Вы находитесь на странице: 1из 10

Journal of Vocational Behavior 74 (2009) 5362

Contents lists available at ScienceDirect

Journal of Vocational Behavior


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

The longitudinal impact of self-efcacy and career goals on objective


and subjective career success
Andrea E. Abele *, Daniel Spurk
Social Psychology Group, University of Erlangen-Nuremberg, Bismarckstr. 6, D 91054 Erlangen, Bavaria, Germany

a r t i c l e

i n f o

Article history:
Received 21 July 2008
Available online 1 November 2008

Keywords:
Occupational self-efcacy
Career-advancement goals
Salary
Status
Career satisfaction
Gender
Longitudinal study

a b s t r a c t
The present research reports on the impact of occupational self-efcacy and of careeradvancement goals on objective (salary, status) and subjective (career satisfaction) career
attainments. Seven hundred and thirty four highly educated and full-time employed professionals answered questionnaires immediately after graduation, three years later, and
seven years later. Controlling for discipline, GPA at masters level, and gender, we found
that occupational self-efcacy measured at career entry had a positive impact on salary
and status three years later and a positive impact on salary change and career satisfaction
seven years later. Career-advancement goals at career entry had a positive impact on salary
and status after three years and a positive impact on status change after seven years, but a
negative impact on career satisfaction after seven years. Women earned less than men, but
did not differ from men in hierarchical status and in career satisfaction. Theoretical implications for socio-cognitive theorizing and for career-success research as well as applied
implications for vocational behavior are discussed.
2008 Elsevier Inc. All rights reserved.

1. Introduction
Research interest in career success both regarding objective parameters (e.g., salary, promotions, hierarchical status) and
regarding subjective ones (e.g., subjective evaluation of ones career) has been high for many years. One main strand of research concerns what predicts success. The present research addresses the inuence on career success of two well-known
individual difference variables, namely self-efcacy beliefs and personal goals. Self-efcacy beliefs (Bandura, 1986, 1997)
and personal goals (Austin & Vancouver, 1996; Little, 1983; Locke & Latham, 2002) are important constructs in socio-cognitive models of career interests and performance (Lent, Brown, & Hackett, 1994). There is considerable research on the inuence of self-efcacy and of personal goals on task performance as well as on job performance (e.g., Stajkovic & Luthans,
1998). There are also ndings on self-efcacy and goals inuencing early phases of an individuals career choice (e.g., Betz
& Hackett, 2006). However, there is almost no research on the inuence these variables have on career success conceptualized as the objective and subjective outcomes an individual receives in his/her career. A recent meta-analysis on determinants and correlates of career success (Ng, Eby, Sorensen, & Feldman, 2005) listed not a single study on this topic. The
aim of the present study is to close this research gap. We will present ndings on the impact that self-efcacy beliefs
(i.e., occupational self-efcacy) and personal occupational goals (i.e., career-advancement goals) have on career outcomes
measured both on an objective level (salary, hierarchical status) and on a subjective level (career satisfaction).
* Corresponding author. Fax: +49 9131 8524731.
E-mail address: abele@phil.uni-erlangen.de (A.E. Abele).
0001-8791/$ - see front matter 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jvb.2008.10.005

54

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

2. Self-efcacy and personal goals


Self-efcacy is dened as individuals beliefs about their capability to perform some behavior or to meet a standard. Individuals with high self-efcacy beliefs set higher goals for themselves, put in more effort, and persist longer on a difcult task
(Bandura, 1986; Bandura, 1997). Generalized self-efcacy can be distinguished from more domain-specic self-efcacy. Personal goals (Little, 1983) are aims of an action (Locke & Latham, 2002) or internally represented desired states (Austin &
Vancouver, 1996). They are assumed to inuence outcomes by directing attention, mobilizing effort, affecting persistence,
and structuring behavior. They allow long-term orientation and regulation of ones actions. According to socio-cognitive theorizing (Brown, Jones, & Leigh, 2005) both self-efcacy beliefs and goals are determinants of successful actions.
3. Career success
Career success is dened as the positive psychological or work-related outcomes or achievements one accumulates as a
result of work experiences (Seibert, Crant, & Kraimer, 1999, p. 417). A conceptual distinction between so-called objective
and subjective measures of career success is very frequently made. Criteria of objective success include salary, salary growth,
promotions, or hierarchical status. Criteria of subjective success are, for instance, career satisfaction, comparative judgments,
or job satisfaction (for discussions see Arnold & Cohen, 2008). Many career researchers argue that it is important to assess
both aspects because the meaning of a career can only be understood if different criteria are taken into account (e.g., Arthur,
Khapova, & Wilderom, 2005; Heslin, 2005). Objective and subjective measures correlate positively, but the correlations are
only moderate. Recent meta-analyses have revealed correlations not higher than .30 (Dette, Abele, & Renner, 2004; Ng et al.,
2005). There are also ndings suggesting that the predictors of objective career success differ from the predictors of subjective success and that even within different facets of objective and subjective success predictors differ. Ng et al. (2005), for
instance, argue that individual difference variables account more for subjective parameters than for objective ones.
4. Inuence of self-efcacy and goals on career success
Day and Allen (2004) reported positive correlations between municipal employees career self-efcacy, current salary,
and subjective career success (similarly Valcour & Ladge, 2008). Kim, Mone, and Kim (2008) reported that Korean employees
self-efcacy correlate positively with salary. In contrast, Lubbers, Loughlin, and Zweig (2005) found no association between
job self-efcacy and hourly wage. Saks (1995) showed that task-related self-efcacy of newly hired entry-level accountants
had a positive effect on job satisfaction 10 months later (similarly, Higgins, Dobrow, & Chandler, 2008 on subjective career
success). Frieze, Olson, Murrell, and Selvan (2006) found that MBA graduates materialistic work values (e.g., making a lot of
money) predicted salary 26 years later. Hence, three cross-sectional studies suggest an inuence of self-efcacy on salary
(Day & Allen, 2004; Kim et al., 2008; Valcour & Ladge, 2008), another study suggests no inuence on salary (Lubbers
et al., 2005). Two longitudinal studies reveal an inuence of self-efcacy on job satisfaction or perceived career success (Higgins et al., 2008; Saks, 1995). One study suggests a longitudinal effect of materialistic goals on salary (Frieze et al., 2006).
None of these studies looked at both self-efcacy and goals.
5. Present research
In the present research we analyzed the longitudinal inuence of occupational self-efcacy and of career-advancement
goals on objective success as well as on subjective success across seven years. With regard to self-efcacy beliefs we were
concerned with occupational self-efcacy. Occupational self-efcacy is the belief in ones capacity and motivation to successfully perform occupational tasks and challenges and to pursue ones occupational career irrespective of the particular eld of
occupation (Higgins et al., 2008). Occupational self-efcacy is neither a broad measure of generalized self-efcacy nor a very
specic measure of particular career interests self-efcacy such as, for instance, occupational condence themes according to
Hollands (1997) RIASEC model (for instance, Betz et al., 2003; Wulff & Steitz, 1996). Occupational self-efcacy rather has an
intermediate level of specicity. We chose such a level because research has shown that a medium level of specicity is
advantageous in predicting specic outcomes (Chen, Gully, & Eden, 2001; Pajares, 1996).
Regarding personal occupational goals we were concerned with career-advancement goals. These are directed at climbing
up the career ladder and at being successful in terms of inuence, material gain, and prestige. There were two reasons for
choosing this goal content. First, prestige-, power-, and achievement-goals are important elements in the conceptualization
of life goals (e.g., Phlmann & Brunstein, 1997), and respective career-advancement goals are important elements in the
work values literature (Super, 1970; Zytowski, 1994). Second, previous research in the realm of motivational forces has
shown that power-related motives and materialistic goals were especially important predictors of high achievement (Frieze
et al., 2006; Winter, 1991; Winter, Riggio, Murphy, & Pirozzolo, 2002).
Fig. 1 depicts our hypotheses, theoretical model, and empirical approach. We operationalized objective career success as
salary and hierarchical status and we operationalized subjective success as career satisfaction. At time 1, immediately after
the participants graduated, we assessed occupational self-efcacy and career-advancement goals. We measured objective
career success after 36 months of professional experience and again after 85 months of professional experience. Career satisfaction was measured once after 85 months of professional experience.

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

55

Fig. 1. Theoretical model.

Hypotheses 1 and 2 concern correlations between the variables considered here. In accord with socio-cognitive theorizing
(Brown et al., 2005) we assumed that occupational self-efcacy and career-advancement goals associate positively. Individuals with high career-advancement goals should also have high occupational self-efcacy beliefs. However, individuals with
high occupational self-efcacy do not necessarily also hold high career-advancement goals. Hence, the relationship between
both variables should be positive, but of moderate size.
Hypothesis 1. Occupational self-efcacy and career-advancement goals correlate positively.
In accord with meta-analytical ndings (Dette et al., 2004; Ng et al., 2005) we assume that the two objective success indicators and the subjective success indicator associate positively.
Hypothesis 2. Salary, hierarchical status, and career satisfaction correlate positively.
Hypotheses 35 concern the impact of the predictors on the career success measures. We predicted that occupational
self-efcacy (Day & Allen, 2004; Kim et al., 2008; Valcour & Ladge, 2008) and career-advancement goals (Frieze et al.,
2006) both impact salary. Because our longitudinal research allows testing the inuence of predictors at two times of measurement (i.e., salary and salary change; hierarchical status and status change) we predicted that occupational self-efcacy
and career-advancement goals would not only inuence salary, but also changes in salary.
Hypothesis 3. Occupational self-efcacy and career-advancement goals both positively inuence salary and salary change.
Career-advancement goals should predict hierarchical status as well as changes in status because status is one of the main
objectives of these goals.
Hypothesis 4. Career-advancement goals positively inuence hierarchical status and changes in status.
We tested the inuence of occupational self-efcacy on status and status change in an exploratory fashion. A positive
inuence is conceivable because of the general motivational impact self-efcacy has on performance and on desired outcomes. There could, however, also be no inuence because occupational self-efcacy need not be accompanied by status
goals.
Derived from previous research (Day & Allen, 2004; Higgins et al., 2008; Saks, 1995) and also derived from ndings in the
eld of optimistic expectations and their inuence on outcome evaluations (e.g., Armor & Taylor, 1998) we predicted that
occupational self-efcacy has a positive inuence on career satisfaction.
Hypothesis 5. Occupational self-efcacy positively inuences career satisfaction.
The inuence of career-advancement goals on career satisfaction was again be tested in an exploratory fashion. There
could be no inuence at all, because career-satisfaction should be due to goal-fulllment and not to the degree of careeradvancement goals. A negative inuence is also conceivable such that individuals with high career-advancement goals

56

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

are less easily satised with their careers than individuals with lower career-advancement goals. Research in the realm of
self-determination theory (Ryan & Deci, 2000), for instance, has shown that goals of nancial success were less related to
subjective well-being than more intrinsic goals (e.g., self-acceptance) (Schmuck, Kasser, & Ryan, 2000).
To ensure that our hypothesis testing was as valid as possible, we controlled for several variables that had the potential to
affect our ndings. First, we included only participants who worked full-time, because it has been shown that working hours
have a strong inuence on, for instance, salary or promotions (Ng et al., 2005). Second, we controlled for the occupational eld
our participants were in, because average wages vary between disciplines like, for instance, teaching vs. medicine or law.
Third, we controlled for gender, because many studies show that women earn lower salaries than men (Abele, 2003; Greene
& DeBacker, 2004; Kirchmeyer, 1998; Ng et al., 2005). Finally, we controlled for our participants masters degree Grade Point
Average (GPA), because this also might have an inuence on career success (Ng et al., 2005).
Summarizing, our model suggests thatcontrolling for gender, GPA, and disciplineoccupational self-efcacy and careeradvancement goals inuence career success outcomes, and that these inuences remain signicant if both variables are considered simultaneously.
Our model posits multivariate relationships over time. The emergence of new analytic methods has provided useful tools for
examining such relationships and we tested our hypotheses by means of structural equation modeling (SEM) using Mplus
(Muthn & Muthn, 1998). Structural equation modeling has several advantages. The measurement model of the predictors
can be included; measurement errors can be taken into account; the specic postulated paths can be tested; and besides providing the path coefcients, a series of overall t statistics shows how well the empirical data t the theoretical model (Kline,
2005).

6. Method
6.1. Overview
We tested our hypotheses with data collected in a prospective longitudinal study with a large sample of professionals (see
also Abele, 2003; Abele & Stief, 2004). They were all highly educated and held a masters degree. Two cohorts of graduates
(consecutive graduation years) completed the rst questionnaire some weeks after they had passed their nal exams. Further measures were taken after one-and-a-half years, after three years, and after seven years. We did not nd any cohort
effects or time of measurement effects (cf. Palmore, 1978). Hence, we report results for the combined data from both cohorts.
6.2. Participants and procedure
Due to address protection reasons we were not allowed to send out the rst questionnaire ourselves. Instead the universitys graduation ofce sent (or gave) it to the graduates. We asked our participants to complete and return the questionnaire
together with their addresses, because the study would be continued some time later. From the 4200 questionnaires given
out 1930 (46%) were sent back to the researchers.
Time 1. Participants were 825 women and 1105 men (mean age 27 years). Most of them were German and about ve percent came from other European countries. Ninety-four percent of the respondents provided their address (N = 1819). At time
1, we assessed gender, GPA, study major, occupational self-efcacy, and career-advancement goals among other variables.
Participants who provided their address did not differ from participants who declined to provide their address with regard
to these variables.
Time 2. Participants received the second questionnaire 18 months later. 102 of the 1819 participants who had provided
their address in the rst questionnaire had moved to an unknown address at time 2. Of the remaining 1717 participants,
1397 (588 women and 809 men; mean age 28.5 years) responded to the second questionnaire (response rate 81.4%). We
do not include data from this testing here, because in the German occupational system medical doctors, people in law professions and teachers have to undergo an obligatory 18 months post-gradual training during which they earn a xedlow
salary. Hence, a large proportion of our sample could not provide career success data for time 2 measures.
Time 3. Of the 1663 participants who could be contacted three years after having left university (54 individuals had
moved to an unknown address), 1,330 (561 women, 769 men; mean age 30 years) responded (response rate 80%). A
drop-out analysis revealed that there were no differences (with regard to gender, age, study major, GPA, occupational
self-efcacy, and career-advancement goals) between participants who answered this questionnaire and those who did
not. Among other variables, we measured salary, hierarchical status, and total working hours per week.
Time 4. Seven years after having left university 1415 participants were contacted (116 individuals had moved to an unknown
address, 132 had declined participation already at time 3). Out of these, 1265 participants (527 women, 738 men; mean age 34
years) completed the questionnaire (response rate 89%). There were no differences (same variables tested as at time 3) between
participants who answered this questionnaire and those who did not. We measured the same variables as at time 3.
Present sample. The present analyses were performed with a sub-sample of these 1265 participants. We excluded participants with incomplete data sets (N = 98 had not participated at time 3). We excluded participants who were not employed
at time 3 and/or time 4 (N = 138), because we did not have data on career success for these participants. For the same reason
we also excluded women in maternal leave (N = 83). Finally, we excluded participants who did not work full-time at time 3

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

57

and/or at time 4 (N = 212) because of two reasons. First, there is a strong correlation of hours worked per week with salary
and with promotions (Ng et al., 2005) and we wanted to study our hypotheses without this confound. Second, in most cases
it was not clear whether participants worked part-time because they wanted to or because they could not get a full-time
employment. This information, however, is extremely important when the inuence of self-efcacy beliefs and personal
goals on career successmediated via working hoursis studied (we nevertheless tested our hypotheses also in the sample
in which part-time employed participants were included (N = 971). The ndings [controlling for working hours] were by and
large the same as the ones reported in the results section).
The above exclusion criteria led to a sample of 734 participants (190 women, 544 men) who worked at least 35 h per
week both three years and seven years after entering their career. More women (337 out of 527, i.e., 64%) than men (194
out of 738, i.e., 26%) were excluded. This is due to the fact that women worked less full-time (51%) than men (92%). This,
in turn, is mainly due to parenthood. Seven years after graduation mothers worked M = 12.41 h per week and fathers worked
M = 38.97, t (572) = 27.14, p < .001.
Our 734 participants had graduated in law (16 women, 33 men), medicine (39 women, 101 men), arts and humanities (28
women, 19 men), natural sciences (13 women, 55 men), economics (43 women, 113 men), engineering (12 women, 179
men), and teaching (39 women, 44 men).
6.3. Measures
Grade point average. We standardized the participants individual GPAs using the average of all individuals who had
passed their masters degree in the respective major and year as the criterion. A value of 0 means that the participant
had the same GPA as the average of all graduates of the respective major and respective year; a positive value means that
the participant had a GPA better than average; a negative value means that the participant had a GPA worse than average.
Occupational self-efcacy. The occupational self-efcacy scale (Abele, Stief, & Andr, 2000) consists of 6 items (sample item
I am condent that I could deal efciently with the challenges of my occupation if I only wanted to). Participants responded on 5-point scales (1 = not at all to 5 = very much). In the present sample a one-factorial solution explains 49% of
the item variance and the internal consistency of the scale was Cronbachs a = .78. The occupational self-efcacy scale shows
construct validity (cf. Abele et al., 2000; similarly see Higgins et al., 2008). It is, for instance, associated with protean career
self-directed career management (r = .42, p < .001; Spurk, 2007). The retest-reliabilities of self-efcacy in working populations range between .62 and .75 depending on the interval considered (cf. Frese, Garst, & Fay, 2007; Higgins et al., 2008).
Career-advancement goals. The scale we used to measure career-advancement goals is adapted from the well established
work values inventory (Super, 1970; German version, Seifert & Bergmann, 1983). It consists of ve items (sample items: I
want to make a lot of money; I want to gain high occupational reputation). Participants rated the importance of these
goals on 5-point scales (1 = not important to 5 = very important). The scale is one-dimensional (54% explained item variance)
and showed good internal consistency, Cronbachs a = .77. Supporting its construct validity, the scale is highly correlated
with other measures of career orientation and its retest reliability is high (Abele & Spurk, 2006).
Objective career success. We measured monthly salary before taxes in thirteen steps from no salary, coded as 0; less than
500, coded as 0.5; less than 1,000, coded as 1; and then in equal steps to less than 10,000, coded as 10; and more
than 10,000, coded as 11. The salary variable could vary between zero and 11. We measured hierarchical status by combining information on three variables: permission to delegate work (0 = no, 1 = yes), project responsibility (0 = no, 1 = yes), and
ofcial leadership position (0 = no, 1 = yes). The hierarchical status variable could vary between zero and 3.
Subjective career success. We measured career satisfaction with a German translation of the career satisfaction scale (Greenhaus, Parasuraman, & Wormley, 1990). The scale comprises ve items (sample item: I am satised with the progress I have
made towards meeting my overall career goals). Participants responded on ve-point scales (1 = not at all to 5 = very much).
The scale is one-dimensional (61% explained item variance) and revealed good internal consistency, Cronbachs a = .83.
Analytical strategy. We tested our hypotheses by means of structural equation modeling (SEM) using Mplus (Muthn &
Muthn, 1998). Regarding our dual assessment of the objective success measures (salary, hierarchical status) we estimated
an autoregressive change model with these variables (see Kline, 2005). Hence, the effects of our predictors on salary and
hierarchical status after seven years of professional experience can be interpreted as the effects on the residual change of
these variables. We tested all models using Maximum Likelihood Estimation with robust standard errors (MLR).

7. Results
7.1. Descriptive ndings
Table 1 displays the means, standard deviations, and intercorrelations of the present measures. Our participants mean
GPA was close to the overall average, M = .07 (SD = .51). This suggests that our sample was representative with regard to
the respective graduates population GPA.
As can be seen in Table 1 our participants salary increased (after 3 years of professional experience M = 4.60; after seven
years of professional experience M = 6.21), t (733) = 27.67, p < .001. Their hierarchical status increased, as well (after 3 years
of professional experience M = 1.16; after seven years of professional experience M = 1.71), t (733) = 13.74, p < .001.

58

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

Table 1
Means, standard deviations, and correlations among study variables (N = 734).
Variable

SD

1
2
3
4
5
6
7
8

.07
3.80
3.23
4.60
6.21
1.16
1.71
3.67

.51
.70
.71
1.02
1.78
.97
1.14
.70

.07
.04
.03
.07
.01
.03
.00

.22
.17
.18
.14
.12
.19

.17
.16
.19
.20
-.03

.48
.29
.27
.08

.23
.37
.31

.49
.05

.18

GPA
Occupational SEF at graduationa
Career-advancement goals at graduationa
Salary after three years of professional experienceb
Salary after seven years of professional experienceb
Status after three years of professional experiencec
Status after seven years of professional experiencec
Career satisfaction after seven years of professional experiencea

Note. For rs > .13, p < .001; for rs > .09, p < .01; for rs > .06, p < .05.
a
Scales from 1 (low) to 5 (high).
b
Values from 0 to 11.
c
Values from 0 to 3; SEF = self-efcacy.

We also tested whether there were gender differences in the predictors or control variables. There were no differences in
GPA and in career-advancement goals, both t < 1. Women had lower occupational self-efcacy (M = 3.69) than men
(M = 3.85), t (732) = 2.72, p < .01, the effect size, however, was low, d = .02.
7.2. Hypotheses testing
Measurement model. We rst modeled the measurement of occupational self-efcacy (6 items), of career-advancement
goals (5 items), and of career satisfaction (5 items). Each construct was represented by one latent factor and these factors
were allowed to correlate. We built parcels (Dwyer, 1983) such that every latent factor had three indicators. The t index
of this measurement model (factor loadings see Appendix 1) is good (v2 = 28.26, df = 24, N = 734; CFI = 1.00, TLI = 1.00;
RMSEA = .02; SRMR = .02).
Hypotheses testing models. We built a model aimed at testing our hypotheses stated above. It contained GPA, gender (dummy coded), the participants discipline (dummy coded), salary (times 3 and 4) and status (times 3 and 4), as well as the three
latent factors of self-efcacy, career-advancement goals, and career satisfaction. We tested paths from the control variables,
e.g., participants gender, GPA, and discipline, to occupational self-efcacy, to career-advancement goals, to salary, to status,
and to career satisfaction. We also tested possible interactions between gender and discipline on the variables considered in
the models. There were none, all ps > .05. We estimated the residual correlation (controls partialled out) between occupational self-efcacy and career-advancement goals and the residual correlations between salary, hierarchical status, and career satisfaction. Finally, we tested paths from occupational self-efcacy and career-advancement goals to salary, status, and
to career satisfaction.
The resulting model has good t statistics (v2 = 229.50, df = 96, N = 734, CFI = .96, TLI = .93, RMSEA = .04, SRMR = .02). The
2
R values show that 19% of variance in salary after three years of professional experience, 16% of salary change, and 7% of
variance in career satisfaction after seven years of professional experience are explained by this model.
We rst present the ndings on the inuence of the control variables, e.g., gender, GPA, and discipline because for claritys
sake we did not include the respective paths into the graphical depiction of the model. Table 2 shows the paths discipline or
professional eld (dummy coded), gender (dummy coded), and GPA had on the variables under study as computed by the
structural equation model described above. As can be seen, occupational self-efcacy did barely differ between disciplines

Table 2
Beta effects of discipline (study major), gender, and GPA on latent variables and success measures.

Discipline
Law
Medicine
Arts & Humanities
Science
Economics
Engineering
Gender
GPA

Occupational
self-efcacy at
graduation

Careeradvancement
goals at
graduation

Salary after 3
years of
professional
experience

Salary after 7
years of
professional
experiencea

Status after 3
years of
professional
experience

Status after 7
years of
professional
experiencea

Career satisfaction
after 7 years of
professional
experience

.04
.11
.05
.01
.09
.17*
.09*
.09*

.24***
.28***
.03
.13*
.38***
.29***
.05
.06

.06
.25***
.01
.03
.33***
.19***
.16***
.02

.05
.13**
.04
.09**
.26***
.20***
.11***
.06

.30***
.42***
.25***
.19***
.42***
.38***
.07
.01

.30***
.45***
.26***
.25***
.46***
.44***
.03
.02

.03
.10
.08
.02
.10
.11
.02
.01

Note. Disciplines dummy coded (teaching is reference category).


a
Controlled for prior measure of the same variable.

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

59

(only participants in the eld of engineering had slightly higher means than teachers). Participants working in the elds of
teaching or arts and humanities had lower career-advancement goals than participants in the other disciplines. After three
years of professional experience participants working in the elds of medicine, economics, and engineering had higher salaries than the other participants. After seven years of professional experience these differences remained by and large the
same. The hierarchical status variable differed between teachers (low) and all other professions. Career satisfaction was
the same across the professions. GPA had an inuence on occupational self-efcacy. There were no gender inuences on status or career satisfaction. However, there was an effect on salary both after three and seven years of professional experience.
Women earned less than men.
Fig. 2 depicts the structural equation model with respect to the postulated hypotheses. For clarity reasons we excluded
the paths from discipline, gender, and GPA (see Table 2), and we also excluded non-signicant paths. Supporting Hypothesis 1
the residual correlation of occupational self-efcacy and career-advancement goals was positive (r = .26, p < .001). Supporting Hypothesis 2 the residual correlations of salary and career satisfaction (r = .26, p < .001), status and career satisfaction
(r = .22, p < .001), salary and status (r = .17, p < .001), and salary change and status change (r = .16, p < .001) were positive.
Hypothesis 3 on the impact of occupational self-efcacy and of career-advancement goals on salary was fully supported
for occupational self-efcacy. Participants high in occupational self-efcacy at career entry earned more three years later
(b = .10, p < .01) and had more increase in salary seven years later (b = .08, p < .05) than participants with lower occupational
self-efcacy at career entry. Hypothesis 3 was partly supported for career-advancement goals. Participants with high careeradvancement goals at career entry earned more three years later (b = .11, p < .01) than those with lower career-advancement
goals. However, salary increase was unaffected by these goals (b = .05, ns). Supporting Hypothesis 4 career-advancement
goals had a positive inuence both on status after three years of professional experience (b = .11, p < .01) and on status
change after seven years of professional experience (b = .08, p < .05). The exploratory test of the inuence of occupational
self-efcacy on status revealed a positive effect after three years (b = .09, p < .05), but no effect on status change after seven
years (b = .02, ns).
Supporting Hypothesis 5 occupational self-efcacy had a positive inuence on career satisfaction (b = .26, p < .001). The
exploratory test of the inuence of career-advancement goals on career satisfaction resulted in a signicantly negative impact (b = .10, p < .01). The higher our participants career-advancement goals had been at graduation, the less satised they
were with their career seven years later.
Since GPA had an inuence on occupational self-efcacy (see Table 2) we also tested whether there might be indirect
inuences of GPA on career outcomes mediated via occupational self-efcacy. We computed signicance tests for indirect
effects (cf. Kline, 2005) and found only one indirect effect of GPA on career satisfaction mediated via occupational self-efcacy (product of the two involved beta-coefcients: .023, p < .05). The test of indirect inuences of gender (mediated via
occupational self-efcacy) revealed no signicant results.

Fig. 2. Structural equation model for salary, hierarchical status, and career satisfaction. Only signicant paths displayed; paths from GPA, gender and
discipline, as well as the measurement model for clarity reasons not displayed; all variables except gender regressed on study major (dummy coded); all
variables regressed on gender and GPA; *p < .05, **p < .01, ***p < .001.

60

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

8. Discussion
Hypotheses testing. Supporting Hypothesis 1 occupational self-efcacy and career-advancement goals correlated moderately. The objective and subjective success measures also correlated moderately, supporting Hypothesis 2. The size of the
cross-sectional associations (residual correlations) ranged between .17 and .26. The association between salary and career
satisfaction was somewhat lower (r = .22) than the one reported in the meta-analysis by Ng et al. (2005; r = .30). The other
associations, however, were similar (salary and hierarchical status r = .17; status and career satisfaction r = .26) to those reported by Ng et al. (2005; salary and promotions: r = .18: status and career satisfaction: r = .22).
Regarding the objective success indicators (salary and hierarchical status) the data generally supported Hypotheses 3 and
4. Individuals occupational self-efcacy and their career-advancement goals at career entry had an impact on salary, salary
change, on their hierarchical status, and on status change. The higher the participants self-efcacy and career advancement
goals had been at career entry, the more they earned and the higher was their status later on. The ndings show that both
variables have an independent impact on objective career success and that the impact is evident even 7 years later. Both
variables added to the prediction of success above the included control measures (3% in case of salary and 2% in case of salary
change; 3% in case of status and 1% in case of status change). Career-advancement goals had a relatively stronger impact on
status than on salary, whereas occupational self-efcacy had a relatively stronger impact on salary than on status.
The variances in objective success explained by the predictors were relatively small, but we had not expected higher percentages of explained variance. There are many inuences on salary and status, and individual differences in socio-cognitive
variables are but one source of inuence. Apart from that we controlled for important sources of inuence like discipline,
GPA, and gender, and in case of salary change and status change we also controlled for the auto-regressor. Most importantly,
we covered a long time span (7 years). Therefore, we believe that even the relatively small demonstrated impact of occupational self-efcacy and of career-advancement goals on objective career attainments is of both theoretical and applied relevance. Furthermore, beta effects between .07 and .11 are absolutely in line with previous ndings on individual differences
as predictors of objective career success. Ng et al. (2005), for example, found population correlations (mainly cross-sectional
studies) of .11 for proactivity with salary and .06 for locus of control with salary. Judge and Hurst (2007) found beta effects of
.11 and .12 in longitudinally predicting salary by core self-evaluations.
Regarding the subjective success indicator (career satisfaction) we found supporting evidence for Hypothesis 5. Participants with higher occupational self-efcacy at graduation were more satised with their careers seven years later than those
with lower occupational self-efcacy. This variable added 5% of variance to the prediction of success above the included control measures. Replicating meta-analytical ndings (Ng et al., 2005) the impact of our social-cognitive variables on subjective
success is, hence, somewhat larger than the impact on objective success.
Our exploratory test of the inuence of career-advancement goals at graduation on career satisfaction seven years later
revealed a negative impact. Although individuals with high career-advancement goals became objectively more successful
they were nevertheless less satised with their careers than individuals with lower career-advancement goals. This is a novel
and provocative nding. One interpretation could rely on an individuals aspiration level. If the aspiration level is very high, it
takes longer time to achieve it and hence it also takes longer time to become satised with what one has achieved. This
would mean that the negative impact of career-advancement goals on career satisfaction might disappear over time. Another
interpretation could rely on self-determination theory (Ryan & Deci, 2000) and its assumption that extrinsic goals like
money and status are less rewarding than more intrinsic goals like, for instance, personal growth or self-acceptance
(Schmuck et al., 2000). If this reasoning is correct, the negative impact of career-advancement goals on career satisfaction
should remain stable over time. Further research has to test these possibilities. In any case, the present data suggest that
high career-advancement goals enhance an individuals objective success, but not his/her career satisfaction.
Gender. In our specic sample of full-time working professionals we found no gender differences in career-advancement
goals; women were as interested in making a career as men. However, despite the same GPAs at graduation women had
slightly lower occupational self-efcacy than men. This is in line with prior research (e.g., Betz & Fitzgerald, 1987).
An important nding of our study concerns the gender effect on salary. It is well known that women on average earn less
than men (Ng et al., 2005). However, these general ndings are often not controlled for eld of occupation (i.e., whether men
and women work in elds of occupation comparable in terms of salary), and they are also often not controlled for working
hours (i.e., whether women and men work the same amount of time). In our present study we controlled for eld of occupation and working time. But even though we only included participants working full-time and even though we controlled
for their eld of occupation (i.e., discipline) we still found that women earned less money than men (see also Abele, 2003;
Greene & DeBacker, 2004; Kirchmeyer, 1998). The gender effect was highly signicant three years after work entry and it
remained highly signicant after seven years of professional experience. Womens somewhat lower occupational self-efcacy did not mediate this effect. Interestingly, gender had no inuence on status. This means that in the present sample womens duties seem to have been similar to those of men. Nevertheless they earned less. We therefore can conclude that the
wage gap between full-time working women and men is not due to differences in career-advancement goals or occupational
self-efcacy; it is not due to the eld of occupation; and it also seems not to be due to the duties men and women fulll in
their jobs.
In contrast, gender had no effect on career satisfaction. This missing gender difference in career-satisfaction may be due
to shifting standards (Biernat & Billings, 2001). Because it is well known that women on average are less successful in their

61

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

careers than men, women may apply lower standards for their careers than for mens careers. Therefore despite of lower
salary they are similarly satised with their careers as men.
GPA and discipline. Even if ndings on GPA and on the disciplines our participants were working in are not the focus of the
present research they are nevertheless worthy of a brief discussion. First, it is interesting to see that the inuence of GPA on
career success was more or less negligible. Second, the objective career attainments studied here were generally high in
elds like economics, engineering, and medicine, and they were generally low in teaching. However, despite these marked
differences between disciplines in objective attainments we found no differences between disciplines in career satisfaction.
We believe that this again has to do with different aspirations. As we showed for career-advancement goals (lower in arts
and humanities and in teaching than in the other disciplines) aspirations are different between disciplines. With lower aspirations individuals will be satised with lower objective attainments, and vice versa. Teachers are a good example. When
choosing this profession, future teachers knew that they will earn relatively little and that there are only few possibilities
for promotions in the organizational context of schools. Hence, they were relatively low on career-advancement goals
and seven years after professional experience teachers were as satised with their careers as other professional groups.
Limitations and research perspectives. The present research has limitations which open perspectives for further investigation. First, our sample was a highly educated one. Future research should test whether the present ndings can be generalized for people with lower human capital (Ng et al., 2005) in terms of education. Second, future research should consider
process variables possibly mediating the inuences of self-efcacy and goals. Third, future research should consider more
individual difference variables. As a recent meta-analysis demonstrated self-efcacy effects, for instance, might be attenuated if personality is taken into account (cf. Judge, Jackson, Shaw, Scott, & Rich, 2007). Fourth, future research should replicate and further analyze the negative impact of career-advancement goals on career satisfaction found here in order to study
whether this is a phenomenon limited to an early career phase or whether it is a more stable inuence. The gender gap in
salary is also worth further examination.
Another research perspective is the study of reciprocal inuences of career success, occupational self-efcacy, and personal occupational goals. It is well conceivable that not only self-efcacy enhances career success, but that also career success enhances occupational self-efcacy. Self-efcacy theory postulates such a cyclical nature of self-efcacy, goals, and goalattainment (Bandura, 1986; see Kammeyer-Mueller, Judge, & Piccolo, 2008, for self-esteem).
Summarizing, the present research shows that socio-cognitive reasoning can be well applied to career-success research.
Since self-efcacy beliefs and personal occupational goals are malleable individual differences these variables are also good
candidates for applied issues of career counseling and training. Enhancement of occupational self-efcacy is generally useful.
Regarding career-advancement goals, however, the distinction between more objective and more subjective career success
should be considered since highly ambitious goals in the realm of career advancement might have both desired and undesired effects.
Appendix A
Factor loadings for latent constructs estimated by the measurement model.
Variable

Factor 1

Factor 2

Factor 3

Occupational self-efcacy at graduation


Career-advancement goals at graduation
Career satisfaction after seven years of professional experience

.73
.91
.83

.82
.65
.90

.74
.76
.69

Note. All factor loadings are signicant on the .001 signicance level.

References
Abele, A. E. (2003). The dynamics of masculine-agentic and feminine-communal traits: Findings from a prospective study. Journal of Personality and Social
Psychology, 85, 768776.
Abele, A. E., & Spurk, D. (2006). Beruiche Laufbahnentwicklung von Akademikerinnen und Akademikern der Universitt Erlangen-Nrnberg (BELA-E).
Vierte Erhebung der prospektiven LngsschnittstudieFragebogen und Grundauswertung [Career development of professionals. Questionnaire and
basic analysis of the fourth wave of measurement] (Tech. Rep. No. 4). Bavaria, Germany: University of Erlangen, Department of Social Psychology.
Abele, A. E., & Stief, M. (2004). Die Prognose des Berufserfolgs von Hochschulabsolventinnen und-absolventen [Predicting Career Success of University
Graduates]. Zeitschrift fr Arbeits-und Organisationspsychologie, 48, 416.
Abele, A. E., Stief, M., & Andr, M. (2000). Zur konomischen Erfassung beruicher Selbstwirksamkeitserwartungen: Neukonstruktion einer BSW-Skala [on
the economic measurement of occupational self-efcacy expectations]. Zeitschrift fr Arbeits-und Organisationspsychologie, 44, 145151.
Armor, D. A., & Taylor, S. E. (1998). Situated optimism: Specic outcome expectancies and self-regulation. In M. P. Zanna (Ed.). Advances in experimental
social psychology (Vol. 30, pp. 309379). New York: Academic Press.
Arnold, J., & Cohen, L. (2008). The psychology of careers in industrial and organizational settings: A critical but appreciative analysis. In G. P. Hodgkinson & J.
K. Ford (Eds.). International review of industrial and organizational psychology (Vol. 23, pp. 144). New York: John Wiley & Sons.
Arthur, M. B., Khapova, S. N., & Wilderom, C. P. M. (2005). Career success in a boundaryless career world. Journal of Organizational Behavior, 26, 177202.
Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure, process, and content. Psychological Bulletin, 120, 338375.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1997). Self-efcacy. The exercise of control. New York: Freeman.
Betz, N. E., & Fitzgerald, L. F. (1987). The career psychology of women. San Diego, CA: Academic Press.

62

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 5362

Betz, N. E., & Hackett, G. (2006). Career self-efcacy: Theory back to the future. Journal of Career Assessment, 14, 311.
Betz, N. E., Borgen, F. H., Rottinghaus, P., Paulsen, A., Halper, C. R., & Harmon, L. W. (2003). The expanded skills condence inventory measuring basic
dimensions of vocational activity. Journal of Vocational Behavior, 62, 76100.
Biernat, M., & Billings, L. S. (2001). Standards, expectancies, and social comparison. In A. Tesser & N. Schwarz (Eds.), Blackwell handbook of social psychology:
intraindividual processes (pp. 257283). Oxford: Blackwell.
Brown, S. P., Jones, E., & Leigh, T. W. (2005). The attenuating effect of role overload on relationships linking self-efcacy and goal level to work performance.
Journal of Applied Psychology, 90, 972979.
Chen, G., Gully, S., & Eden, D. (2001). Validation of a new general self-efcacy scale. Organizational Research Methods, 4, 6283.
Day, R., & Allen, T. D. (2004). The relationship between career motivation and self-efcacy with proteg career success. Journal of Vocational Behavior, 64,
7291.
Dette, E. D., Abele, A. E., & Renner, O. (2004). Zur Denition und Messung von BerufserfolgTheoretische berlegungen und metaanalytische Befunde zum
Zusammenhang von externen und internen Laufbahnerfolgsmaen [Dening and measuring occupational successtheoretical considerations and
meta-analytical ndings on the relationship between external and internal measures]. Zeitschrift fr Personalpsychologie, 3, 170183.
Dwyer, J. H. (1983). Statistical models for the social and behavioral sciences. New York: Oxford University Press.
Frese, M., Garst, H., & Fay, D. (2007). Making things happen: Reciprocal relationships between work characteristics and personal initiative in a four-wave
longitudinal structural equation model. Journal of Applied Psychology, 92, 10841102.
Frieze, I. H., Olson, J. E., Murrell, A. J., & Selvan, M. S. (2006). Work values and their effect on work behavior and work outcomes in female and male
managers. Sex Roles, 54, 8393.
Greene, B. A., & DeBacker, T. K. (2004). Gender and orientations toward the future: Links to motivation. Educational Psychology Review, 16, 91120.
Greenhaus, J. H., Parasuraman, S., & Wormley, W. M. (1990). Effects of race on organizational experiences, job performance evaluations, and career
outcomes. Academy of Management Journal, 33, 6486.
Heslin, P. A. (2005). Conceptualizing and evaluating career success. Journal of Organizational Behavior, 26, 113136.
Higgins, M. C., Dobrow, S. R., & Chandler, D. (2008). Never quite good enough: The paradox of sticky developmental relationships for elite university
graduates. Journal of Vocational Behavior, 72, 207224.
Holland, J. L. (1997). Making vocational choices: A theory of vocational personalities and work environments (3rd ed.). Englewood Cliffs, NJ: Prentice Hall.
Judge, T. A., & Hurst, C. (2007). Capitalizing on ones advantages: Role of core self-evaluations. Journal of Applied Psychology, 92, 12121227.
Judge, T. A., Jackson, C. L., Shaw, J. C., Scott, B. A., & Rich, B. L. (2007). Self-efcacy and work-related performance: The integral role of individual differences.
Journal of Applied Psychology, 92, 107127.
Kammeyer-Mueller, J. D., Judge, T. A., & Piccolo, R. F. (2008). Self-esteem and extrinsic career success: Test of a dynamic model. Applied Psychology: An
International Review, 57, 204224.
Kim, S., Mone, M. A., & Kim, S. (2008). Relationships among self-efcacy, pay-for-performance perceptions, and pay satisfaction: A Korean examination.
Human Performance, 21, 158179.
Kirchmeyer, C. (1998). Determinants of managerial career success: Evidence and explanation of male/female differences. Journal of Management, 24,
673692.
Kline, R. B. (2005). Principles and practice of structural equation modeling. New York: The Guilford Press.
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, 79122.
Little, B. R. (1983). Personal projects: A rationale and a method for investigation. Environment and Behavior, 15, 273309.
Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57,
705717.
Lubbers, R., Loughlin, C., & Zweig, D. (2005). Young workers job self efcacy and affect: Pathways to health and performance. Journal of Vocational Behavior,
67, 199214.
Muthn, L. K., & Muthn, B. O. (1998). Mplus users guide. 3rd ed.. Los Angeles, CA: Muthn & Muthn.
Ng, T. W. H., Eby, L. T., Sorensen, K. L., & Feldman, D. C. (2005). Predictors of objective and subjective career success. A meta-analysis. Personnel Psychology,
58, 367408.
Pajares, F. (1996). Self-efcacy beliefs in academic settings. Review of Educational Measurement, 66, 543578.
Palmore, E. (1978). When can age, period, and cohort be separated? Social Forces, 57, 282295.
Phlmann, K., & Brunstein, J. (1997). Goals: Ein Fragebogen zur Messung von Lebenszielen [Goals: A questionnaire for assessing life goals]. Diagnostica, 43,
6379.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American
Psychologist, 55, 6878.
Saks, A. M. (1995). Longitudinal eld investigation of the moderating and mediating effects of self-efcacy on the relationship between training and
newcomer adjustment. Journal of Applied Psychology, 80, 211225.
Schmuck, P., Kasser, T., & Ryan, R. M. (2000). Intrinsic and extrinsic goals: Their structure and relationship to well-being in German and US college students.
Social Indicators Research, 50, 225241.
Seibert, S. E., Crant, J. M., & Kraimer, M. L. (1999). Proactive personality and career success. Journal of Applied Psychology, 84, 416427.
Seifert, K. H., & Bergmann, C. H. (1983). Deutschsprachige Adaptation des Work Value Inventory von Super [German adaptation of Supers work value
inventory]. Zeitschrift fr Arbeits-und Organisationspsychologie, 27, 160172.
Spurk, D. (2007). The interrelationship of occupational self-efcacy, career goals, and career orientations. Unpublished raw data.
Stajkovic, A. D., & Luthans, F. (1998). Self-efcacy and work-related performance: A meta-analysis. Psychological Bulletin, 124, 240261.
Super, D. E. (1970). Work values inventory. Boston: Houghton.
Valcour, M., & Ladge, J.J. (2008). Family and career path characteristics as predictors of womens objective and subjective career success: Integrating
traditional and protean career explanations. Journal of Vocational Behavior, 73, 300-309.
Winter, D. G. (1991). A motivational model of leadership: Predicting long-term management success from TAT measures of power motivation and
responsibility. Leadership Quarterly, 2, 6780.
Winter, D. G., Riggio, R. E., Murphy, S. E., & Pirozzolo, F. J. (2002). The motivational dimensions of leadership: Power, achievement, and afliation. In Multiple
intelligences and leadership. (pp. 119138). Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers.
Wulff, M., & Steitz, J. (1996). A measure of career self-efcacy. Perceptual and Motor Skills, 82, 240242.
Zytowski, D. G. (1994). A super contribution to vocational theory: Work values. The Career Development Quarterly, 43, 2531.