Discussion 1

Journal of Vocational Behavior 74 (2009) 53–62

Contents lists available at ScienceDirect

Journal of Vocational Behavior

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j v b

The longitudinal impact of self-efficacy 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-efficacy Career-advancement goals Salary Status Career satisfaction Gender Longitudinal study

0001-8791/$ - see front matter � 2008 Elsevier Inc doi:10.1016/j.jvb.2008.10.005

* Corresponding author. Fax: +49 9131 8524731. E-mail address: [email protected] (A.E.

a b s t r a c t

The present research reports on the impact of occupational self-efficacy and of career- advancement goals on objective (salary, status) and subjective (career satisfaction) career attainments. Seven hundred and thirty four highly educated and full-time employed pro- fessionals answered questionnaires immediately after graduation, three years later, and seven years later. Controlling for discipline, GPA at master’s level, and gender, we found that occupational self-efficacy 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 impli- cations 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 one’s career) has been high for many years. One main strand of re- search concerns what predicts success. The present research addresses the influence on career success of two well-known individual difference variables, namely self-efficacy beliefs and personal goals. Self-efficacy beliefs (Bandura, 1986, 1997) and personal goals (Austin & Vancouver, 1996; Little, 1983; Locke & Latham, 2002) are important constructs in socio-cogni- tive models of career interests and performance (Lent, Brown, & Hackett, 1994). There is considerable research on the influ- ence of self-efficacy and of personal goals on task performance as well as on job performance (e.g., Stajkovic & Luthans, 1998). There are also findings on self-efficacy and goals influencing early phases of an individual’s career choice (e.g., Betz & Hackett, 2006). However, there is almost no research on the influence these variables have on career success conceptual- ized as the objective and subjective outcomes an individual receives in his/her career. A recent meta-analysis on determi- nants 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 findings on the impact that self-efficacy beliefs (i.e., occupational self-efficacy) 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).

. All rights reserved.

Abele).

54 A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 53–62

2. Self-efficacy and personal goals

Self-efficacy is defined as individuals’ beliefs about their capability to perform some behavior or to meet a standard. Indi- viduals with high self-efficacy beliefs set higher goals for themselves, put in more effort, and persist longer on a difficult task (Bandura, 1986; Bandura, 1997). Generalized self-efficacy can be distinguished from more domain-specific self-efficacy. Per- sonal goals (Little, 1983) are aims of an action (Locke & Latham, 2002) or internally represented desired states (Austin & Vancouver, 1996). They are assumed to influence outcomes by directing attention, mobilizing effort, affecting persistence, and structuring behavior. They allow long-term orientation and regulation of one’s actions. According to socio-cognitive the- orizing (Brown, Jones, & Leigh, 2005) both self-efficacy beliefs and goals are determinants of successful actions.

3. Career success

Career success is defined 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 findings suggesting that the predictors of objective career success differ from the predictors of subjec- tive 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. Influence of self-efficacy and goals on career success

Day and Allen (2004) reported positive correlations between municipal employees’ career self-efficacy, current salary, and subjective career success (similarly Valcour & Ladge, 2008). Kim, Mone, and Kim (2008) reported that Korean employees’ self-efficacy correlate positively with salary. In contrast, Lubbers, Loughlin, and Zweig (2005) found no association between job self-efficacy and hourly wage. Saks (1995) showed that task-related self-efficacy 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 influence of self-efficacy on salary (Day & Allen, 2004; Kim et al., 2008; Valcour & Ladge, 2008), another study suggests no influence on salary (Lubbers et al., 2005). Two longitudinal studies reveal an influence of self-efficacy on job satisfaction or perceived career success (Hig- gins 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-efficacy and goals.

5. Present research

In the present research we analyzed the longitudinal influence of occupational self-efficacy and of career-advancement goals on objective success as well as on subjective success across seven years. With regard to self-efficacy beliefs we were concerned with occupational self-efficacy. Occupational self-efficacy is the belief in one’s capacity and motivation to success- fully perform occupational tasks and challenges and to pursue one’s occupational career irrespective of the particular field of occupation (Higgins et al., 2008). Occupational self-efficacy is neither a broad measure of generalized self-efficacy nor a very specific measure of particular career interests self-efficacy such as, for instance, occupational confidence themes according to Holland’s (1997) RIASEC model (for instance, Betz et al., 2003; Wulff & Steitz, 1996). Occupational self-efficacy rather has an intermediate level of specificity. We chose such a level because research has shown that a medium level of specificity is advantageous in predicting specific 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 influence, 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., Pöhlmann & 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-efficacy and career-advancement goals. We measured objective career success after 36 months of professional experience and again after 85 months of professional experience. Career sat- isfaction was measured once after 85 months of professional experience.

Fig. 1. Theoretical model.

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 53–62 55

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-efficacy and career-advancement goals associate positively. Individ- uals with high career-advancement goals should also have high occupational self-efficacy beliefs. However, individuals with high occupational self-efficacy 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-efficacy and career-advancement goals correlate positively.

In accord with meta-analytical findings (Dette et al., 2004; Ng et al., 2005) we assume that the two objective success indi- cators and the subjective success indicator associate positively.

Hypothesis 2. Salary, hierarchical status, and career satisfaction correlate positively.

Hypotheses 3–5 concern the impact of the predictors on the career success measures. We predicted that occupational self-efficacy (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 influence of predictors at two times of mea- surement (i.e., salary and salary change; hierarchical status and status change) we predicted that occupational self-efficacy and career-advancement goals would not only influence salary, but also changes in salary.

Hypothesis 3. Occupational self-efficacy and career-advancement goals both positively influence 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 influence hierarchical status and changes in status.

We tested the influence of occupational self-efficacy on status and status change in an exploratory fashion. A positive influence is conceivable because of the general motivational impact self-efficacy has on performance and on desired out- comes. There could, however, also be no influence because occupational self-efficacy need not be accompanied by status goals.

Derived from previous research (Day & Allen, 2004; Higgins et al., 2008; Saks, 1995) and also derived from findings in the field of optimistic expectations and their influence on outcome evaluations (e.g., Armor & Taylor, 1998) we predicted that occupational self-efficacy has a positive influence on career satisfaction.

Hypothesis 5. Occupational self-efficacy positively influences career satisfaction.

The influence of career-advancement goals on career satisfaction was again be tested in an exploratory fashion. There could be no influence at all, because career-satisfaction should be due to goal-fulfillment and not to the degree of career- advancement goals. A negative influence is also conceivable such that individuals with high career-advancement goals

56 A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 53–62

are less easily satisfied 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 financial 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 findings. First, we included only participants who worked full-time, because it has been shown that working hours have a strong influence on, for instance, salary or promotions (Ng et al., 2005). Second, we controlled for the occupational field 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’ master’s degree Grade Point Average (GPA), because this also might have an influence on career success (Ng et al., 2005).

Summarizing, our model suggests that—controlling for gender, GPA, and discipline—occupational self-efficacy and career- advancement goals influence career success outcomes, and that these influences remain significant if both variables are con- sidered 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 (Muthén & Muthén, 1998). Structural equation modeling has several advantages. The measurement model of the predictors can be included; measurement errors can be taken into account; the specific postulated paths can be tested; and besides pro- viding the path coefficients, a series of overall fit statistics shows how well the empirical data fit 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 master’s degree. Two cohorts of graduates (consecutive graduation years) completed the first questionnaire some weeks after they had passed their final exams. Fur- ther measures were taken after one-and-a-half years, after three years, and after seven years. We did not find 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 first questionnaire ourselves. Instead the univer- sity’s graduation office 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 five per- cent 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-efficacy, 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 first 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 pro- fessions and teachers have to undergo an obligatory 18 months post-gradual training during which they earn a fixed—low— 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-efficacy, 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 partic- ipants 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) 53–62 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 influence of self-efficacy beliefs and personal goals on career success—mediated via working hours—is studied (we nevertheless tested our hypotheses also in the sample in which part-time employed participants were included (N = 971). The findings [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 GPA’s using the average of all individuals who had passed their master’s 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-efficacy. The occupational self-efficacy scale (Abele, Stief, & Andrä, 2000) consists of 6 items (sample item ‘‘I am confident that I could deal efficiently with the challenges of my occupation if I only wanted to”). Participants re- sponded 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 Cronbach’s a = .78. The occupational self-efficacy 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-efficacy in working popula- tions 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 five 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, Cronbach’s 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 combin- ing information on three variables: permission to delegate work (0 = no, 1 = yes), project responsibility (0 = no, 1 = yes), and official 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 (Green- haus, Parasuraman, & Wormley, 1990). The scale comprises five items (sample item: ‘‘I am satisfied with the progress I have made towards meeting my overall career goals”). Participants responded on five-point scales (1 = not at all to 5 = very much). The scale is one-dimensional (61% explained item variance) and revealed good internal consistency, Cronbach’s a = .83.

Analytical strategy. We tested our hypotheses by means of structural equation modeling (SEM) using Mplus (Muthén & Muthén, 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 findings

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.

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

Variable M SD 1 2 3 4 5 6 7

1 GPA .07 .51 2 Occupational SEF at graduationa 3.80 .70 .07 3 Career-advancement goals at graduationa 3.23 .71 .04 .22 4 Salary after three years of professional experienceb 4.60 1.02 .03 .17 .17 5 Salary after seven years of professional experienceb 6.21 1.78 .07 .18 .16 .48 6 Status after three years of professional experiencec 1.16 .97 .01 .14 .19 .29 .23 7 Status after seven years of professional experiencec 1.71 1.14 .03 .12 .20 .27 .37 .49 8 Career satisfaction after seven years of professional experiencea 3.67 .70 .00 .19 -.03 .08 .31 .05 .18

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-efficacy.

58 A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 53–62

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-efficacy (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 first modeled the measurement of occupational self-efficacy (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 fit 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 (dum- my 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-efficacy, career-advancement goals, and career satisfaction. We tested paths from the control variables, e.g., participants’ gender, GPA, and discipline, to occupational self-efficacy, 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 occupa- tional self-efficacy and career-advancement goals and the residual correlations between salary, hierarchical status, and ca- reer satisfaction. Finally, we tested paths from occupational self-efficacy and career-advancement goals to salary, status, and to career satisfaction.

The resulting model has good fit statistics (v2 = 229.50, df = 96, N = 734, CFI = .96, TLI = .93, RMSEA = .04, SRMR = .02). The R2 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 first present the findings on the influence of the control variables, e.g., gender, GPA, and discipline because for clarity’s sake we did not include the respective paths into the graphical depiction of the model. Table 2 shows the paths discipline or professional field (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-efficacy did barely differ between disciplines

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

Occupational self-efficacy at graduation

Career- advancement 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

Discipline Law .04 .24*** �.06 .05 .30*** .30*** �.03 Medicine .11 .28*** .25*** .13** .42*** .45*** �.10 Arts & Humanities .05 .03 .01 .04 .25*** .26*** �.08 Science .01 .13* .03 .09** .19*** .25*** �.02 Economics .09 .38*** .33*** .26*** .42*** .46*** �.10 Engineering .17* .29*** .19*** .20*** .38*** .44*** �.11 Gender .09* �.05 .16*** .11*** .07 .03 .02 GPA .09* .06 .02 .06 �.01 .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) 53–62 59

(only participants in the field of engineering had slightly higher means than teachers). Participants working in the fields 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 fields of medicine, economics, and engineering had higher sal- aries 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 influence on occupational self-efficacy. There were no gender influences on sta- tus 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-significant paths. Supporting Hypothesis 1 the residual correlation of occupational self-efficacy and career-advancement goals was positive (r = .26, p < .001). Support- ing 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-efficacy and of career-advancement goals on salary was fully supported for occupational self-efficacy. Participants high in occupational self-efficacy 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-efficacy at career entry. Hypothesis 3 was partly supported for career-advancement goals. Participants with high career- advancement 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 influence 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 influence of occupational self-efficacy 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-efficacy had a positive influence on career satisfaction (b = .26, p < .001). The exploratory test of the influence of career-advancement goals on career satisfaction resulted in a significantly negative im- pact (b = �.10, p < .01). The higher our participants’ career-advancement goals had been at graduation, the less satisfied they were with their career seven years later.

Since GPA had an influence on occupational self-efficacy (see Table 2) we also tested whether there might be indirect influences of GPA on career outcomes mediated via occupational self-efficacy. We computed significance tests for indirect effects (cf. Kline, 2005) and found only one indirect effect of GPA on career satisfaction mediated via occupational self-effi- cacy (product of the two involved beta-coefficients: .023, p < .05). The test of indirect influences of gender (mediated via occupational self-efficacy) revealed no significant results.

Fig. 2. Structural equation model for salary, hierarchical status, and career satisfaction. Only significant 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) 53–62

8. Discussion

Hypotheses testing. Supporting Hypothesis 1 occupational self-efficacy and career-advancement goals correlated moder- ately. 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 re- ported 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-efficacy 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-efficacy and career advancement goals had been at career entry, the more they earned and the higher was their status later on. The findings 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-efficacy 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 per- centages of explained variance. There are many influences on salary and status, and individual differences in socio-cognitive variables are but one source of influence. Apart from that we controlled for important sources of influence 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 occupa- tional self-efficacy and of career-advancement goals on objective career attainments is of both theoretical and applied rel- evance. Furthermore, beta effects between .07 and .11 are absolutely in line with previous findings 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. Partici- pants with higher occupational self-efficacy at graduation were more satisfied with their careers seven years later than those with lower occupational self-efficacy. This variable added 5% of variance to the prediction of success above the included con- trol measures. Replicating meta-analytical findings (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 influence 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 satisfied with their careers than individuals with lower career-advancement goals. This is a novel and provocative finding. One interpretation could rely on an individual’s 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 satisfied 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 individual’s objective success, but not his/her career satisfaction.

Gender. In our specific 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-efficacy than men. This is in line with prior research (e.g., Betz & Fitzgerald, 1987).

An important finding 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 findings are often not controlled for field of occupation (i.e., whether men and women work in fields 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 field of occu- pation and working time. But even though we only included participants working full-time and even though we controlled for their field 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 significant three years after work entry and it remained highly significant after seven years of professional experience. Women’s somewhat lower occupational self-effi- cacy did not mediate this effect. Interestingly, gender had no influence on status. This means that in the present sample wo- men’s 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-efficacy; it is not due to the field of occupation; and it also seems not to be due to the duties men and women fulfill 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

A.E. Abele, D. Spurk / Journal of Vocational Behavior 74 (2009) 53–62 61

careers than men, women may apply lower standards for their careers than for men’s careers. Therefore despite of lower salary they are similarly satisfied with their careers as men.

GPA and discipline. Even if findings 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 influence of GPA on career success was more or less negligible. Second, the objective career attainments studied here were generally high in fields 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 aspi- rations individuals will be satisfied 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 satisfied with their careers as other professional groups.

Limitations and research perspectives. The present research has limitations which open perspectives for further investiga- tion. First, our sample was a highly educated one. Future research should test whether the present findings can be general- ized for people with lower ‘‘human capital” (Ng et al., 2005) in terms of education. Second, future research should consider process variables possibly mediating the influences of self-efficacy and goals. Third, future research should consider more individual difference variables. As a recent meta-analysis demonstrated self-efficacy effects, for instance, might be attenu- ated if personality is taken into account (cf. Judge, Jackson, Shaw, Scott, & Rich, 2007). Fourth, future research should repli- cate 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 influence. The gender gap in salary is also worth further examination.

Another research perspective is the study of reciprocal influences of career success, occupational self-efficacy, and per- sonal occupational goals. It is well conceivable that not only self-efficacy enhances career success, but that also career suc- cess enhances occupational self-efficacy. Self-efficacy theory postulates such a cyclical nature of self-efficacy, goals, and goal- attainment (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-efficacy 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-efficacy 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 unde- sired effects.

Appendix A

Factor loadings for latent constructs estimated by the measurement model.

Variable

Factor 1

Factor 2

Factor 3

Occupational self-efficacy at graduation

.73

.82

.74

Career-advancement goals at graduation

.91

.65

.76

Career satisfaction after seven years of professional experience

.83

.90

.69

Note. All factor loadings are significant on the .001 significance 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, 768–776.

Abele, A. E., & Spurk, D. (2006). Berufliche Laufbahnentwicklung von Akademikerinnen und Akademikern der Universität Erlangen-Nürnberg (BELA-E). Vierte Erhebung der prospektiven Längsschnittstudie—Fragebogen 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 für Arbeits-und Organisationspsychologie, 48, 4–16.

Abele, A. E., Stief, M., & Andrä, M. (2000). Zur ökonomischen Erfassung beruflicher Selbstwirksamkeitserwartungen: Neukonstruktion einer BSW-Skala [on the economic measurement of occupational self-efficacy expectations]. Zeitschrift für Arbeits-und Organisationspsychologie, 44, 145–151.

Armor, D. A., & Taylor, S. E. (1998). Situated optimism: Specific outcome expectancies and self-regulation. In M. P. Zanna (Ed.). Advances in experimental social psychology (Vol. 30, pp. 309–379). 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. 1–44). 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, 177–202. Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure, process, and content. Psychological Bulletin, 120, 338–375. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1997). Self-efficacy. 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) 53–62

Betz, N. E., & Hackett, G. (2006). Career self-efficacy: Theory back to the future. Journal of Career Assessment, 14, 3–11. Betz, N. E., Borgen, F. H., Rottinghaus, P., Paulsen, A., Halper, C. R., & Harmon, L. W. (2003). The expanded skills confidence inventory measuring basic

dimensions of vocational activity. Journal of Vocational Behavior, 62, 76–100. 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. 257–283). Oxford: Blackwell. Brown, S. P., Jones, E., & Leigh, T. W. (2005). The attenuating effect of role overload on relationships linking self-efficacy and goal level to work performance.

Journal of Applied Psychology, 90, 972–979. Chen, G., Gully, S., & Eden, D. (2001). Validation of a new general self-efficacy scale. Organizational Research Methods, 4, 62–83. Day, R., & Allen, T. D. (2004). The relationship between career motivation and self-efficacy with protegé career success. Journal of Vocational Behavior, 64,

72–91. Dette, E. D., Abele, A. E., & Renner, O. (2004). Zur Definition und Messung von Berufserfolg—Theoretische Überlegungen und metaanalytische Befunde zum

Zusammenhang von externen und internen Laufbahnerfolgsmaßen [Defining and measuring occupational success—theoretical considerations and meta-analytical findings on the relationship between external and internal measures]. Zeitschrift für Personalpsychologie, 3, 170–183.

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, 1084–1102. 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, 83–93. Greene, B. A., & DeBacker, T. K. (2004). Gender and orientations toward the future: Links to motivation. Educational Psychology Review, 16, 91–120. 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, 64–86. Heslin, P. A. (2005). Conceptualizing and evaluating career success. Journal of Organizational Behavior, 26, 113–136. 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, 207–224. 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 one’s advantages: Role of core self-evaluations. Journal of Applied Psychology, 92, 1212–1227. Judge, T. A., Jackson, C. L., Shaw, J. C., Scott, B. A., & Rich, B. L. (2007). Self-efficacy and work-related performance: The integral role of individual differences.

Journal of Applied Psychology, 92, 107–127. 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, 204–224. Kim, S., Mone, M. A., & Kim, S. (2008). Relationships among self-efficacy, pay-for-performance perceptions, and pay satisfaction: A Korean examination.

Human Performance, 21, 158–179. Kirchmeyer, C. (1998). Determinants of managerial career success: Evidence and explanation of male/female differences. Journal of Management, 24,

673–692. 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, 79–122. Little, B. R. (1983). Personal projects: A rationale and a method for investigation. Environment and Behavior, 15, 273–309. 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,

705–717. Lubbers, R., Loughlin, C., & Zweig, D. (2005). Young workers job self efficacy and affect: Pathways to health and performance. Journal of Vocational Behavior,

67, 199–214. Muthén, L. K., & Muthén, B. O. (1998). Mplus useŕs guide. 3rd ed.. Los Angeles, CA: Muthén & Muthén. 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, 367–408. Pajares, F. (1996). Self-efficacy beliefs in academic settings. Review of Educational Measurement, 66, 543–578. Palmore, E. (1978). When can age, period, and cohort be separated? Social Forces, 57, 282–295. Pöhlmann, K., & Brunstein, J. (1997). Goals: Ein Fragebogen zur Messung von Lebenszielen [Goals: A questionnaire for assessing life goals]. Diagnostica, 43,

63–79. Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American

Psychologist, 55, 68–78. Saks, A. M. (1995). Longitudinal field investigation of the moderating and mediating effects of self-efficacy on the relationship between training and

newcomer adjustment. Journal of Applied Psychology, 80, 211–225. 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, 225–241. Seibert, S. E., Crant, J. M., & Kraimer, M. L. (1999). Proactive personality and career success. Journal of Applied Psychology, 84, 416–427. Seifert, K. H., & Bergmann, C. H. (1983). Deutschsprachige Adaptation des Work Value Inventory von Super [German adaptation of Super’s work value

inventory]. Zeitschrift für Arbeits-und Organisationspsychologie, 27, 160–172. Spurk, D. (2007). The interrelationship of occupational self-efficacy, career goals, and career orientations. Unpublished raw data. Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A meta-analysis. Psychological Bulletin, 124, 240–261. Super, D. E. (1970). Work values inventory. Boston: Houghton. Valcour, M., & Ladge, J.J. (2008). Family and career path characteristics as predictors of women’s 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, 67–80. Winter, D. G., Riggio, R. E., Murphy, S. E., & Pirozzolo, F. J. (2002). The motivational dimensions of leadership: Power, achievement, and affiliation. In Multiple

intelligences and leadership. (pp. 119–138). Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers. Wulff, M., & Steitz, J. (1996). A measure of career self-efficacy. Perceptual and Motor Skills, 82, 240–242. Zytowski, D. G. (1994). A super contribution to vocational theory: Work values. The Career Development Quarterly, 43, 25–31.