta-analysis found that the average correlation was .38

Personality/ 1
Affective Traits

6 Self Efficacy
7 Expectations

3
9
8
12

Goal and Efficacy 10

Relevant Participation 2
in/Progress at Goal Work
Environmental Satisfaction
Supports, Resources, Directed Activity
and Obstacles
11 13

4
Work Conditions

Fig. 1. Lent and Brown’s (2006) model of work satisfaction.

214 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223

across 12 studies. The Lent and Brown (2006) model proposes that the link between self-efficacy and work satisfaction is
most pronounced when self-efficacy is measured in job or goal-specific terms. Work task-related self-efficacy has been found
to correlate moderately to strongly with job satisfaction in organizational (Chen, Goddard, & Casper, 2004) and teacher sam-
ples (Caprara, Barbaranelli, Borgogni, & Steca, 2003). Academic goal self-efficacy has also been found to relate strongly to
academic satisfaction (Lent et al., 2005; Lent et al., 2007) in college student samples.
Research on working conditions has dominated job satisfaction research for many years, and the work conditions con-
struct is a fourth component of the Lent and Brown (2006) model. Lent (2008) divided research on work conditions into sev-
eral categories, two of which include person–environment (P–E) fit and perceived organizational support. Recent research on
P–E fit has examined workers’ perceptions of needs/supplies fit (NS) and person/occupation fit (PO). NS fit pertains to the
congruence between an employee’s needs and the rewards provided by his or her work; PO fit refers to the congruence be-
tween an employee’s values and that of the organizational culture (Cable & DeRue, 2002). In a recent meta-analysis, Kristof-
Brown et al. (2005) found that PO fit (.44, k = 65) and NS fit (.61, k = 32) each correlated substantially with job satisfaction.
The degree to which employees feel supported by their work organization is another important aspect of work conditions
(Lent, 2008). Rhoades and Eisenberger (2002) found that perceived organizational support (POS) related strongly (r = .59)
to job satisfaction in a meta-analysis of 21 studies.
The final component of Lent and Brown’s (2006) work satisfaction model concerns the degree to which one receives sup-
port versus barriers specifically in relation to his or her work goals and self-efficacy. This aspect of the work satisfaction mod-
el has received the least amount of research to this point. Some research (e.g., Babin & Boles, 1996; Baruch-Feldman,
Brondolo, Ben-Dayan, & Schwartz, 2002) has related general work support received from co-workers, supervisors, family
members, and friends to job satisfaction, with correlations ranging from .17 to .41. Other research has found goal support
to be strongly related to satisfaction in non-work (e.g., academic, social) domains (Lent et al., 2005; Lent et al., 2007).

1.2. Relations among the predictors

In addition to the direct paths proposed between the predictors and work satisfaction, the Lent and Brown (2006) model
proposed a network of relations among the predictor variables.

1.2.1. Positive affect and work-related self-efficacy (path 6)

A number of studies have linked positive affect to generalized self-efficacy and non-work domain self-efficacy (Caprara &
Steca, 2006; Judge & Ilies, 2004). However, there is a dearth of studies exploring the relation of positive affect to work do-
main-specific self-efficacy. In a conceptually relevant study, Lent et al. (2005) found moderate to strong relations of positive
affect to domain-specific academic and social self-efficacy. Additionally, self-efficacy was found to mediate the relation of
positive affect to satisfaction. This suggests that the relation of positive affect to domain satisfaction may operate partly
through an individual’s perceived abilities in that domain.

1.2.2. Positive affect and goal support (path 7)

Positive affect has been found to be related to general social support (Lakey & Scoboria, 2005). A few studies have found
weak to moderate relations between trait affect and social support specifically within the work domain (Swanson & Power,
2001; Wong, Cheuk, & Rosen, 2000). Lent et al. (2005) found positive affect to be moderately related to support received
within academic and social environments. Additionally, environmental support was found to partially mediate the relation
of positive affect to satisfaction.

1.2.3. Work-related self-efficacy and goal progress (path 8)

Lent et al. (2005), Lent et al. (2007) found that domain-specific self-efficacy correlated strongly with goal progress. The
findings of both studies also indicated that goal progress partially mediated the relation of self-efficacy to satisfaction. This
suggests that the relation of self-efficacy to domain satisfaction operates partly through goal progress. However, neither
study was specifically tied to the work domain.

1.2.4. Goal support and work-related self-efficacy (path 9)

A number of studies have found domain specific supports to correlate moderately with self-efficacy for completing aca-
demic tasks (Lent et al., 2001; Lent et al., 2003; Lent et al., 2005). Empirical research on the connection between goal-specific
support and self-efficacy has been less common. However, a few studies have found these constructs to be moderately cor-
related and have also reported that self-efficacy mediates the relation between environmental supports and domain satis-
faction (Lent et al., 2005; Lent et al., 2007). That is, those who felt support to achieve their academic goals had higher
levels of self-efficacy and, in turn, higher academic satisfaction.

1.2.5. Goal support and goal progress (path 10)

The Lent and Brown (2006) model suggests that progress at work goals depends partly on the environmental support one
receives for one’s work-related goals. Lent et al. (2005; Study 1) and Lent et al. (2007) found environmental support to cor-
relate moderately to strongly with goal progress within academic domains. They also reported that goal progress partially
mediated the goal support–satisfaction relation.
 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223 215

1.2.6. Work-related self-efficacy, goal progress, goal support, and work conditions (paths 11–13)
The final three proposed links in the Lent and Brown (2006) model each involve the work conditions construct which,
according to Lent (2008), may be represented by indicators or person-environment fit (e.g., NS fit, PO fit) and/or perceived
organizational support (POS). Lent and Brown proposed that work conditions partly mediates the relations of self-efficacy
and goal support to work satisfaction, and that goal progress partly mediates the relation of work conditions to work satis-
faction. Although relevant linkages have been studied in the academic and social domains (Lent et al., 2005; Lent et al., 2007),
there is a need for research examining the relations of work conditions to self-efficacy, goal progress, and goal support spe-
cifically in samples of employed workers.

1.3. Hypotheses

Based on Lent and Brown’s (2006) work satisfaction model and prior research, we hypothesized that the five sets of pre-
dictor variables would each produce positive direct paths to work satisfaction (see Fig. 1). Also in keeping with this model,
we posited several sets of mediating relationships – in particular, that (a) progress at work-related goals will partially medi-
ate the relation of work-related self-efficacy, goal support, and work conditions to work satisfaction; (b) work related self-
efficacy will partially mediate the relation of goal support and positive affect to work satisfaction; (c) work conditions will
partially mediate the relation of self efficacy and goal support to work satisfaction; and (d) goal support will partially medi-
ate the relation of positive affect to work satisfaction. We also expected that the Lent and Brown (2006) model would provide
good overall fit to the data.

1.4. The structure of work satisfaction

A secondary purpose of this study was to explore the structure of work satisfaction vis-à-vis perceptions of organizational
support and subjective P–E fit. Lent (2008) had categorized the latter two constructs as alternative indicators of how workers
view the adequacy of their work conditions (e.g., positive work conditions could be indexed as ratings of strong POS or as the
degree to which one’s needs/values are congruent with the organizational environment). Both of these indictors were as-
sumed to be distinct from, and predictive of, work satisfaction. However, Edwards, Cable, Williamson, Lambert, and Shipp
(2006) had questioned whether measures of subjective P–E fit are, in fact, empirically distinct from work satisfaction. In their
view, asking whether one is getting what one wants from one’s work environment may be conceptually equivalent to asking
whether one is happy with work one’s work situation.
To explore this issue empirically, we preceded our test of the integrative model by examining the dimensionality of work
satisfaction, POS, and subjective P–E fit. In particular, we tested the fit to the data of five alternative measurement models.
The first was a one-factor model in which measures of work satisfaction, P–E fit (NS and PO fit), and POS were all set to load
on the same latent factor. Second, we tested a three-factor model in which the measures of work satisfaction, P–E fit, and POS
composed three distinct latent dimensions. In our third through fifth models, we examined alternative two-factor models in
which (a) work satisfaction and P–E fit composed one factor and POS composed a second factor, (b) work satisfaction and
POS consisted of one factor and P–E fit was a separate factor, and (c) POS and P–E fit formed one factor, with work satisfac-
tion as a second factor.
The Edwards et al. (2006) position might suggest that the third model (work satisfaction and subjective P–E fit represent
the same underlying construct) is the most plausible. Lent’s (2008) conceptual scheme would favor the fifth model. Because
the other models are also theoretically reasonable, we examined them as well. Our plan was, first, to identify the most
empirically defensible representation of work satisfaction and then to incorporate it within the subsequent test of the inte-
grative model of work satisfaction.

2. Method

2.1. Participants

The sample consisted of 366 teachers employed full time (295 women, 70 male, one sex-unidentified) in the North Car-
olina Association of Independent Schools. Their ages ranged from 21 to 70 with an average of 44.59 years (SD = 11.34) and
their teaching experience ranged from 1 to 42 years (M = 15.69; SD = 10.06). In terms of race/ethnicity, 342 of the partici-
pants were White (93%), 6 were Native American (2%), 5 were Latino/a (1%), 4 identified as ‘‘other” (1%), 3 were Asian Amer-
ican (1%), and 3 were African American (1%). The sample included elementary (n = 138, 38%), middle (n = 133, 36%), and high
school teachers (n = 93, 26%).

2.2. Measures

2.2.1. Work satisfaction

Two measures of work satisfaction were used as observed indicators of the work satisfaction construct: the five-item
version of the Brayfield and Rothe (1951) Index of Job Satisfaction (Judge, Locke, Durham, & Kluger, 1998) and the Teacher
216 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223

Satisfaction Scale (TSS; Lim-Ho & Tung-Au, 2006). Participants were asked to respond to the items of the Index of Job
Satisfaction on a 7-point scale ranging from strongly disagree to strongly agree. Sample items included, ‘‘I feel fairly well
satisfied with my present job” and ‘‘Most days I am satisfied with my work”. In terms of validity, this scale correlated
strongly with the Job Descriptive Index (r = .89), observer ratings of job satisfaction (r = .59), and life satisfaction (r = .68)
(Judge et al., 1998). It has been found to yield internal consistency reliability coefficients ranging from .82 to .95 (Ilies &
Judge, 2003; Judge et al., 2005; Judge & Ilies, 2004). In the current study, the internal consistency reliability estimate was .81.
The TSS adapted the five items of Diener et al.’s (1985) general satisfaction with life scale to apply specifically to satis-
faction with teaching. Participants were asked to respond to items on a 5-point scale ranging from strongly disagree to
strongly agree. Sample items included, ‘‘In most ways, being a teacher is close to my ideal” and ‘‘My conditions of being a
teacher are excellent”. This scale correlated positively with the Brayfield and Rothe (1951) Index of Job Satisfaction
(r = .50) and Warr’s Job Satisfaction Scale (r = .47), and negatively with psychological distress (r = .27) and anxiety
(r = .34) (Lim-Ho & Tung-Au, 2006). For the current study, the internal consistency reliability of this scale was .86.

2.2.2. Positive affect

This construct was measured with the Positive Affect (PA) items of the Positive and Negative Affect Scales (PANAS; Wat-
son et al., 1988). Participants were presented with a list of 10 emotions (e.g., inspired, attentive) and asked to indicate the
extent to which they have felt each of them during the past few weeks, using a 5-point scale ranging from very slightly to
extremely. In the development study, the internal consistency reliability coefficient for the Positive Affect (PA) scale was
.88 and the scale correlated negatively, as expected, with state anxiety (r = .35) and depressive symptoms (r = .36) (Wat-
son et al., 1988). PA has also been found to correlate with intrinsic motivation (r = .32) and interest in work activities (r = .57)
(Bye, Pushkar, & Conway, 2007). In the current study, the internal consistency reliability of this scale was .92.

2.2.3. Work-related goal progress

Progress at work-related goals was measured with a 5-item scale adapted from Lent et al. (2005). The original instrument
was used to assess progress at a particular goal, and was not specific to the work domain. For the current study, participants
were asked to write down their most important work-related goal and to respond to each statement on a 5-point scale rang-
ing from strongly disagree to strongly agree. Although participants were likely to have multiple work related goals, a focus was
placed on one goal to avoid confusion in answering items if different levels of progress were made on different goals. Par-
ticipants were provided the definition of a work goal and also were given two examples of goals. Sample items include, ‘‘I am
making good progress on my work goal” and ‘‘My pursuit of my work goal has been productive”. Lent et al. (2005) found the
scale to yield an internal consistency reliability estimate of .89. Additionally, the scale correlated moderately to strongly with
domain-specific goal self-efficacy (r = .46), domain satisfaction (r = .66), and life satisfaction (r = .42) (Lent et al., 2005). In the
current study, we found an internal consistency reliability estimate of .86 for this scale.

2.2.4. Work-related goal support

The degree to which participants felt supported in the pursuit of their work-related goals was measured with a modified
version of a scale originally designed to measure marital partner goal support (Brunstein, Dangelmayer, & Schultheiss, 1996).
The items for the goal support scale were adjusted to reflect support received from three sets of individuals in the partici-
pants’ work environment: fellow teachers, immediate supervisors, and school administrators. Each source of support was
assessed with its own subscale. Sample items are, ‘‘my fellow teachers reliably assist my attempts to accomplish my goal
when I ask them to do so” and ‘‘my fellow teachers behave in ways that conflict with my attempts to accomplish this goal”.
Each item was rated on a 7-point scale, ranging from completely disagree to completely agree. Brunstein et al. (1996) found
marital partner goal support to correlate between .34 and .50 with relationship mood and .21 with relationship goal com-
mitment. In the current study, the internal consistency reliability estimates for teacher support, supervisor support, and
administrative support were, respectively, .61, .71, and .82.

2.2.5. Work-related self-efficacy

Self-efficacy was assessed with three scales. First, we modified a goal self-efficacy instrument (Karoly & Reuhlman, 1995)
to assess the degree to which participants felt capable of achieving their most important work-related goal. The original
instrument was used to assess self-efficacy with regard to a particular goal and was not specific to the work domain. Sample
items included, ‘‘I possess the necessary skills to attain my work goal” and ‘‘I have the ability to reach my work goal”. Par-
ticipants responded to each item on a 4-point scale ranging from not at all accurate to extremely accurate. Karoly and Reuhl-
man (1995) reported reliability estimates for this scale ranging from .80 to .87 for health, interpersonal, and academic goals.
Additionally, Karoly and Ruehlman (1995) found their goal self-efficacy measure correlated with self monitoring (r = .49),
planning (r = .38), and depression (r = .50); Lent et al. (2005) found that it related moderately to domain satisfaction,
r = .39. In the current study, this scale produced an internal consistency reliability estimate of .91.
Second, the Teacher Self-efficacy Scale-Short Form (Tschannen-Moran & Woolfolk-Hoy, 2001) was used to assess self-
efficacy regarding teaching-specific behaviors. Participants were asked to rate, along a 5-point scale, the degree to which
they felt they could perform 12 teaching-related behaviors. Sample items include, ‘‘How much can you do to motivate
students who show low interest in school work” and ‘‘To what extent can you craft good questions for your students”.
Tschannen-Moran and Woolfolk-Hoy (2001) reported an internal consistency reliability estimate of .90 and found the
 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223 217

total scale to correlate positively with a previous measure of teacher self-efficacy (r = .48), teacher locus of control
(r = .33), and responsibility for student achievement (r = .46). We found an internal consistency reliability estimate of
.87 for this scale in the present study.
The third self-efficacy measure, the Personal Efficacy Beliefs Scale (PEBS; Riggs, Warka, Babasa, Betancourt, & Hooker,
1994), was used to tap confidence regarding more general work tasks. Participants indicated their perceived efficacy on
10 items along a 5-point scale ranging from strongly disagree to strongly agree. Sample items included, ‘‘I have confidence
in my ability to do my job”, and ‘‘I have all the skills needed to perform my job very well”. Previous research (Lam, Chen,
& Schaubroeck, 2002; Schaubroeck, Lam, & Xie, 2000) has found this instrument to yield internal consistency reliability esti-
mates of .90 to .95, and to correlate positively with job satisfaction (r = .22 to .34) (Lubbers, Loughlin, & Zweig, 2005). We
obtained an internal consistency reliability estimate of .82 for this scale in the current sample.

2.2.6. Work conditions

Three measures were used to index participants’ perceptions of the favorability of their work environments. Person/
organization (PO) fit and needs/supplies (NS) fit (Cable & DeRue, 2002) tapped two aspects of perceived fit between one-
self and one’s work environment. Both scales contained three items (e.g., ‘‘My personal values match my organization’s
values and culture”, PO fit; ‘‘There is a good fit between what my job offers me and what I am looking for in a job”, NS
fit). Participants were asked to respond to each item on a 7-point scale ranging from strongly disagree to strongly agree.
Cable and Derue (2002) found that the PO fit and NS fit scales each produced internal consistency estimates ranging
from .89 to .91. In terms of validity, they each correlated substantially with occupational commitment (.33 to .43)
and perceived organizational support (.40 to .53). In the current study, the internal consistency reliabilities for PO Fit
and NS fit were, respectively, .96 and .92.
The third aspect of work conditions was assessed with the Perceived Organizational Support Scale-Short Form (SPOS;
Eisenberger, Huntington, Hutchison, & Sowa 1986), reflecting the degree to which participants feel supported by their orga-
nization. The SPOS contains the 16 highest-loading items from the original 36-item POS scale. Participants responded to each
item on a 7-point scale from strongly disagree to strongly agree. The original instrument development study found the SPOS to
yield internal consistency reliability estimates ranging from .80 to .93, and to correlate with absenteeism (Eisenberger et al.,
1986). In a meta-analysis, perceived organizational support was found to correlate highly with organizational commitment
(r = .60) and moderately with job involvement (r = .33) (Rhoades & Eisenberger, 2002). The scale produced an internal con-
sistency estimate of .95 in the current study.

2.3. Procedure

The survey was administered via the internet to teachers in schools that were members of the North Carolina Asso-
ciation of Independent Schools (NCAIS). This school district is composed of 3500 teachers affiliated with 82 different
schools across the state. Administrators at each school were asked by the Executive Director of NCAIS to forward infor-
mation about the electronic survey to the teachers at their schools via email. Upon accessing the survey website, teach-
ers were provided with a description of the study and an informed consent form; they were then directed to the survey
itself. As an incentive for participation, participants were given the option to enter into a raffle for one of five $100 gift
certificates. Over the course of a one-month period, three emails containing the survey link and cover letter were sent
from the Executive Director to the school administrators, who then could decide whether to forward the survey link on
to the teachers at their school. Teachers at 26 (32%) of the schools within the district participated in the survey, with an
approximately 38% response rate among the teachers at these 26 schools. The schools were all non-parochial and were
distributed throughout North Carolina, although the majority (20 of 26) of them were located in the greater Charlotte,
Greensboro, or Raleigh metropolitan areas.

3. Results

3.1. Preliminary analyses

When computing descriptive statistics for each of the 13 measured variables, we observed that a large number of
them were negatively skewed and kurtotic. To create more normally distributed scores, the variables were subjected
to rank transformations and then converted to z scores (McDonald, 1999). These transformed values were used in all
subsequent analyses. Table 1 contains the original means and standard deviations and the correlations among the
rank-transformed scales. Our subsequent analyses (confirmatory factor analysis and structural equation modeling) in-
volved latent factors. Each factor was represented by multiple observed indicators. Self-efficacy and goal support were
each represented by three indictors, and work satisfaction was indexed by two measures. Item parcels were used to cre-
ate two indictors each for the positive affect and goal progress constructs, with items being randomly assigned to each
parcel (Little, Cunningham, Shahar, & Widaman, 2002). Five indicators of work conditions (PO fit, NS fit, and three item
parcels derived from the SPOS) were also used in the initial measurement models exploring the structure of work
satisfaction.
218 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223

Table 1
Correlations among independent and dependent variables.

M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. Job Satisfaction 31.16 3.62 –
2. Teaching Satisfaction 20.55 3.49 .64 –
3. Positive Affect Parcel 1 19.79 3.10 .55 .52 –
4. Positive Affect Parcel 2 19.72 3.27 .57 .52 .87 –
5. Goal Progress Parcel 1 11.81 2.13 .38 .30 .34 .36 –
6. Goal Progress Parcel 2 7.43 1.53 .28 .21 .31 .32 .74 –
7. Goal Self-efficacy 13.61 2.25 .26 .25 .37 .36 .37 .37 –
8. Task Self-efficacy 41.49 5.28 .30 .30 .36 .36 .18 .20 .51 –
9. Teacher Self-efficacy 50.59 5.34 .38 .41 .52 .51 .28 .24 .37 .50 –
10. Coworker Support 11.21 2.46 .35 .38 .27 .28 .35 .27 .26 .12 .27 –
11. Supervisor Support 11.64 2.58 .35 .36 .25 .26 .28 .22 .19 .09* .13 .48 –
12. Administration Support 11.16 2.84 .46 .41 .34 .40 .40 .34 .27 .10* .24 .55 .65 –
13. Needs-Supplies Fit 17.54 3.53 .65 .69 .43 .44 .36 .25 .22 .13 .27 .42 .45 .55 –
14. Person-Organization Fit 17.48 3.63 .51 .52 .33 .35 .24 .16 .17 .11 .26 .35 .38 .50 .67 –
15. Perceived Org. Support 89.95 17.07 .55 .56 .36 .41 .28 .20 .16 .12 .34 .39 .49 .68 .62 .67 –

Note: N = 338.
*
p > .05.

3.2. The structure of work satisfaction and work conditions

Prior to testing the structural model in Fig. 1, we examined the dimensionality of the work satisfaction and work condi-
tions constructs, using confirmatory factor analyses (CFA). These analyses were performed with covariance matrices and the
maximum likelihood (ML) estimation procedures of EQS 6.1 (Bentler & Wu, 2005). For each factor, one observed variable
loading was fixed to 1 and the other loadings were freely estimated. Our CFAs were tested with the following goodness
of fit indices: v2, comparative fit index (CFI), and root mean square error of approximation (RMSEA). A non-significant v2
suggests that the model fits the data adequately, though the v2 is problematic as a fit index because it is sensitive to sample
size. The CFI determines if the hypothesized model is a better fit to the data than a null model, where no relations among the
constructs are expected. Values of CFI range from 0 to 1 and Hu and Bentler (1999) have suggested a minimum cutoff of .95.
RMSEA assesses the degree of complexity in the model and whether or not a simpler solution is warranted. Values close to
.06 may be taken as indicating adequate model-data fit (Hu & Bentler, 1999).
The first set of CFAs was aimed at identifying the underlying structure of work satisfaction and the two aspects of work
conditions (subjective P–E fit and POS). Given evidence of multivariate non-normality (Mardia’s normalized estimate = 6.08),
robust ML estimation was employed. As described earlier, we examined the fit of five plausible model variations. First, we
tested a one-factor model in which (a) the two measures of work satisfaction, (b) the two fit indicators (PO fit and NS fit), and
(c) the three POS parcels were all set to load on a single factor. This model yielded unsatisfactory fit indices, Sartorra–Bentler
(S–B) v2(14) = 205.52, p < .001; CFI = .87; RMSEA = .20. A three-factor structure, in which the work satisfaction, P–E fit, and
POS indicators were each modeled as separate but correlated dimensions, was next tested. The fit indices for this model were
S–B v2(11) = 42.06, p < .001; CFI = .98; RMSEA = .09. Although the RMSEA value reflected less than ideal fit, the CFI suggested
that model-data fit was acceptable and a v2 difference test indicated significantly improved fit compared to the one-factor
model, DS–B v2(3) = 122.38, p < .001.
Three two-factor variations were next tested: (a) work satisfaction and P–E fit composed one factor, with POS as a second
factor (S–B v2[13] = 49.43, p < .001; CFI = .98; RMSEA = .09); (b) work satisfaction and POS representing one factor, with P–E
fit as a second factor (S–B v2[13] = 171.08, p < .001; CFI = .90; RMSEA = .19); and (c) POS and P–E fit forming one factor, with
work satisfaction as a second factor (S–B v2[13] = 157.36, p < .001; CFI = .90; RMSEA = .18). While each of these models im-
proved over the fit of the one-factor model, only the first model produced CFI and RMSEA values that rivaled that of the
three-factor model.
These findings, therefore, suggested that the three-factor model and the two-factor model representing P–E fit as loading
on the same latent dimension as work satisfaction were both tenable. In comparing these two models, we noted that the
three-factor model yielded better fit, as judged by the v2 difference test, DS–B v2[2] = 7.44, p < .05; however, inspection
of the factor correlations in the three-factor model revealed that the correlation between work satisfaction and P–E fit ap-
proached one (.92), raising questions about the uniqueness of these two factors. Given the magnitude of this correlation, the
comparable CFI and RMSEA indices of the two models, and the value of parsimony, we considered the two-factor model as
offering the most reasonable representation of work satisfaction and work conditions. This conclusion favors Edwards et al.’s
(2006) position that work satisfaction and subjective P–E fit may be manifestations of the same construct and suggests that
POS is a distinct but related construct.

3.3. Test of the predictive model

Having determined that the indicators of P–E fit were not clearly distinguishable from work satisfaction, we used the POS
item parcels alone to represent work conditions in our subsequent model testing, omitting NS and PO fit as largely redundant
 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223 219

indicators of work satisfaction (rather than work conditions). To assess the factor structure of the remaining variables, we
used CFA to test a six-factor CFA representation of the hypothesized model in Fig. 1. Given evidence of multivariate non-nor-
mality (Mardia’s normalized estimate = 9.20), robust ML estimation was employed.
The six-factor model produced satisfactory values of CFI (.98) and RMSEA (.05), suggesting good fit to the data of this mea-
surement model (S–B v2[75] = 144.57, p < .001). An alternative, five-factor model was also tested and compared against the
six-factor model. Specifically, because perceived organizational support has the potential to overlap, conceptually and
empirically, with the indicators of goal support, we tested a model in which the POS parcels, teacher support, supervisor sup-
port, and administrative support were all set to load on a common factor and self-efficacy, goal progress, positive affect, and
satisfaction were modeled as separate but correlated factors. This model yielded non-optimal indications of fit (S–B
v2[80] = 311.56, p < .001, CFI = .92, RMSEA = .09) and poorer fit relative to the six-factor model, D v2(5) = 210.73, p < .001.
Examination of variable-factor loadings for the six factor model revealed that each variable loaded substantially (P60) on
its appropriate factor. We, therefore, concluded that the six-factor measurement model provided an adequate basis for test-
ing the structural model of work satisfaction.
The six-factor, or target, structural model (shown in Fig. 1) provided good fit to the data (S–B v2[77] = 150.16, p < .001,
CFI = .97, RMSEA = .05). This structural model did not differ significantly from the 6-factor measurement model in terms
of relative model fit (DS–B v2[2] = 5.91, p > .05), supporting the adequacy of its representation of the relations among the
constructs. Structural coefficients, shown in Fig. 2, indicated that positive affect, self-efficacy, and work conditions each pro-
duced significant paths to work satisfaction. Work conditions produced the largest coefficient. However, neither goal pro-
gress nor goal support explained unique variation in work satisfaction. The full model accounted for 75% of the variance
in work satisfaction.
To test mediation within the context of the full model, bootstrap procedures were utilized (Mallinckrodt, Wei, Russell, &
Abraham, 2006; Shrout & Bolger, 2002). The Lent and Brown (2006) model hypothesized eight mediational paths. However,
because some of these paths did not receive support in the structural model test, we examined only three mediational paths:
(a) self-efficacy mediates the relation of positive affect and work satisfaction, (b) work conditions mediate the relation of
self-efficacy and work satisfaction, and (c) work conditions mediate the relation of goal support to work satisfaction.
We followed the bootstrapping recommendations by Shrout and Bolger (2002) to develop more reliable estimates of the
standard errors of the indirect effects within the model. In keeping with their recommendations, we first created 1000 bootstrap
samples using EQS 6.1. Second, we ran the six factor model 1000 times with each of these randomly generated samples. Third,
based on the mean parameter estimates from these 1000 randomly generated models, we calculated the indirect effects for each
of the mediational hypotheses. For example, to calculate the indirect effect of positive affect on work satisfaction, the 1000 path

Positive
Affect

.36*
Self-
.66* Efficacy
.46* .17*

.03
.37*

.53* .04
Goal Goal Work
Support Progress Satisfaction

.14*
-.25*

.70*

Work
.46*
Conditions

.06

Fig. 2. N = 338. Parameter estimates for the Lent and Brown (2006) structural model. *p < .05.
220 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223

coefficients from positive affect to self-efficacy were multiplied by the 1000 path coefficients from self-efficacy to work satis-
faction. This same technique was utilized for the other two mediational hypotheses explored. Finally, the standard errors and
confidence intervals of these indirect effect estimates were calculated. Shrout and Bolger (2002) suggest that if the 95% confi-
dence interval for the indirect effect does not include zero, then the indirect effect is significant (p < .05).
Support was found for two of the three mediated paths. The mean indirect effect of positive affect on work satisfaction, as
mediated by self-efficacy, was found to be .10 (SE = .06, CI = .01, .24); the mean indirect effect of goal support on work sat-
isfaction, as mediated by work conditions, was found to be .40 (SE = .10, CI = .22, .58); and the mean indirect effect of self-
efficacy on work satisfaction, as mediated by work conditions, was found to be .09 (SE = .05, CI = .01, .19). Because the con-
fidence intervals of the latter two indirect effects did not include zero, each of these mediational paths was found to be sta-
tistically significant (p < .05).
In sum, while the CFA was consistent with the six factor representation of the constructs in the Lent and Brown (2006)
model, the results from the structural model offered mixed support for the hypotheses. In particular, the structural model
provided good overall fit to the data. Work conditions, self-efficacy, and positive affect were each found to explain unique
variance in work satisfaction, though goal progress and goal support did not yield significant direct paths to work satisfac-
tion. Additionally, work conditions were found to fully mediate the goal support-work satisfaction relation and to partially
mediate the self-efficacy-work satisfaction relation.

4. Discussion

The current study was aimed at testing the Lent and Brown (2006) model of work satisfaction with a sample of school
teachers. As a preliminary step to testing this model, we sought to explore the dimensionality of work satisfaction relative
to two aspects of work conditions, subjective P–E fit and perceived organizational support. Edwards et al. (2006) had cau-
tioned that subjective P–E fit might not be empirically distinguishable from work satisfaction, whereas Lent (2008) had im-
plied that subjective P–E fit and perceived organizational support could be seen as alternative aspects of work conditions,
both of which may be distinct from work satisfaction.
We tested a series of measurement models, finding the most tenable and parsimonious model to be one depicting sub-
jective P–E fit (NS and PO fit) as loading on the same factor as work satisfaction, with perceived occupational support loading
on a separate but correlated factor. Although a three-factor model also fit the data well, the very high correlation observed
between subjective P–E fit and work satisfaction when they were modeled as separate factors suggested that it is, in fact,
difficult to differentiate them, at least as they were operationalized in this study. Thus, there appears to be merit to Edwards
et al.’s (2006) contention that asking how well one’s work environment fulfills his or her needs/values may, in essence, be an
alternative way of tapping work satisfaction rather than representing a unique construct.
Given the findings of our initial confirmatory factor analyses, we moved on to examining the tenability of the full, inte-
grative model of work satisfaction. We first tested a six-factor measurement model in which self-efficacy, goal progress, goal
support, work conditions (as indexed by perceived organizational support), positive affect, and work satisfaction were mod-
eled as distinct but correlated latent constructs. This six-factor model fit the data well and significantly better than did a five-
factor alternative measurement model.
We next tested the hypothesized direct and indirect effects of the predictors on work satisfaction. The structural model
provided good overall fit to the data, though not all specified paths were consistent with expectations. In particular, three of
the five predictors – positive affect, self-efficacy, and work conditions – were each found to explain unique variance in the
prediction of work satisfaction. The relation of self-efficacy to work satisfaction is in line with past findings in the academic,
social, and work domains (e.g., Caprara et al., 2003; Judge & Bono, 2001; Lent et al., 2005; Lent et al., 2007) and suggests that
teachers with more confidence in performing their work-related tasks and fulfilling their work related goals may be more
satisfied with their work. This is the first study of which we are aware that specifically included goal self-efficacy in the pre-
diction of work satisfaction along with task self-efficacy and general teacher self-efficacy. Because self-efficacy is viewed as a
relatively modifiable variable (Lent & Brown, 2006), this finding may have useful implications for counselors working with
teachers who are dissatisfied with their jobs.
The finding that positive affect relates to work satisfaction is consistent with findings from meta-analyses (Connolly &
Viswesvaran, 2000; Thoresen et al., 2003), and suggests that people with higher levels of trait positive affect are more likely
to have positive attitudes towards their work. The linkage of perceived organizational support to work satisfaction is, like-
wise, consistent with meta-analytic findings (Rhoades & Eisenbeger, 2002). Although we found that goal progress and goal
support each produced significant bivariate relations with work satisfaction, which is consistent with past findings (e.g., Ba-
bin & Bolles, 1996; Baruch-Feldman et al., 2002; Maier & Brunstein, 2001; Wiese & Freund, 2005), these two variables did not
yield unique direct paths to work satisfaction apart from the other predictors.
Taken together, the set of predictor variables within the social cognitive model accounted for 75% of the variance in work
satisfaction. The large percentage of variance accounted for in work satisfaction may be attributable to the integrative nature
of the model (i.e., the predictors in the model had each been found to be moderately to strongly linked to work domain sat-
isfaction in prior research) and the fact that the structural equation model controlled for measurement error. In addition to
the direct paths to work satisfaction, we found that the relations of goal support and self-efficacy to work satisfaction were
mediated by work conditions.
 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223 221

4.1. Implications

The results of the current study may have implications for future researchers studying work satisfaction and practitioners
working with individuals dissatisfied with their jobs. From a research standpoint, the results from the Lent and Brown (2006)
model point to the importance of studying predictors of work satisfaction within the context of a unified model rather than
only in bivariate form, where predictors are examined in isolation from one another. The present findings, along with prior
research (e.g., Judge, Bono, & Locke, 2000; Judge et al., 2005), point especially to the predictive utility of work conditions, self-
efficacy, and positive affect.
Future research on the social cognitive model might take several directions. First, it would be valuable to extend the mod-
el to other samples of teachers (including those who are likely to be less satisfied with, or have less tenure in, teaching than
did the current sample) as well as to different types of occupational groups, which would aid in establishing the model’s
range of generalizability. Second, further research is needed to explore the role of goal support and goal progress as predic-
tors of work satisfaction. Third, while further cross-sectional studies would be useful, longitudinal and intervention studies
are needed to establish the temporal predominance or causal linkages among the variables. Fourth, additional research is
needed to clarify the factor structure of subjective P–E fit vis-à-vis work satisfaction.
The results of the current study may also have implications for counselors working with individuals who are dissatisfied
with their jobs. First, as self-efficacy is considered a relatively modifiable variable (Lent & Brown, 2006), clients who present
with job dissatisfaction might be assisted in strengthening their self-efficacy (and skills) in relation to work tasks in which
they perceive themselves to be deficient or aspects of goal pursuit over which they may assert some measure of personal
control. Lent (2005) noted a number of strategies that counselors might use to bolster a client’s work-related self-efficacy,
including helping clients to develop achievable work-related goals and reinforcing progress on these goals. These activities
may allow clients to gain positive work experiences and receive positive social feedback.
Second, although trait affect and work conditions can be somewhat challenging targets in counseling, assessment proce-
dures may identify particular clients who can benefit from a focus on these issues. For example, clients who exhibit generally
low levels of positive affect (or high levels of negative affect) may profit from affect management strategies that deal with
the cognitive and behavioral concomitants of trait affect (Brown, Ryan, & McPartland, 1996). Other methods may prove use-
ful within the context of counseling or organizational interventions, such as building undeveloped skills, redesigning jobs to
add valued elements, seeking social support from appropriate sources, and (where necessary) considering alternate jobs that
offer a better fit to one’s current work personality. Lyubomirsky, Sheldon, and Schkade (2005) also suggest several general
strategies for promoting satisfaction that could be adapted to work settings.

4.2. Limitations

The current study contains a number of limitations which may temper the interpretation or application of the findings. In
particular, this study involved a sample of teachers who were from one school system, were mostly White and female, expe-
rienced favorable working conditions, and tended to be highly satisfied with their jobs. Such sample characteristics render as
uncertain efforts to generalize the findings to other, more diverse samples of teachers. Second, the cross-sectional nature of
the data precluded study of temporal predominance or causality among the variables and may have resulted in an overes-
timation of mediational effects (Maxwell & Cole, 2007). Third, all of the instruments in the current study were self-report
measures, reflecting a mono-method bias.
Fourth, the study contained limited representations of some of the latent constructs. For example, positive affect alone
was used to represent personality or affective traits. It would be useful to include additional aspects of trait affect (e.g., neg-
ative affect) or personality (e.g., extraversion) in future model tests. Likewise, only goal support was used to assess ‘‘goal and
efficacy relevant environmental supports, resources, and obstacles”. Future tests of the model should focus on resources and
obstacles as well as support. Finally, participants were asked to think only of their most important work goal in assessing the
goal progress construct. Although this procedure helped to clarify the goal referent, it failed to take into account the multiple
goals that many participants may have at work – and the possibility that progress on one’s set of goals may be more predic-
tive of work satisfaction than is progress at a single goal.
In sum, the current results provide preliminary support for the application of the integrative social cognitive model to the
understanding of work satisfaction in teachers. Teachers who were confident in accomplishing their work related tasks and
goals, felt supported by their organization, and possessed high levels of positive affect tended to be the most satisfied with
their work. Although the model fit the data well overall, not all of the hypotheses regarding individual predictors or medi-
ational relations were supported by the findings. Further research is necessary to test the tenability of the integrative model
in more diverse occupational samples.

References

Babin, B. J., & Boles, J. S. (1996). The effects of perceived co-worker involvement and supervisor support on service provider role stress, performance and job
satisfaction. Journal of Retailing, 72, 57–75.
Baruch-Feldman, C., Brondolo, E., Ben-Dayan, D., & Schwartz, J. (2002). Sources of social support and burnout, job satisfaction, and productivity. Journal of
Occupational Health Psychology, 7, 84–93.
Bentler, P.M., & Wu, E.J.C. (2005). EQS 6.1 for Windows, Multi-variate Software, Encino, CA.
222 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223

Brayfield, A. H., & Rothe, H. F. (1951). An index of job satisfaction. Journal of Applied Psychology, 35, 307–311.
Brief, A. P., & Weiss, H. M. (2002). Organizational behavior: Affect in the workplace. Annual Review of Psychology, 53, 279–307.
Brown, S. D., Ryan, N. E., & McPartland, E. B. (1996). Why are so many people happy and what do we do for those who aren’t? A reaction to Lightsey (1996).
The Counseling Psychologist, 24, 751–757.
Brunstein, J. C., Dangelmayer, G., & Schultheiss, O. C. (1996). Personal goals and social support in close relationships: Effects on relationship mood and
marital satisfaction. Journal of Personality and Social Psychology, 71, 1006–1019.
Bye, D., Pushkar, D., & Conway, M. (2007). Motivation, interest and positive affect in traditional and nontraditional undergraduates. Adult Education
Quarterly, 60, 1275–1288.
Cable, D. M., & DeRue, D. S. (2002). The convergent and discriminant validity of subjective fit perceptions. Journal of Applied Psychology, 87, 875–884.
Caprara, G. V., Barbaranelli, C., Borgogni, L., & Steca, P. (2003). Efficacy beliefs as determinants of teachers’ job satisfaction. Journal of Educational Psychology,
95, 821–832.
Caprara, G. V., & Steca, P. (2006). The contribution of self-regulatory efficacy beliefs in managing affect and family relationships to positive thinking and
hedonic balance. Journal of Social & Clinical Psychology, 25, 603–627.
Chen, G., Goddard, T., & Casper, W. J. (2004). Relating general and work-specific self-evaluations and work-related control beliefs to an expanded job
attitudes criterion. Applied Psychology: An International Review, 53, 349–370.
Connolly, J. J., & Viswesvaran, C. (2000). The role of affectivity in job satisfaction: A meta-analysis. Personality and Individual Differences, 29, 265–281.
Dawis, R. V., & Lofquist, L. H. (1984). A psychological theory of work adjustment. Minneapolis: University of Minnesota Press.
Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71–75.
Edwards, J. R., Cable, D. M., Williamson, I. O., Lambert, L. S., & Shipp, A. (2006). The phenomenology of fit: Linking the person and environment to the
subjective experience of fit. Journal of Applied Psychology, 91, 802–827.
Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). Perceived organizational support. Journal of Applied Psychology, 71, 500–507.
Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16,
250–279.
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.
Ilies, R., & Judge, T. A. (2003). On the heritability of job satisfaction: The mediating role of personality. Journal of Applied Psychology, 88, 750–759.
Judge, T. A., & Bono, J. E. (2001). Relationship of core self-evaluations traits – self-esteem, generalized self-efficacy, locus of control, and emotional stability –
with job satisfaction and job performance: A meta-analysis. Journal of Applied Psychology, 86, 80–92.
Judge, T. A., Bono, J. E., Erez, A., & Locke, E. A. (2005). Core self evaluations and job and life satisfaction: The role of self concordance and goal attainment.
Journal of Applied Psychology, 90, 257–268.
Judge, T. A., Bono, J. E., & Locke, E. A. (2000). Personality and job satisfaction: The mediating role of job characteristics. Journal of Applied Psychology, 85,
237–249.
Judge, T. A., Erez, A., Bono, J. E., & Thoresen, C. J. (2002). Are measures of self-esteem, neuroticism, locus of control, and generalized self-efficacy indicators of
a common core construct? Journal of Personality and Social Psychology, 83, 693–710.
Judge, T. A., Erez, A., Bono, J. E., & Thoresen, C. J. (2003). The Core Self-Evaluations Scale (CSES): Development of a measure. Personnel Psychology, 56,
303–331.
Judge, T. A., Heller, D., & Mount, M. K. (2002). Five-factor model of personality and job satisfaction: A meta analysis. Journal of Applied Psychology, 87,
530–541.
Judge, T. A., & Ilies, R. (2004). Affect and job satisfaction: A study of their relationship at work and at home. Journal of Applied Psychology, 89, 661–673.
Judge, T. A., Locke, E. A., Durham, C. C., & Kluger, A. N. (1998). Dispositional effects on job and life satisfaction: The role of core self-evaluations. Journal of
Applied Psychology, 83, 17–34.
Karoly, P., & Ruehlman, L. S. (1995). Goal cognition and its clinical implications: Development and preliminary validation of four motivational assessment
instruments. Assessment, 2, 113–129.
Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C. (2005). Consequences of individuals’ fit at work: A meta-analysis of person–job, person–
organization, person–group, and person–supervisor fit. Personnel Psychology, 58, 281–342.
Lakey, B., & Scoboria, A. (2005). The relational contribution of trait and social influences to the links among perceived social support, affect, and self-esteem.
Journal of Personality, 73, 361–388.
Lam, S. S. K., Chen, X. P., & Schaubroeck, J. (2002). Participative decision making and employee performance in different cultures: The moderating effects of
allocentrism/idiocentrism and efficacy. Academy of Management Journal, 45, 905–914.
Lent, R. W. (2008). Understanding and promoting work satisfaction. An integrative view. In S. D. Brown & R. W. Lent (Eds.), Handbook of counseling
psychology (4th ed.). New York: Wiley.
Lent, R. W. (2004). Toward a unifying theoretical and practical perspective on well-being and psychosocial adjustment. Journal of Counseling Psychology, 51,
482–509.
Lent, R. W., & Brown, S. D. (2006). Integrating person and situation perspectives on work satisfaction: A social-cognitive view. Journal of Vocational Behavior,
69, 236–247.
Lent, R. W., Brown, S. D., Brenner, B., Chopra, S. B., Davis, T., Talleyrand, R., & Suthakaran, V. (2001). The role of contextual supports and barriers in the choice
of math/science educational options: A test of social cognitive hypotheses. Journal of Counseling Psychology, 48, 474–483.
Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive theory of career and academic interest, choice, and performance. Journal of
Vocational Behavior, 45, 79–122.
Lent, R. W., Brown, S. D., Schmidt, J., Brenner, B., Lyons, H., & Treistman, D. (2003). Relation of contextual supports and barriers to choice behavior in
engineering majors: Test of alternative social cognitive models. Journal of Counseling Psychology, 50, 458–465.
Lent, R. W., Singley, D., Sheu, H., Gainor, K. A., Brenner, B. R., Treistman, D., & Ades, L. (2005). Social-cognitive predictors of domain and life satisfaction:
Exploring the theoretical precursors of subjective well-being. Journal of Counseling Psychology, 52, 429–442.
Lent, R. W., Singley, D., Sheu, H., Schmidt, J. A., & Schmidt, L. C. (2007). Relation of social-cognitive factors to academic satisfaction in engineering students.
Journal of Career Assessment, 15, 87–97.
Lim-Ho, C., & Tung-Au, W. (2006). Teaching satisfaction scale: Measuring job satisfaction of teachers. Educational and Psychological Measurement, 66,
172–185.
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural
Equation Modeling, 9, 151–173.
Locke, E. A., & Latham, G. P. (1990). Work motivation and satisfaction: Light at the end of the tunnel. Psychological Science, 1, 240–246.
Lubbers, R., Loughlin, C., & Zweig, D. (2005). Common pathways to health and performance: Job-self-efficacy and affect among young workers. Journal of
Vocational Behavior, 67, 199–214.
Lyubomirsky, S., Sheldon, K. M., & Schkade, D. (2005). Pursuing happiness: The architecture of sustainable change. Review of General Psychology, 9, 111–131.
Maier, G. W., & Brunstein, J. C. (2001). The role of personal work goals in newcomers’ job satisfaction and organizational commitment: A longitudinal
analysis. Journal of Applied Psychology, 86, 1034–1042.
Mallinckrodt, M., Wei, M., Russell, D. W., & Abraham, T. W. (2006). Advances in testing the statistical significance of mediation effects. Journal of Counseling
Psychology, 53, 372–378.
Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12, 23–44.
McDonald, R. P. (1999). Test theory: A unified treatment. Mahwah, NJ: Erlbaum.
 R.D. Duffy, R.W. Lent / Journal of Vocational Behavior 75 (2009) 212–223 223

Perie, M., & Baker, D. P. (1997). Job satisfaction among America’s teachers: Effects of workplace conditions background characteristics and teacher compensation.
Washington, DC: National Center for Education Statistics.
Porter, L. W., & Lawler, E. E. (1968). Managerial attitudes and performance. Homewood, IL: Dorsey Press.
Rhoades, L., & Eisenberger, R. (2002). Perceived organizational support: A review of the literature. Journal of Applied Psychology, 87, 698–714.
Riggs, M. L., Warka, J., Babasa, B., Betancourt, R., & Hooker, S. (1994). Self-efficacy and outcome expectancy scales for job-related applications. Educational
and Psychological Measurement, 54, 793–802.
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on hedonic and eudaimonic well-being. Annual Review of
Psychology, 52, 141–166.
Schaubroeck, J., Lam, S. S. K., & Xie, J. L. (2000). Collective efficacy versus self-efficacy in coping responses to stressors and control: A cross-cultural study.
Journal of Applied Psychology, 85, 512–525.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7,
422–445.
Spector, P. E. (1997). Job satisfaction: Application, assessment, causes, and consequences. Thousand Oaks, CA: Sage.
Swanson, V., & Power, K. (2001). Employees’ perceptions of organizational restructuring: The role of social support. Work & Stress, 15, 161–178.
Thoresen, C. J., Kaplan, S. A., Barsky, A. P., Warren, C. R., & de Chermont, K. (2003). The affective underpinnings of job perceptions and attitudes: A meta-
analytic review and integration. Psychological Bulletin, 129, 914–945.
Tschannen-Moran, M., & Woolfolk Hoy, A. (2001). Teacher efficacy: Capturing an elusive construct. Teaching and Teacher Education, 17, 783–805.
Vroom, V. H. (1964). Work and motivation. New York: Wiley.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of
Personality and Social Psychology, 54, 1063–1070.
Weston, R., & Gore, P. A. (2006). A brief guide to structural equation modeling. The Counseling Psychologist, 34, 719–751.
Wiese, B. S., & Freund, A. M. (2005). Goal progress makes one happy, or does it? Longitudinal findings from the work domain. Journal of Occupational and
Organizational Psychology, 78, 287–304.
Wong, K. S., Cheuk, W. H., & Rosen, S. (2000). The influences of job stress and supervisor support on negative affects and job satisfaction in kindergarten
principals. Journal of Social Behavior & Personality, 15, 85–98.