Journal of Vocational Behavior 60, 321–335 (2002)

doi:10.1006/jvbe.2001.1835, available online at http://www.idealibrary.com on

Subjective Career Success and Emotional Well-Being:

Longitudinal Predictive Power of Selection,
Optimization, and Compensation

Bettina S. Wiese, Alexandra M. Freund, and Paul B. Baltes

Max Planck Institute for Human Development, Berlin, Germany

In a 3-year longitudinal study, we found in a sample of young professionals (N = 82;

44% male; age range: 28 to 39 years) that self-reported behaviors reflecting selection,
optimization, and compensation (SOC) predicted global and work-specific subjective well-
being (multiple correlations ranged from R = .22 to R = .44). In addition to optimization
(i.e., implementing goal-relevant means), it was especially the degree of compensation
(i.e., investing goal-relevant means to counteract losses) that predicted how emotionally
balanced individuals felt and how satisfied they were with their work situation 3 years
later. These longitudinal predictions were quite robust when controlling for personality
variables (NEO). Results are consistent with previous cross-sectional findings and demon-
strate how the SOC framework might be successfully applied to the domain of vocational
behavior. C 2002 Elsevier Science (USA)
Key Words: selection; optimization; compensation; goal striving; life management; well-
being; job satisfaction; personality.

In current Western societies, many people view high achievement in the work
domain as a key factor for successful life management. To foster career devel-
opment, processes of lifelong learning and the development of problem-solving
strategies are considered to be of prime importance (Hall & Mirvis, 1995; Super,
1980). This is particularly true in times where advancing globalization and fast
technical progress quickly change professional affordances (Fay & Frese, 2000;
Howard, 1995). The investigation of effective self-regulatory and proactive strate-
gies to support career development (e.g., Kanfer & Heggestad, 1999; Seibert,
Crant, & Kraimer, 1999), therefore, seems to be a promising research field for the
next decades.
The present study represents the longitudinal extension of a previously pub-
lished cross-sectional study (Wiese, Freund, & Baltes, 2000). In that earlier study,
positive cross-sectional associations were found between strategies of proactive
life management and success in work and partnership. For a subsample of this

Bettina S. Wiese is now at the Darmstadt University of Technology.

Address correspondence and reprint requests to Bettina S. Wiese, Darmstadt University of Techno-
logy, Institute of Psychology, Hochschulstrasse 1, D-64289 Darmstadt, Germany. Fax: +49-61 51-16
68 11. E-mail: wiese@psychologie.tu-darmstadt.de.

321
0001-8791/02 $35.00

C 2002 Elsevier Science (USA)
All rights reserved.
322 WIESE, FREUND, AND BALTES

cross-sectional investigation, the present study tested the longitudinal prediction

of this association. The key outcome, subjective career success, was measured
3 years later. The next section briefly describes the theory of adaptive develop-
ment, SOC, on which this study is based.

SELECTION, OPTIMIZATION, AND COMPENSATION:

AN APPLICATION TO THE CAREER DOMAIN
Taking a life-span approach, the SOC model proposes that there are three funda-
mental processes contributing to successful life management: selection, optimiza-
tion, and compensation (SOC; M. Baltes & Carstensen, 1996, 1998; P. Baltes,
1997; P. Baltes & Baltes, 1990; Freund & Baltes, 2000; Marsiske, Lang, Baltes,
& Baltes, 1995; Wiese, 2000; Wiese et al., 2000).
Embedded in an action-theoretical framework that stresses the active role of
the individual for shaping his or her life, the SOC model postulates that selection
(i.e., developing, elaborating, and committing oneself to personal goals) provides
the basis for focusing resources on a delineated number of life domains in which
to achieve one’s goals (optimization; i.e., the acquisition, refinement, and appli-
cation of goal-relevant skills or resources) or in which to maintain one’s goals
in the face of losses in goal-relevant means (i.e., compensation). As Savickas
(2001) elaborated, the SOC model might be especially useful in the domain of
career development by presenting a conceptual framework of how choices are
made and implemented. Super (1980; Super, Savickas, & Super, 1996), for in-
stance, pointed out that recognizing decision questions, exploring possibilities for
improving one’s occupational situation, committing oneself to action plans, and
evaluating outcomes to modify or maintain career plans are central strategies for
fostering successful career development.
Recently, another attempt to apply the SOC framework to work settings has been
undertaken by B. Baltes and Dickson (2001). They demonstrated the applicability
of SOC in the domains of organizational leadership and work–family conflict. As
one example of how SOC can be applied to the domain of organizational lead-
ership, B. Baltes and Dickson (2001) pointed out that—in the face of resource
restrictions—successful managers clearly select and reward the most valued sub-
ordinates, which can be interpreted as a way of selective optimization of working
power. Moreover, they argued that the SOC model is applicable on the individual
and the organizational levels. In a situation of changing market affordances, for
instance, new organizational and marketing means have to be found to achieve
selected organizational goals.
Similarly, concerning research on work–family conflict (for a summary see
Kossek & Ozeki, 1998), there is a lack of models focusing on how individuals
successfully manage these conflicts in everyday life. As the conflicts between the
work and the family domain are mainly resource-based (Wiese & Freund, 2000),
using behaviors that are geared toward focusing limited resources on selected
goals should promote the successful management of work–family conflicts. Sup-
porting this view, Wiese (2000) found that (so-far-childless) young adults, who
 LONGITUDINAL PREDICTIVE POWER OF SOC 323

reported using the life-management strategies of SOC in the partnership domain,

experienced lower work–family conflict.

Empirical Findings on Self-Reported SOC

There is empirical evidence that self-reported SOC strategies relate to success-
ful life management. In three questionnaire studies comprising a total age range
of 14 to 103 years, Freund and Baltes (1998, 2001) demonstrated not only that
self-reported SOC behaviors showed meaningful convergent and discriminant rela-
tions to other variables (e.g., tenacious vs flexible goal adjustment, Brandtstädter &
Renner, 1990; action vs state orientation, Kuhl & Beckmann, 1994; “Big Five” per-
sonality traits, Costa & McCrae, 1985; thinking styles, Sternberg, 1994), but also
that the SOC strategies related positively to subjective indicators of successful life
management (e.g., life satisfaction, positive emotions, and absence of loneliness).
In addition, SOC exhibited a fair degree of uniqueness in predicting well-being
when controlling for alternative personality and goal-related variables (Freund &
Baltes, 2001). Concerning age, Freund and Baltes (2001) reported that middle-
aged adults (43–67 years) showed higher endorsement of SOC than younger
(18–43 years) or older adults (67–89 years). Abraham and Hanssson (1995) ap-
plied the SOC framework to the occupational domain. In a sample of employed
adults aged 40 to 69 years, they found a positive correlation between use of SOC
strategies and subjective ratings of competence maintenance and goal attainment
in the work situation.
In addition, there is evidence for the functionality of SOC that can be derived
from other work in which SOC-like behaviors are studied: e.g., research on the
functionality of setting specific goals (e.g., Locke, Shaw, Saari, & Latham, 1981),
on the positive career effects of showing personal initiative (i.e., being self-starting,
proactive, and persistent; Fay & Frese, 2001), on modeling others as an effective
learning strategy (Bandura, 1986), and on the ability to delay gratification (Mischel,
1996).

Objective of Study
In a cross-sectional design with a sample of N = 206 young adults (25 to
36 years), we found in past research that self-reported SOC behaviors related
positively to well-being (Wiese et al., 2000). In that study, SOC was assessed by
self-report using a general and two domain-specific (work, partnership) versions
of the SOC questionnaire (P. Baltes, Baltes, Lang, & Freund, 1999). Individuals
reporting SOC behaviors scored higher on multiple subjective indicators of global
and work-related as well as partnership-related success. More specifically, opti-
mization and compensation evinced positive relations with general and domain-
specific well-being, whereas selection turned out to be of minor importance. These
findings were quite robust when controlling for rival predictors, including “Big
Five” traits and control beliefs.
Because of the cross-sectional nature of that study, however, it was impossi-
ble to determine causal relationships: Do SOC-related behaviors lead to higher
324 WIESE, FREUND, AND BALTES

well-being or does well-being influence the strategies of goal-selection and goal-

pursuit captured in selection, optimization, and compensation? To address this
question, we conducted a 3-year follow-up of this study. The central aim of
this follow-up was to test whether the self-reported use of SOC predicts general
well-being as well as satisfaction and subjective attainment in the work domain
over an interval of 3 years. Furthermore, we investigated whether the positive
association of SOC with subjective criteria of successful life outcomes is also
longitudinally robust when controlling for other predictors such as personality
traits.

METHOD
Participants and Procedures
In our first cross-sectional questionnaire study 206 so-far-childless, employed
young adults took part (Wiese et al., 2000). To 180 of these participants, we
sent a questionnaire by mail 3 years later (the others had not be contacted, either
because they had filled out T1 instruments anonymously or because they were
already known to have relocated to an unknown address since we contacted a
subsample of T1 participants for another laboratory experiment; Wiese & Freund,
2001). Eighty-two returned completed questionnaires. Again, having relocated to
an unknown address had been the most prevalent reason for not participating in
the longitudinal study. This is not surprising because the sample comprised young
adults who, for occupational as well as for family reasons, often relocate.
The follow-up sample comprised 36 males and 44 females, ranging in age from
28 to 39 years (M = 33 years, SD = 3 years). Sixty-eight of them were employed
outside the home (working as physicians, lawyers, scientists, bank employees,
hotel managers, or policemen) and five persons were self-employed. Those who
were not working were either on maternal leave (8 females) or currently unem-
ployed (1 male). Overall, the educational level of the sample was rather high, with
63.4% holding an advanced college degree. Sixty-five participants reported having
a partner. Fourteen participants had become parents since the first measurement
occasion 3 years ago.
When comparing participants of the follow-up study with those who dropped
out, there were no differences on measures such as neuroticism, extraversion,
openness to experience, conscientiousness, social desirability, and educational
level (measures are described below or/and in Wiese, 2000). Participants of the
follow-up study, however, had higher scores on agreeableness (M = 5.34, SD =
0.86) than drop-outs [M = 5.10, SD = 0.86; t(204) = −2.10, p < .05]. In
addition, the portion of females taking part in the follow-up was significantly higher
than in the cross-sectional study [χ 2 (1, n = 206) = 4.21, p < .05]. Concerning
self-reported SOC at T1, participants of the follow-up study had higher scores on
selection (M = 1.45, SD = 0.80) than drop-outs [M = 1.20, SD = 0.80; t(204) =
2.19, p < .05]. There were no significant group differences, however, concerning
optimization [t(204) = 1.04, ns] and compensation [t(204) = 0.15, ns].
 LONGITUDINAL PREDICTIVE POWER OF SOC 325

Measures
We first describe instruments from the first time of measurement (T1; see also
Wiese, 2000; Wiese et al., 2000), followed by the instruments used at the second
time of measurement (T2).
Measurement point 1 (T1). SOC was assessed using a short version of a ques-
tionnaire developed by P. Baltes et al. (1999). Each SOC component was assessed
by three items. Each item consists of two alternative statements, one reflecting se-
lection (e.g., “I concentrate all my energy on a few things”), optimization (e.g., “I
keep working on what I have planned until I succeed”), or compensation (“When
things don’t go as well as they used to, I keep trying other ways until I can achieve
the same results as I used to”), the other statement reflecting an alternative behav-
ioral strategy (e.g., “I divide my energy among many things” as a statement not
describing selection). These statements were framed as self-descriptions of two
fictitious persons. Participants were asked to indicate first to which person they
believed themselves to be more similar. Subsequently, they indicated the degree of
similarity on a 4-point scale. For the analyses, a 5-point score (0–4) was computed
for each item, allocating a value of zero to choices of statements not reflecting SOC.
Then, mean scores for selection (M = 1.16, SD = 1.04, α = .56), optimization
(M = 2.27, SD = 0.96, α = .52), and compensation (M = 2.39, SD = 0.83,
α = .47) were built.
As one indicator of successful life management, emotional balance, reflecting
a central affective component of well-being, was assessed using a German short
version of the Positive and Negative Affect Schedule (PANAS; Watson, Clark, &
Tellegen, 1988). This scale consists of 10 adjectives denoting positive (e.g., en-
thusiastic) and 10 items denoting negative affect (e.g., afraid). Participants were
asked to indicate on a 5-point scale how often they had experienced each of the
emotions during the past year. To index emotional balance, mean negative af-
fect was subtracted from mean positive affect (M = 1.20, SD = 0.61, α =
.70).
Successful life management in the occupational domain was assessed using
two instruments: (1) a three-item, 7-point Likert-type scale reflecting subjective
success in the work domain: Did you come closer to achieving your goals in
the occupational domain this past year? How successful have you been so far
in pursuing your occupational goals? and How satisfied are you with your life
management in the occupational domain?, M = 4.99, SD = 1.27, α = .79); and
(2) a job-satisfaction scale (Giegler, 1985) consisting of 10 items (7-point rating
scale; M = 5.12, SD = 1.12, α = .89).
As control variables we included the “Big Five” personality traits because they
are known to have substantial links to various indicators of successful life manage-
ment. With a short version of the NEO-FFI (6 items per scale; 7-point rating; Costa
& McCrae, 1985) we assessed neuroticism (M = 3.27, SD = 0.96, α = .63), ex-
traversion (M = 4.93, SD = 0.95, α = .63), openness to experience (M = 4.78,
SD = 0.75, α = .26), agreeableness (M = 5.20, SD = 0.88, α = .63), and
conscientiousness (M = 5.00, SD = 0.92, α = .68).
326 WIESE, FREUND, AND BALTES

Measurement point 2 (T2): Three years later. To assess selection (M = 1.19,

SD = 1.03, α = .58), optimization (M = 2.38, SD = 0.99, α = .72), and
compensation (M = 2.31, SD = 0.86, α = .55), we again used the SOC ques-
tionnaire. As in the cross-sectional study, emotional balance (M = 1.10, SD =
0.80, α = .84) was measured by the PANAS. Successful life management in the
occupational domain was assessed with the scale developed for measuring subjec-
tive success in the work domain (M = 5.04, SD = 1.50, α = .89) and by Giegler’s
(1985) job-satisfaction scale (M = 5.36, SD = 1.17, α = .92). Concerning the
subjective success ratings, participants were asked to think about the time between
the last and the present questionnaire study.

RESULTS
As the cross-sectional positive relations between SOC and general, as well as
work-related, well-being are described in detail in Wiese et al. (2000; see also
Wiese, 2000), we focus only on the longitudinal relations. Let us first address the
longitudinal correlations of the measures used in the current study (see Table 1).
Correlational stability is of moderate size (ranging between .42 and .58) for both
the measures of SOC and the measures of subjective well-being or success in the
work domain. In light of the small number of items, however, this moderate size is
acceptable. Mean-level comparisons revealed no significant decreases or increases
over the 3-year period (see Table 1).
Turning to the longitudinal correlations between SOC and outcomes (see
Table 2), SOC at T1 predicts emotional balance, subjective success, and satisfac-
tion in the work domain. More specifically, emotional balance at T2, as well as job
satisfaction, could be predicted by optimization and compensation, whereas sub-
jective success in the work domain was solely predicted by compensation. When
entering all three of the SOC variables in multiple-regression analyses, SOC vari-
ables measured at T1 explained 18% of the variance of emotional balance at T2
(multiple R = .44, p < .01, one-tailed), 8% of the variance of job satisfaction at
T2 (multiple R = .29, p < .05, one-tailed), but no significant amount of variance
in work-related subjective success ratings at T2 (5%; multiple R = .22, ns).

TABLE 1
Longitudinal Correlations within Variables over 3 Years

Time 1 Time 2

Variable rt1/t2 M SD M SD t test

Selection .42∗∗ 1.34 1.03 1.19 1.03 t(81) = 1.21, ns

Optimization .58∗∗ 2.36 1.02 2.38 0.99 t(81) = −0.20, ns
Compensation .51∗∗ 2.34 0.80 2.31 0.86 t(81) = 0.38, ns
Emotional Balance .47∗∗ 1.26 0.86 1.12 0.76 t(80) = 1.56, ns
Subjective Work Success .42∗∗ 4.99 1.38 5.04 1.50 t(81) = −0.29, ns
Job Satisfaction .57∗∗ 5.38 1.10 5.36 1.17 t(81) = 0.16, ns
∗∗ p < .01.
 LONGITUDINAL PREDICTIVE POWER OF SOC 327

TABLE 2
Longitudinal Correlations between SOC (T1) and Indicators of Subjective Well-Being (T2)

SOC (T1)
Multiple
Well-Being (T2) Selection Optimization Compensation R

Emotional Balance .01 .22∗∗ .43∗∗ .44∗∗

Subjective Success in the Work Domain .01 .13 .22** .22
Job Satisfaction −.08 .21∗ .25∗∗ .29∗
∗p < .05, one-tailed.
∗∗ p > .01, one-tailed.

To analyze whether SOC at T1 predicts the change in subjective indicators

of successful life management, another set of multiple-regression analyses were
computed (see Tables 3 to 5). In these analyses, we controlled for the stable aspects
of the outcomes (autocorrelation T1/T2) and entered SOC in a second step. As
shown in Table 3, the SOC variables measured at T1 significantly contributed to the
prediction of emotional balance even when entered in the regression analysis after
emotional balance measured at T1 ( R 2 = .12, p < .01, one-tailed, Model 1). An
additional significant portion of variance in emotional balance could be explained
by subsequently entering SOC measured at T2 ( R 2 = .14, p < .01, one-tailed,
Model 1). Note that even when controlling for the autocorrelation and the T2
assessment of SOC strategies, SOC measured at T1 added unique variance to the
longitudinal prediction of well-being (see Table 3, R 2 = .06, p < .05, one-
tailed; Model 2).
With regard to job satisfaction and subjective success ratings in the work do-
main, however, entering SOC measured at T1 did not significantly contribute to

TABLE 3
Hierarchical Multiple Regression Analyses: Predictions from Selection, Optimization,
and Compensation on Emotional Well-Being

Emotional Balance (Time 2)

R R 2

Model 1
Step 1: Emotional Balance (Time 1) .47∗∗
Step 2: SOC (Time 1) .59∗∗ .12∗∗
Step 3: SOC (Time 2) .70∗∗ .14∗∗
Model 2
Step 1: Emotional Balance (Time 1) .47∗∗
Step 2: SOC (Time 2) .66∗∗ .21∗∗
Step 3: SOC (Time 1) .70∗∗ .06∗∗
∗p < .05, one-tailed.
∗∗ p < .01, one-tailed.
328 WIESE, FREUND, AND BALTES

TABLE 4
Hierarchical Multiple Regression Analyses: Predictions from Selection, Optimization,
and Compensation on Subjective Success in the Work Domain

Subjective Success in the

Work Domain (Time 2)

R R 2

Model 1
Step 1: Subjective Success in the Work Domain (Time 1) .43∗∗
Step 2: SOC (Time 1) .47∗∗ .14∗∗
Step 3: SOC (Time 2) .60∗∗ .03
Model 2
Step 1: Subjective Success in the Work Domain (Time 1) .43∗∗
Step 2: SOC (Time 2) .58∗∗ .16∗∗
Step 3: SOC (Time 1) .60∗∗ .02
∗p < .05, one-tailed.
∗∗ p < .01, one-tailed.

the prediction of work-related well-being over and above the contribution of the
domain-specific well-being autocorrelations and its covariance with the T2 pre-
dictors (r = .42 to .58, see Table 1). Here, only the concurrent measures of SOC
(T2) added unique predictive variance (see Tables 4 and 5).
Finally, we investigated whether SOC at T1 uniquely contributes to 3-year
changes in general and work-specific well-being above and beyond the “Big Five”
personality traits. We used hierarchical regression analyses comprising four steps:
In step 1 we entered the well-being autocorrelation, in step 2 we added the NEO
variables, in step 3 we included SOC measured at Time 1, and in step 4 we
added SOC measured at Time 2. Remarkably, after controlling for the predictive

TABLE 5
Hierarchical Multiple Regression Analyses: Predictions from Selection, Optimization,
and Compensation on Job Satisfaction

Job Satisfaction (Time 2)

R R 2

Model 1
Step 1: Job Satisfaction (Time 1) .58∗∗
Step 2: SOC (Time 1) .62∗∗ .05
Step 3: SOC (Time 2) .69∗∗ .10∗∗
Model 2
Step 1: Job Satisfaction (Time 1) .58∗∗
Step 2: SOC (Time 2) .66∗∗ .11∗∗
Step 3: SOC (Time 1) .69∗∗ .04
∗p < .05, one-tailed.
∗∗ p < .01, one-tailed.
 LONGITUDINAL PREDICTIVE POWER OF SOC 329

contributions of the autocorrelation of emotional balance (step 1; R = .47,

p < .01, one-tailed) and the NEO variables (step 2; R = .59, p < .01; R 2 = .12;
p < .01), SOC measured at T1 still added to the prediction of emotional balance
3 years later (step 3; R = .67, p < .01; R 2 = .10, p < .01). For job satisfaction
and subjective success ratings in the occupational domain, however, entering SOC
measured at T1 after entering well-being autocorrelations and the NEO variables
in steps 1 and 2 ( job satisfaction: R = .63, p < .01; subjective success: R = .52,
p < .01, one-tailed) evinced no significant unique increases in predictive power
in step 3 ( job satisfaction: R = .66, p < .01, R 2 = .03, ns; subjective success:
R = .55, p < .01, R 2 = .03, ns). But note that the concurrent measures of SOC
(T2) significantly contributed to the prediction of work-related well-being over
and above the contributions of the autocorrelations and the NEO variables (step 4;
emotional balance: R = .73, p < .01, R 2 = .09, p < .01; job satisfaction:
R = .72, p < .01, R 2 = .09, p < .01; subjective success in the work domain:
R = .63, p < .01, R 2 = .10, p < .01).

DISCUSSION
In general, the results demonstrate the longitudinal predictive power of SOC.
In this instance, the outcomes were general and work-related well-being. In our
study, we gave special attention to the question of whether SOC has a predictive
role for general and work-specific well-being over and above dispositional traits
(i.e., “Big Five”). With regard to subjective well-being, numerous studies (e.g.,
Schmutte & Ryff, 1997; Tokar & Subich, 1997; Salgado, 1997) have demonstrated
the predictive power of the “Big Five” personality factors.
Recent personality concepts, however, also include more dynamic concepts
(e.g., individual differences in goal setting and planning; Cantor & Kihlstrom,
1981; Frese, Stewart, & Hannover, 1987; Mischel, 1990). SOC-related behaviors
fall into such dynamic personality processes as they represent action styles or
strategies. In cross-sectional empirical studies, Freund and Baltes (1998) as well as
Wiese et al. (2000) showed that controlling for the influence of enduring personality
traits (i.e., “Big Five”) did not reduce the significance of the positive correlations
between SOC and well-being. The present data present first evidence that this is
also true for longitudinal prediction.
Concerning openness to experience, we have to stress that this finding has to
be interpreted with caution since, in this study, the scale turned out to be rather
low in internal consistency, which could be seen as an indicator of low reliability.
It is known, however, that also conceptually openness to experience is the most
heterogeneous factor of the “Big Five” for which low internal consistency is to
be expected (McCrae, 1996; De Raad & Van Heck, 1994; De Raad, Perugini,
Hrebickova, & Szarota, 1998). In McCrae et al. (1999) the German version of
the openness scale in its full length showed the lowest internal consistency when
compared to other NEO traits. The even lower internal consistency in our study
probably results from the fact that we used a short version of this scale. For reasons
of completeness, however, we left this scale in our NEO battery.
330 WIESE, FREUND, AND BALTES

As another limitation of this study, note that the follow-up sample evinced sig-
nificantly higher scores on agreeableness than those who dropped out. It is, of
course, not surprising to find agreeable persons to be more willing to support our
research. In addition, it has been shown in studies on convergent and discriminant
validity of self-reported SOC by Freund and Baltes (2001) as well as in the present
study (r = −.00 to .10, ns; see also Wiese et al., 2000) that there is no significant
link between agreeableness and SOC. The critical point, however, is that agreeable-
ness (measured at T1) is associated with job satisfaction (T1: r = .22, p < .01;
T2: r = .30, p < .01). Could this have affected our findings? We argue that the
positive link between SOC and agreeableness points to the possibility that we even
underestimated the relation between SOC and job satisfaction since the variance
of job satisfaction is probably restricted in our follow-up sample.
We stress that a shared predictive variance between SOC and personality traits
is part of the SOC model (Freund & Baltes, 2001; Wiese, 2000; Wiese et al.,
2000). Conscientiousness, for instance, describes a person as being a good time
manager, highly organized, ambitious, and hard-working. Of course, positive cor-
relations between conscientiousness and SOC behaviors are to be expected (here:
r = .38 to .40, p < .01; see Wiese et al., 2000) and can be interpreted as indi-
cators of convergent validity (see also Freund & Baltes, 2001). This holds also
true for the positive correlations between extraversion and optimization (r = .31,
p < .01; see Wiese et al., 2000) as well as between extraversion and compensation
(r = .26, p < .01; see Wiese et al., 2000) since persons who are outgoing, opti-
mistic, and self-confident have probably fewer difficulties in being active regulators
of their development. Further indicators of validity are the negative correlations
between neuroticism and optimization (r = −.34, p < .01; see Wiese et al.,
2000) as well as between neuroticism and compensation (r = −.31, p < .01;
see Wiese et al., 2000) since persons who have higher scores on the neuroticism
scale describe themselves as feeling rather pessimistic and helpless when con-
fronted with life management problems. But note that aside from the theoretically
meaningful shared variance of SOC with personality traits, the SOC strategies
contain uniqueness and therefore cannot be reduced to well-known personality
variables.
To prevent a possible misconception, we want to add one further observation on
the question of shared versus unique predictive variance of SOC in comparison with
constructs such as the NEO. It would be inappropriate for the further development
of SOC to ignore the predictive variance it shares with the NEO and evaluate its
theoretical and empirical utility only in terms of uniqueness of prediction. Since
SOC has been shown to have such uniqueness, both in past and present work, the
view on this question is intended to underscore a more general issue of innovation
in theory and practice. To ignore the predictive variance that SOC shares with other
predictors (such as the NEO) would give undue preference to constructs that are
part of the established field of psychological constructs. As new constructs such as
SOC are developed, some of their variance is expected to be associated with other
behavioral constructs. We consider such shared associations in predictive validity
 LONGITUDINAL PREDICTIVE POWER OF SOC 331

to be a strength and not a weakness. At the same time, obtaining additional evidence
for uniqueness, as we do, supports the case for the usefulness of our “new” approach
to understand modes of life management.
In this vein, we conducted additional analyses. We asked whether similar covari-
ance and unique variance existed with other well-established personality constructs.
As the earlier study did also comprise, for instance, data on control beliefs and
self-efficacy beliefs (measured according to Krampen, 1991; see Wiese, 2000)—a
group of variables also known to be related to well-being (see Bandura, 1995;
Schwarzer, 1992)—we could tackle this question. Whereas control beliefs and
self-efficacy beliefs measured at Time 1 did not predict work-related well-being
in a 3-year interval, they were predictive of emotional balance (Multiple R = .32,
p < .05). They did not, however, significantly add ( R 2 = .04, ns) to well-
being (Time 2) over and above SOC (Time 1). This result strengthens our conclu-
sion that SOC carries uniqueness in the longitudinal prediction of well-being. In
addition, this result supposes that believing in ones’ competencies is not enough
to be successful if such beliefs are not followed by adaptive means of regulating
goal-relevant activities in the sense of SOC.
In addition, one might ask whether—in addition to the longitudinal influence
of SOC on well-being—there is also an impact of well-being on the subsequent
implementation of SOC-relevant strategies. Post hoc analyses revealed that neither
selection nor compensation at T2 could be predicted by general or domain-specific
well-being (measured at T1). However, emotional balance (r = .29, p < .01) and
job satisfaction (r = .22, p < .01) at T1 predicted the use of optimizing strategies
at T2. Since there has been a positive cross-sectional link between optimization
and well-being at T1 (Wiese et al., 2000), it is not surprising to find this predictive
power of well-being on the future use of optimization.
Whether individuals have insight into the underlying causal processes (i.e., in-
ferring that optimization leads to positive outcomes and therefore trying to continue
optimizing), however, remains to be investigated. In fact, the long time interval in
between the two measurements leaves us with open questions on the mechanisms
and processes that constitute the operation of SOC. Although longitudinal, our
design does not really allow us to analyze and understand the functional processes
that took place during the intervening period of 3 years. Therefore, from a concep-
tual as well as from a methodological point of view, more fine-grained analyses
comprising more measurement points are required (see Schmitz, 2000).
Furthermore, one should note that the present study focuses on subjective mea-
sures of developmental success. Subjective evaluation of one’s well-being in the
work domain can be viewed as a meaningful outcome when interested in success-
ful life management. An individual’s well-being is, in and of itself, an important
hallmark of successful development. After all, who would call a person a success-
ful life manager who feels down and dissatisfied with him- or herself and his or
her work life? To get a more complete picture of how SOC exerts an influence
on career development and occupational outcomes, however, the use of objec-
tive measures (e.g., performance, absenteeism, and turnover) seems desirable. Of
332 WIESE, FREUND, AND BALTES

course, even objective measures are not unambiguous when it comes to evaluate
development (Vondracek, Lerner, & Schulenberg, 1983; M. Baltes & Carstensen,
1996). In fact, they rather seem to represent occupational outcomes that indicate
occupational adjustment than indicators of a successful developmental progress.
However, the exclusive reliance on self-report measures is a clear limitation of
the present study. In the work domain, the associations between well-being and
objective occupational outcomes are often of only moderate, sometimes even negli-
gible, size (see Iaffaldano & Muchinsky, 1985; Judge, Thoresen, Bono, & Patton,
2001). Therefore, one cannot infer from a positive relation between SOC and
subjective well-being that there is also a positive link to objective indicators of
successful life management in the occupational domain. This link awaits to be
investigated in future empirical studies.
The inclusion of objective criteria would also allow to test whether selection
would have a stronger effect on objective than on subjective measures. In fact,
as in the cross-sectional analyses (for a detailed discussion see Wiese et al.,
2001), selection turned out to be of minor importance for subjective well-being.
One interpretation is that selection implies some loss aspect because focusing on
the most important goals excludes alternative options and pathways (P. Baltes,
1997; Freund & Baltes, 2001). This loss aspect might counteract feelings of
well-being.
In addition, one could integrate family life as another central domain of life
management into the present research strand (Wiese & Freund, 2000). In fact,
originally it was planned to also analyze and describe the longitudinal predictive
power of SOC on partnership-related well-being since cross-sectionally, positive
associations between SOC and well-being have been found (Wiese et al., 2000).
It turned out, however, that partnership-related well-being could not be predicted
over the 3-year interval. Only compensation at Time 1 significantly predicted
satisfaction with the current partnership (r = .21, p < .05, one-tailed). One
has to take into account, however, that in the life phase considered, partnership
instability is rather high. Therefore, partnership-related ratings do not necessarily
refer to the same person. Here, again, it seems desirable to implement a design with
multiple measurements in shorter time intervals (e.g., several weeks or months).
Such a design might also decrease the drop-out rate (which has been rather high
in the present study) since participants would know beforehand that they will be
contacted several times.
A cross-domain approach, however, should not be restricted to an individual
level (i.e., asking for life management in the occupational and private domain).
Since life management is also embedded in social contexts, it would also be desir-
able to include collective processes of life management (Berg, Meegan, & Deviney,
1998; M. Baltes & Carstensen, 1998). This would also be in line with the SOC
approach, which represents a life-span model that builds on a conception of devel-
opment as a multicausal, multidimensional, and highly contextualized process. In
fact, in vocational psychology it has been called for the adaptation of models which
stress a dynamic interaction between developing individuals and their changing
environments (Vondracek et al., 1983).
 LONGITUDINAL PREDICTIVE POWER OF SOC 333

Finally, future research should focus on intervention strategies for promoting

SOC behaviors. There is evidence training programs can increase SOC-consistent
behaviors (e.g., planning career outcomes; Marko & Savickas, 1998). Thus, the
development and evaluation of intervention strategies focusing on SOC in the
occupational domain seems to be a promising endeavor.
In conclusion, our study provides initial longitudinal evidence that SOC con-
tributes to general well-being and subjective career success. Our results add to
those research approaches in organizational psychology that outline the central
role of agentic personality constructs (e.g., proactive personality and personal ini-
tiative; cf. Bateman & Crant, 1993; Frese, Fay, Hilburger, Leng, & Tag, 1997;
Seibert et al., 1999). The SOC framework, however, offers more than just another
agentic personality construct focussing on self- and action regulation useful for
predicting career success. In its meta-theoretical formulation, the SOC model al-
lows one to comprehensively integrate theoretical concepts and empirical findings
of developmental, vocational, and organizational psychology.

REFERENCES
Abraham, J. D., & Hansson, R. O. (1995). Successful aging at work: An applied study of selection,
optimization, and compensation through impression management. Journal of Gerontology, 50,
94–103.
Baltes, B. B., & Dickson, M. W. (2001). Using life-span models in industrial-organizational psychology:
The theory of selective optimization with compensation. Applied Developmental Sciences, 5,
51–61.
Baltes, M. M., & Carstensen, L. L. (1996). The process of successful ageing. Ageing and Society, 16,
397–422.
Baltes, M. M., & Carstensen, L. L. (1998). Social psychological theories and their application to aging:
From individual to collective selective optimization with compensation. In V. L. Bengston &
K. W. Schaie (Eds.), Handbook of theories of aging (pp. 209–226). New York: Springer.
Baltes, P. B. (1997). On the incomplete architecture of human ontogeny: Selection, optimization, and
compensation as foundation of developmental theory. American Psychologist, 52, 366–380.
Baltes, P. B., & Baltes, M. M. (1990). Psychological perspectives on successful aging: The model
of selective optimization with compensation. In P. B. Baltes & M. M. Baltes (Eds.), Successful
aging: Perspectives from the behavioral sciences (pp. 1–34). Cambridge, UK: Cambridge Univ.
Press.
Baltes, P. B., Baltes, M. M., Freund, A. M., & Lang, F. (1999). The measurement of selection, opti-
mization, and compensation (SOC) by self-report: Technical report 1999. Berlin, Germany: Max
Planck Institute for Human Development.
Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (Ed.). (1995). Self-efficacy in changing societies. New York: Cambridge Univ. Press.
Bateman, T. S., & Crant, J. M. (1993). The proactive component of organizational behavior. Journal
of Organizational Behavior, 14, 103–118.
Berg, C. A., Meegan, S. P., & Deviney, F. P. (1998). A social-contextual model of coping with everyday
problems across the life-span. International Journal of Behavioral Development, 22, 239–261.
Brandtstädter, J., & Renner, G. (1990). Tenacious goal pursuit and flexible goal adjustment: Explication
and age-related analysis of assimilative and accomodative strategies of coping. Psychology and
Aging, 5, 58–67.
Cantor, N., & Kihlstrom, J. F. (1981). Personality, cognition, and social interaction. Hillsdale, NJ:
Erlbaum.
Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory manual. Odessa, FL: Psy-
chological Assessment Resources.
334 WIESE, FREUND, AND BALTES

De Raad, B., & Van Heck, G. (1994). The fifth of the Big Five [Special Issue]. European Journal of
Personality, 8(4).
De Raad, B., Perugini, M., Hrebickova, M., & Szarota, P. (1998). Lingua franca of personality:
Taxonomies and structures based on the psycholexical approach. Journal of Cross Cultural
Psychology, 29, 212–232.
Fay, D., & Frese, M. (2000). Self-starting behavior at work: Toward a theory of personal initiative. In
J. Heckhausen (Ed.), Motivational psychology of human development: Developing motivation and
motivating development (pp. 307–324). Amsterdam: Elsevier.
Fay, D., & Frese, M. (2001). The concept of personal initiative: An overview of validity studies. Human
Performance, 14, 97–124.
Frese, M., Fay, D., Hilburger, T., Leng, K., & Tag, A. (1997). The concept of personal initiative:
Operationalization, reliability and validity in two German samples. Journal of Occupational and
Organizational Psychology, 70, 139–161.
Frese, M., Stewart, J., & Hannover, B. (1987). Goal orientation and planfulness: Action styles as
personality concepts. Journal of Personality and Social Psychology, 52, 1182–1194.
Freund, A. M., & Baltes, P. B. (1998). Selection, optimization, and compensation as strategies of life
management: Correlations with subjective indicators of successful aging. Psychology and Aging,
13, 531–543.
Freund, A. M., & Baltes, P. B. (2000). The orchestration of selection, optimization, and compensation:
An action-theoretical conceptualization of a theory of developmental regulation. In W. J. Perrig
& A. Grob (Eds.), Control of human behaviour, mental processes and consciousness (pp. 35–58).
Mahwah, NJ: Erlbaum.
Freund, A. M., & Baltes, P. B. (2001). Life management strategies of selection, optimization, and
compensation: Measurement by self-report and construct validity. Manuscript submitted for pub-
lication.
Giegler, H. (1985). Rasch-Skalen zur Messung von Arbeits- und Berufszufriedenheit, Betriebsklima
und Berufsbelastung auf seiten der Betroffenen [Rasch scales for measuring job and occupational
satisfaction, working conditions, and occupational stress for the person affected]. Zeitschrift für
Sozialpsychologie, 16, 13–28.
Hall, D. T., & Mirvis, P. H. (1995). Careers as lifelong learning. In A. Howard (Ed.), The changing
nature of work (pp. 323–361). San Francisco, CA: Jossey–Bass.
Howard, A. (1995). A framework for work change. In A. Howard (Ed.), The changing nature of work
(pp. 3–44). San Francisco, CA: Jossey–Bass.
Iaffaldano, M. T., & Muchinsky, P. M. (1985). Job satisfaction and job performance: A meta-analysis.
Psychological Bulletin, 97, 251–273.
Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K. (2001). The job satisfaction–job performance
relationship: A qualitative and quantitative review. Psychological Bulletin, 127, 376–407.
Kanfer, R., & Heggestad, E. D. (1999). Individual differences in motivation: Traits and self-regulatory
skills. In P. L. Ackerman & P. C. Kyllonen (Eds.), Learning and individual differences: Pro-
cess, trait, and content determinants (pp. 293–313). Washington, DC: American Psychological
Association.
Kossek, E. E., & Ozeki, C. (1998). Work-family conflict, policies, and the job-life satisfaction rela-
tionship: A review and directions for organizational behavior–human resources research. Journal
of Applied Psychology, 83, 139–149.
Krampen, G. (1991). Der Fragebogen zu Kompetenz- und Kontrollüberzeugungen [Questionnaire for
the measurement of self-efficacy and control beliefs]. Göttingen, Germany: Hogrefe.
Kuhl, J., & Beckmann, J. (1994). Volition and personality: Action versus state orientation.
Göttingen/Seattle: Hogrefe.
Locke, E. A., Shaw, K. N., Saari, L. M., & Latham, G. P. (1981). Goal setting and task performance:
1969–1980. Psychological Bulletin, 90, 125–152.
Marko, K. W., & Savickas, M. (1998). Effectiveness of a career time perspective intervention. Journal
of Vocational Behavior, 52, 106–119.
Marsiske, M., Lang, F. R., Baltes, P. B., & Baltes, M. M. (1995). Selective optimization with compen-
sation: Life-span perspectives on successful human development. In R. A. Dixon & L. Bäckman
 LONGITUDINAL PREDICTIVE POWER OF SOC 335

(Eds.), Compensating for psychological deficits and declines: Managing losses and promoting
gains (pp. 35–79): Mahwah, NJ: Erlbaum.
McCrae, R. R. (1996). Social consequences of experiential openness. Psychological Bulletin, 120,
323–337.
McCrae, R. R., Costa, P. T., De Lima, M. P., Simoes, A., Barbaranelli, C., Chae, J. H., & Piedmont,
R. L. (1999). Age differences in personality across the adult life span: Parallels in five cultures.
Developmental Psychology, 35, 466–477.
Mischel, W. (1990). Personality dispositions revisited and revised: A view after three decades. In L. A.
Pervin (Ed.), Handbook of personality: Theory and research (pp. 11–134). New York: Guilford
Press.
Mischel, W. (1996). From good intentions to willpower. In P. M. Gollwitzer & J. A. Bargh (Eds.), The
psychology of action: Linking cognition and motivation to behavior (pp. 197–218). New York:
Guilford Press.
Salgado, J. F. (1997). The five factor model of personality and job performance in the European
Community. Journal of Applied Psychology, 82, 30–43.
Savickas, M. L. (2001). Toward a comprehensive theory of career development: Dispositions, con-
cerns, and narratives. In F. T. L. Leong & A. Barak (Eds.), Contemporary models in vocational
psychology: A volume in honor of Samuel H. Osipow (pp. 295–320). Mahwah, NJ: Erlbaum.
Schmutte, P. S., & Ryff, C. D. (1997). Personality and well-being: Reexamining methods and meanings.
Journal of Personality and Social Psychology, 73, 549–559.
Schmitz, B. (2000). Auf der Suche nach dem verlorenen Individuum: Vier Theoreme zur Aggregation
von Prozessen [On search for lost individual: Four theorems on the aggregation of processes].
Psychologische Rundschau, 51, 83–92.
Schwarzer, R. (Ed.). (1992). Self-efficacy: Thought control of action. Washington, DC: Hemisphere.
Seibert, S. E., Crant, M., & Kraimer, M. L. (1999). Proactive personality and career success. Journal
of Applied Psychology, 84, 416–427.
Sternberg, R. J. (1994). Thinking styles: Theory and assessment at the interface between intelligence
and personality. In R. J. Sternberg & P. Ruzgis (Eds.), Personality and intelligence (pp. 169–187).
New York: Cambridge Univ. Press.
Super, D. E. (1980). A life-span, life-space approach to career development. Journal of Vocational
Behavior, 16, 282–298.
Super, D. E., Savickas, M. L., & Super, C. M. (1996). The life-span, life-space approach to careers. In
D. Brown, L. Brooks, & Associates (Eds.), Career choice and development: Applying contempo-
rary theories to practice (3rd ed., pp. 121–178). San Francisco, CA: Jossey–Bass.
Tokar, D. M., & Subich, L. M. (1997). Relative contributions of congruence and personality dimensions
on job satisfaction. Journal of Vocational Behavior, 50, 482–491.
Vondracek, F. W., Lerner, R., & Schulenberg, J. E. (1983). The concept of development in vocational
theory and intervention. Journal of Vocational Behavior, 23, 179–202.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of
positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology,
54, 1063–1070.
Wiese, B. S. (2000). Berufliche und familiäre Zielstrukturen [Work-related and family-related goal
structures]. Münster, Germany: Waxmann.
Wiese, B. S., & Freund, A. M. (2000). The interplay of work and family in young and middle adulthood.
In J. Heckhausen (Ed.), Motivational psychology of human development: Developing motivation
and motivating development (pp. 233–249). Amsterdam: Elsevier.
Wiese, B. S., & Freund, A. M. (2001). Zum Einfluss persönlicher Prioritätensetzungen auf die
Stimulspräferenz: Erste empirische Befunde [The influence of setting personal priorities on mea-
sures of preference: First empirical evidence]. Zeitschrift für Experimentelle Psychologie, 48,
57–73.
Wiese, B. S., Freund, A. M., & Baltes, P. B. (2000). Selection, optimization, and compensation: An
action-related approach to work and partnership. Journal of Vocational Behavior, 57, 273–300.

Received March 13, 2001; published online November 16, 2001