An Empirical Investigation Career Success PDF

PERSONNEL PSYCHOLOGY
1995,48
AN EMPIRICAL INVESTIGATION OF THE

PREDICTORS OF EXECUTIVE CAREER SUCCESS
TIMOTHY A. JUDGE
Department of Management and Organizations
University of Iowa
DANIEL M. CABLE
College of Business and Management
Georgia Institute of Technology
JOHN W. BOUDREAU
Center for Advanced Human Resource Studies
Cornell University
ROBERT D. BRETZ, JR.
Department of Management and Organizations
University of Iowa
This study examined the degree to which demographic, human capital,

motivational, organizational, and industrylregion variables predicted
executive career success. Career success was assumed to comprise ob-
jective (pay, ascendancy) and subjective (job satisfaction, career satis-
faction) elements. Results obtained from a sample of 1,388U.S.exec-
utives suggested that demographic, human capital, motivational, and
organizational variables explained significant variance in objective ca-
reer success and in career satisfaction. Particularly interesting were
findings that educational level, quality, prestige, and degree type all
predicted financial success. In contrast, only the motivational and or-
ganizational variables explained significant amounts of variance in job
satisfaction. These findings suggest that the variables that lead to ob-
jective career success often are quite different from those that lead to
subjectively defined success.
What factors lead some executives to be more successful in their ca-

reers than others? This interesting and important question has been only
partially answered through prior research. In fact, examination of the
relevant literatures reveals that knowledge of executive career success
The majority of the work on this paper was completed while the first, second, and
fourth authors were at Cornell University. The authors thank Sharon Voros of Paul Ray
Berndtson for assistancewith this study. The authors also appreciate the helpful comments
of three anonymous reviewers on earlier drafts of this paper. The Center for Advanced
Human Resource Studies, Cornell University,provided funding for this study.
Correspondence and requests for reprints should be addressed to Timothy A. Judge,
Department of Management and Organizations, College of BusinessAdministration, Uni-
versity of Iowa, Iowa City IA 52242-1000.
COPYRIGHT 0 1995 PERSONNEL PSYCHOLOGY, INC
485
486 PERSONNELPSYCHOLOGY
can be enhanced in several ways. First, researchers have predicted ca-

reer success primarily by using a few variables in a piecemeal fashion,
without considering the relative effects of manifold sets of theoretically
based variables (e.g., Gattiker & Larwood, 1989; Judge & Bretz, 1994).
Furthermore, although executive career success has generated consid-
erable interest in the business press, little rigorous empirical research
is available. Third, little research has examined executives’ satisfaction
with their careers, and research that is available often has relied exclu-
sively on common-method, self-report data (cf. Cox & Cooper, 1989;
Gattiker & Larwood, 1986,1988; Judge & Bretz, 1994). Finally, almost
no research has simultaneously examined both the objective (e.g., com-
pensation) and subjective (e.g., career satisfaction) aspects of career suc-
cess (Gattiker & Larwood, 1989), although both appear to be essential
to a complete treatment of this issue.
Accordingly, the present study proposes and tests a comprehensive
model of executive career success that includes both objective and sub-
jective elements. The predictors within this model are derived from past
research, and include a wider range of theoretically relevant variables
than have been included in any single prior study. Thus, results from
the test of the hypothesized model should provide the most comprehen-
sive evidence to date regarding the predictors of career success among
executives.
Conceptual Model of Executive Career Success
Consistent with Judge and Bretz (1994) and London and Stumpf
(1982), we define career success as the positive psychological or work-
related outcomes or achievements one has accumulated as a result of
one’s work experiences. As Jaskolka, Beyer, and Trice (1985) noted, ca-
reer success is an evaluative concept, so judgments of career success de-
pend on who does the judging. Career success as judged by others is de-
termined on the basis of relatively objective and visible criteria (Jaskolka
et al., 1985). Researchers often refer to this type of career success as ob-
jective success because it can be measured by observable exoteric metrics
such as salary and number of promotions (Gattiker & Larwood, 1988;
Judge & Bretz, 1994; Kotter, 1982). Thus, we define objective career
success as observable career accomplishmentswhich can be measured
against the metrics of pay and ascendancy (London & Stumpf, 1982).
Career success also can be judged by the individual pursuing the ca-
reer. Most research on career success typically has focused on objective
success (e.g., Kotter, 1982), rather than individual appraisals of their
own success (Gattiker & Larwood, 1989). Even more rare is research
that considers objective and subjective dimensions together (Gattiker
TIMOTHY A. JUDGE ET AL. 487
& Larwood, 1989). Past research has suggested that many individuals
who are extrinsically successful do not feel successful or satisfied with
their achievements (Korman, Wittig-Berman, & Lang, 198l), so it is im-
portant to consider both objective and subjective evaluations of career
success (Howard & Bray, 1988; Gattiker & Larwood, 1989). Accord-
ingly, our model includes subjective career success, defined as individu-
als’feelings of accomplishment and satisfactionwith their careers. There
is a link between objective success and subjective appraisals in that indi-
viduals define their success based, in part, on their objective accomplish-
ments. In fact, past research generally has found that objective and sub-
jective success are positively but moderately related (Bray & Howard,
1980; Harrell, 1969; Judge & Bretz, 1994).
Because a career is a sequence of work-related positions (jobs) oc-
cupied throughout a person’s life (London & Stumpf, 1982), we define
subjective career success to include current job satisfaction just as the
career includes the current job. Consistent with Locke (1976), overall
job satisfaction is defined as “a pleasurable or positive emotional state
resulting from an appraisal of one’s job or job experiences” (p. 1300).
Career satisfaction, in turn, is defined as the satisfaction individuals de-
rive from intrinsic and extrinsic aspects of their careers, including pay,
advancement, and developmental opportunities (Greenhaus, Parasura-
man, & Wormley, 1990).
Figure 1 displays the hypothesized model of career success. Consis-
tent with Judge and Bretz (1994) and Whitely, Dougherty, and Dreher
(1991), we assume that objective career success consists of compensation
and ascendancy (number of promotions). As the figure shows, we hy-
pothesize that several categories of variables (i.e., demography, human
capital, motivation, organization, and industryhegion) predict objective
career success. We discuss each category of predictors in turn.
Demographic variables. According to Pfeffer (1983), the demography
of an organization’s members may influence many behavioral patterns
and outcomes, including promotions and salary attainment. Thus, de-
mographic variables need to be taken into account when investigating
the predictors of career success. Several studies have found that de-
mographic variables explain more variance in career success than other
sets of influences (Gattiker & Larwood, 1988, 1989; Gould & Penley,
1984). One of the most obvious and consistent findings regarding demo-
graphic influences is that age positively predicts objective success (Cox &
Nkomo, 1991; Gattiker & Larwood, 1988,1989; Gutteridge, 1973; Har-
rell, 1969; Jaskolka et al., 1985), presumably because extrinsic outcomes
accrue over time.
Another relatively consistent finding is that married individuals
achieve higher levels of objective success than unmarried individuals
-IndividualCharacteristics
Objective Career Success Subjective Career Success

* Job Satisfaclioii
NurnberofPrornollons -Career Satisfaction
- H o w ofWorkDesired
Organizational I Industry
Characteristics
OrganizationSize
- OrganimtionSuccess
* Public Organization
* Industry Sector
.Region-
Figure 1: Conceptual Model of Career Success.
(Judge & Bretz, 1994; Pfeffer & Ross, 1982). As Pfeffer and Ross (1982)
pointed out, marriage may act as a signal to organizations of the exis-
tence of positive attributes in the individual, such as stability, responsi-
bility, and maturity (Bloch & Kuskin, 1978). Furthermore, spouses often
act as resources for managers because they can assist with household re-
sponsibilities, offer emotional support, and provide consultation on job-
related matters (Pfeffer & Ross, 1982). On the other hand, a spouse with
a job outside the home diminishes the resources that can be devoted to
the manager’s career (Pfeffer & Ross, 1982). Thus, marital status should
positively predict objective success whereas having a spouse employed
outside the home should negatively predict objective career success (Pf-
effer & Ross, 1982). Additionally, research has suggested that because
hours devoted to dependent care and other household responsibilities
represent time away from work, the time spent on such responsibilities
negatively affects career success (Bielby & Bielby, 1988). Thus, time de-
voted to dependent responsibilities should negatively predict objective
career success.
Numerous studies have found that compared to White managers,
minority managers receive lower evaluations in terms of estimated job
qualifications, performance ratings, and pay and promotions (Cox &
Nkomo, 1991; Greenhaus et al., 1990). A considerabIe amount of re-
search on gender differences in career progression has revealed simi-
lar findings in terms of pay, performance ratings, and promotions (e.g.,
Carlson & Swartz, 1988). Conversely,some research suggests that in cer-
tain situations women and minorities receive more favorable treatment
with respect to promotions and pay raises than White men (Gerhart &
Milkovich, 1989; Tsui & Gutek, 1984).
Thus, evidence suggests that women and minorities are treated dif-
ferently (and sometimes more favorably) than their White male coun-
terparts. However, when levels of career attainment are evaluated-as
opposed to the outcomes of specific personnel decisions-the evidence
is also fairly clear that women and minorities have lower levels of career
success than White men (Cox & Nkomo, 1991). Accordingly, we expect
that minority and female executives will have lower levels of objective
career success than White and male executives.
Human capital variables. Human capital theory posits that the labor
market rewards investments individuals make in themselves, and that
these investments lead to higher ascendancy rates and salaries (Becker,
1964). Here we define human capital to include the cumulative edu-
cational, personal, and professional experiences that might enhance an
executive’s value to an employer. Level of education is the human cap-
ital attribute that has been the subject of the most research. Research
from the labor economics and careers literatures indicates that returns
from educational attainment in terms of pay and promotions are sig-

nificant (Jaskolka et al., 1985; Pfeffer & Ross, 1982; Psacharopoulos,
1985; Whitely et al., 1991). Thus, we predict a positive relationship be-
tween level of education and objective career success. It also appears
important to examine the effect of the educational content (e.g., execu-
tive’s major field of study) because research suggests that organizations
reward business, law, and engineering degrees more than other types
of education (e.g., Swinyard & Bond, 1980; Useem & Karabel, 1986).
Thus, we expect that executives with degrees in business, engineering,
and law will have higher levels of objective success than executiveswith
degrees in other areas.
Although research has revealed a great deal of information concern-
ing the relationship between quantity of education and career success,
less is known about the effects of educational quality on career out-
comes (for an exception see Solmon, 1973). Descriptive studies suggest
that successful executives are disproportionately graduates from well-
regarded universities (Swinyard & Bond, 1980; Warner & Abegglen,
1955), so the role of educational quality in executive career attainment
is an important yet unexplored issue. As noted by Useem and Kara-
be1 (1986), an educational institution may bestow three distinct types
of human capital upon its graduates: scholastic capital (the amount of
knowledge acquired), social capital (personal contacts, network ties, in-
culcation of achievement motivation), and cultural capital (the value so-
ciety places on symbols of prestige). The quality of the school attended,
in terms of research and instruction, resources, quality of students, and
so on, would seem to provide a future executive with scholastic capital.
Thus, the quality of the university from which the executive earned his
or her highest degree should positively predict objective success.
On the other hand, the quality of the school per se may or may not
provide social and cultural capital. These latter forms of capital would
seem to be prevalent in universities that have achieved a certain level
of prestige and status, such as Ivy League universities (Dumhoff, 1967).
Analyses of educational institutions’ status have found that Ivy League
universities are disproportionatelyrepresented, and estimates of univer-
sities with the most prestige usually include most or all Ivy League uni-
versities (Useem & Karabel, 1986). Although there is likely a positive
relationshipbetween the status and quality of a university, some universi-
ties’ reputations surpass their actual quality, and other universities’ true
value exceeds their reputation. Because Ivy League universities have a
high degree of status, because such universities are likely to be particu-
larly beneficial in bestowing social and cultural capital upon their grad-
uates (Useem & Karabel, 1986), and because graduates from these uni-
versities may benefit from policies of nepotism or favoritism beyond any
human capital acquired (Thelin, 1976), we predict that controlling for

educational quality, being a graduate of a prestigious (i.e., Ivy League)
university positively predicts objective career success.
Other than education, we expect other human capital variables to
predict objective career success. Research suggests that job tenure and
total time in one’s occupation are positively related to career attainment
(Cox & Harquail, 1991; Gutteridge, 1973; Jaskolka et al., 1985; Judge
& Bretz, 1994; Pfeffer & Ross, 1982; Whitely et al., 1991). Along with
amount of experience, type of experience may be relevant in predict-
ing career success. Specifically, it is becoming more important for ex-
ecutives to have international work experience (Cava & Mayer, 1993),
suggesting that organizations are more likely to reward and promote ex-
ecutives who have had international exposure (Kets de Vries & Mead,
1992). Thus, we expect that having job and occupational tenure and in-
ternational experience positively predict objective career success. An
important characteristic of professionals that should affect their career
success is their level of accomplishment in their job and career (Hough,
1984). One indicator of an executive’s “portable” value, or market
value, is an assessment of executives’ cumulative accomplishments and
future potential. An organization specializing in assessing the mar-
ketability of executives, such as an executive search firm, could provide
an estimate of executives’cumulative accomplishments. Thus, we expect
that executives’ accomplishment ratings should be positively related to
their objective career success.
Finally, an attribute that is expected to positively influence execu-
tives, objective career success is their appointments to the boards of di-
rectors of other firms. From a resource dependency perspective, exec-
utives on external boards play the important role of establishing inter-
firm coordination and serving as boundary spanners who cope with envi-
ronmental uncertainty (Edstrom & Galbraith, 1977; Haunschild, 1993).
Coping with uncertainty and controlling external information is believed
to confer power, often leading to scarce resources, such as pay and pro-
motions (Pfeffer, 1981). Thus, service on an external board of directors
should positively predict objective career success.
Motivational variables. Wolfle (1973) concluded that most studies
have not adequately considered the role of motivation in predicting earn-
ings, and Whitely et al. (1991) argued that motivational variables are
likely to be influential in predicting career success. Two variables in-
cluded by Whitely et al. as indicators of motivation were hours worked
per week and work centrality. Considerable research supports the re-
lationship between the number of hours worked per week and salary
and ascendancy (Cox & Cooper, 1989; Gutteridge, 1973; Harrell, 1969;
492 PERSONNEL PSYCHOLOGY
Judge & Bretz, 1994; Whitely et al., 1991). In the present study, we as-
sessed not only the number of hours worked per week, but also the num-
ber of evenings worked. Although hours worked and evenings worked
are related, working late at the office is a somewhat unique signal of mo-
tivation because of the family sacrificesit entails, and because of the pos-
itive impressionsit may generate among colleagues and superiors (Judge
& Bretz, 1994). Because both suggest high levels of motivation (Cox &
Cooper, 1989), we expect hours worked and evenings worked to pos-
itively predict objective success. In addition to time actually spent at
work, it is possible that the desire to spend time at work predicts career
success. Cox and Cooper (1989), in trying to discover the motivation be-
hind successful executives’ long work hours, found that these executives
enjoyed working long hours. Extrapolating from their findings, execu-
tives who desire to work more hours find their work motivating, and thus
should have a greater probability of success than other executives.
It seems logical that work centrality, or the degree of importance that
working has to the identity of an individual (England & Whitely, 1990),
positively relates to career attainment because individualswho see their
work as a central part of their lives should be more willing to make sig-
nificant investments in their work and in their careers. In fact, England
and Whitely (1990) found that the group of individuals who had the high-
est work centrality also had the highest net incomes. Another relevant
motivational variable is ambition. Howard and Bray (1988) found that
ambition, or the desire to get ahead, was one of the best predictors of
advancement in their study of AT&T managers. A positive relationship
between ambition and career success has been found in several other
studies of managers and executives (Canning & Montmarquette, 1991;
Cox & Cooper, 1989). Thus, we expect that the greater the number of
levels executives desire to advance, the greater will be their objective
success.
Organizational, industrial, and regional variables. Pfeffer (1991) em-
phasized the influence of structural variables, including both industry
and organizational characteristics, on individual outcomes such as per-
formance, turnover, and salaries. Several organizational-levelvariables
seemed reasonable to examine. One such variable is organization size.
Researchers have demonstrated that larger organizations pay employees
more than smaller organizations (see Brown & Medoff, 1989). How-
ever, because this finding may be attributed to causes that vary with firm
size, such as ability to pay, higher-quality workers, or lack of monitor-
ing ability, not all research has supported this relationship (e.g., Whitely
et al., 1991). Researchers also have argued that larger firms have a
greater number of job vacancies available, and thus have more promo-
tion opportunities (Dalton & Kesner, 1985; Whitely et al., 1991). How-
ever, it is not clear that there are more promotions available per indi-
vidual employee in larger organizations, because there are also more
people competing for the same promotions (e.g., Konda & Stewman,
1980; Pfeffer, 1983; Stewman & Konda, 1983). In fact, evidence has
been found for both a positive (Cox & Harquail, 1991) and a negative
(Cox & Nkomo, 1991) relationship between organization size and pro-
motion levels. Thus, size was included as a relevant variable to the pre-
diction of pay and promotions, but no projections were made about the
relationship between organization size and objective success.
Another relevant organizationalvariable is organization success. Al-
though the reported effects of firm performance on executive pay range
from a direct relationship (e.g., Murphy, 1985) to no relationship (e.g.,
Kerr & Bettis, 1987), most research suggests that organizational perfor-
mance positively influences executive earnings (Gomez-Mejia & Wel-
bourne, 1989). Thus, we expect a positive relationship between organi-
zation success and objective career success. We also examine whether
executives whose organizations’ stock is publicly traded are more suc-
cessful than those who work in private organizations. Although the ef-
fect of public status has not been investigated in the context of career suc-
cess, researchers have indicated that executives’compensation should be
related to the complexity and exposure of their organizations (Gomez-
Mejia & Balkin, 1992, p. 169), both of which should be greater in public
firms.
Organization size, success, and public visibility reflect factors asso-
ciated with the executive’s organization. However, executives also exist
within a broader labor market, which may reflect geographic and indus-
try differences in pay and career patterns (Campbell, Dunnette, Lawler,
& Weick, 1970; Gomez-Mejia & Welbourne, 1989; Gutteridge, 1973;
Judge & Bretz, 1994). Thus, our model includes these variables because
they have been suggested by past research, and to control for unmea-
sured factors that may be associated with industry and region.
Subjective career success. As noted earlier, subjective career success
can be conceptualized as consisting of two components: current job sat-
isfaction and career satisfaction. Figure 1shows a link between objective
and subjective career success. Based on past research that has found that
objective and subjective career success are positively related (Bray &
Howard, 1980; Harrell, 1969; Judge & Bretz, 1994), we believe that ob-
jective successwill positively predict subjective career success. Although
the causal direction of this relationship could be argued to be reciprocal,
in this study we assume that objective career success predicts subjective
success for several reasons. First, research has clearly established that
pay and promotion opportunities affect job and career attitudes (e.g.,
Gattiker & Larwood, 1988; Locke, 1976). The opposite causal direc-
tion-from subjective to objective success-is possible, but such a link
has not been directly demonstrated in the literature. Second, the tem-
poral ordering of the measurement of our variables was consistent with
the hypothesized ordering in Figure 1 (e.g., pay was measured prior to
job and career satisfaction), so our model is temporally correct (at least
with respect to pay). Although we use objective career success to pre-
dict subjective success, we do not suggest that any link between these
constructs can be inferred to be causal.
Past research has suggested that many of the variables that influ-
ence objective career success do not similarly influence subjective suc-
cess (Cox & Harquail, 1991; Judge & Bretz, 1994). As with job satisfac-
tion (e.g., Hulin, 1991; Judge & Locke, 1993), we expect that frames of
reference predict judgments of career success. Frames of reference are
self-referents-versus other-referents-where individuals evaluate their
inputs and outcomes against their own expectations (not against what
others receive; Hulin, 1991). The desirability of a particular level of ex-
trinsic outcomes likely depends on what standard or reference point the
executive uses. Demographic, human capital, and motivational factors,
because they serve as career inputs, may influence the internal standards
by which career success is judged. Thus, it is likely that these variables
act as frames of reference in evaluating job and career outcomes (Judge
& Locke, 1993).
Age and experience (job and occupation) may act as frames of refer-
ence in evaluating career outcomes because older and more experienced
executives may find a particular level of objective success (e.g., earn-
ing a $100,000salary and four promotions) less satisfylng than would a
younger or less experienced executive. In fact, empirical data support a
negative relationship between career satisfaction and age and tenure,
when controlling for extrinsic factors (Cox & Harquail, 1991; Cox &
Nkomo, 1991). Similarly, because individuals use their goals as criteria
against which they evaluate their success, those who set high goals (are
ambitious) have been found to be less satisfied with their current situa-
tion (Judge & Locke, 1993). Thus, we expect that ambition negatively
predicts job and career satisfaction. Another potentially relevant frame
of reference is gender. As Greenberg and McCarty (1990) noted, sev-
eral studies have shown that women have lower expectations regarding
pay and promotions than do men. This suggests that female executives
may be equally satisfied with a lesser level of objective outcomes (cf.
Dreher & Ash, 1990) or, equivalently,more satisfied with an equal level
of objective outcomes, compared to male executives. A comparable ar-
gument could be made with respect to race.
Although past research has not directly assessed the effects of other
variables that might act as frames of reference, we extrapolated from
Hulin’s (1991) job satisfaction model to formulate possible relationships.
First, in addition to tenure (discussed above), we propose that variables
serving as career inputs (e.g., education, hours worked) will negatively
predict career satisfaction when outcomes are held constant. For ex-
ample, if two executives earn similar salaries, we would expect the one
who has an undergraduate degree from a university of average quality
or prestige and who works relatively few hours per week to be more sat-
isfied than an executive who has earned a graduate degree from a high
quality or prestigious university and who works many hours per week
Similarly, a particular level of objective outcomes should be less satisfy-
ing to a highly accomplished executive. Thus, as with education quality
and prestige, holding outcomes constant, we expected rating of executive
accomplishmentsto negatively predict subjective career success.
Despite a number of the variables that are hypothesized to predict
objective career success, no comparable hypothesis can be made with
respect to subjective career success. For example, we have no basis to
offer directional hypotheses concerning the relationship between indus-
tryhegion variables and subjective success. Thus, with some variables
no specific directional effect on subjective career success was expected.
However, they were included in the model to preserve comparabilitybe-
tween the objective and subjective career success equations. Further, it
is possible that industry or region variables predict subjective success if
they happen to operate as frames of reference in the same way as career
inputs.
The data source of this study served as the basis for two other pub-
lications. One paper (Bretz, Boudreau, & Judge, 1994) focused on the
antecedents of job search behavior and the degree to which job search
relates to turnover decisions. The other paper (Judge, Boudreau &
Bretz, 1994) tested a causal model of executive job and life attitudes
(involving job stress, work-family conflict, job satisfaction, and life satis-
faction). The conceptual foundation, methodology, criterion variables,
and practical implicationsof these prior studies are quite different from
the present study. Thus, they could not feasibly be combined without
detracting from their scientificcontribution. However, because the data
source is the same and because all three studies focus on the same sam-
ple of executives, it is important to acknowledge the common data source
while also pointing out the distinctiveness of the studies (American Psy-
chological Association, 1994).
Method
Sample and Procedures
Subjects were executives contained in the database of Paul Ray

Berndtson, one of the largest executive search firms in the U.S.As is
typical of high-level executives, the vast majority of subjects were White
(97%) and male (93%). Average age of the executives was 45.5 years.
Ninety-one percent of executives were married; 43% of executives had a
spouse who was working outside the home. The average executive spent
55.7 hours per week in paid work and spent 4.9 hours per week caring
for dependents. Average annual salary was $126,890 (SD = $89,721);
average total pay, including bonuses, was $155,951 (SD = $133,642).
On average, executives had earned 6.4 promotions in their career, their
last promotion occurred 3.2 years ago, and they were positioned 2 levels
below the chief executive officer of their organization. Seventy percent
of respondents’ highest degree was an undergraduate degree, and 30%
of respondents had earned a master’s degree or higher. Roughly 9% of
the sample received their degree from an Ivy League university. The
distribution of degree type was as follows: business = 50%; engineering
= 16%; law = 2%; other = 32%. The average number of employees in
the executive’s organization was 11,690 and 12% of executives worked
in companies whose stock was publicly traded.
Paul Ray Berndtson’s database was used to identity the target sam-
ple and to collect archival data on the executives. The database con-
tained information concerning executiveswho had been identifiedby the
search firm as potential candidates for past and current position open-
ings. The search firm does not accept applications from executives, but
rather identifies candidates for inclusion in the database from a variety
of sources (10-K reports, industry publications & directories, etc.). Sur-
veys were mailed to a sample of 3,581 executives (a 50%random sample
of the data base). Accompanying the survey was a cover letter from the
chief executive officer of Paul Ray Berndtson soliciting the executives’
participation, and a stamped envelope addressed to the authors. We en-
coded surveys so that those returned could be matched with information
contained in the search firm’s data base. Executives were told in the
cover letter that their responses were confidential (the authors would
not know the names of the respondents and the search firm would not
have access to individual responses). Of the surveys that were mailed,
1,388 usable surveyswere returned, representing a response rate of 39%.
A MANOVA model, simultaneouslyconsidering the interrelated effects
of all variables, revealed no differences between respondents and non-
respondents concerning the study variables in the database (education,
salary, promotions, experience, marital status, age, race, sex, industry,

and region). This suggests that the sample was representative of all ex-
ecutives in the database.
Measures
Objective career success. Information on annual salary-as well as

bonuses, stock options, and other forms of cash compensation-was ob-
tained from Paul Ray Berndtson’s database. Although we used total an-
nual cash compensation as the measure of pay, annual salary and total
cash compensation were highly correlated with total pay (r=.94). The
search firm took numerous steps to ensure the accuracy of the compen-
sation data, as it is a critical piece of information in their placement pro-
cess. Archival salary was closely related to self-reported salary (the av-
erage deviation between self- and archival reports of salary was $1,497,
only a 1% deviation); to preserve independence in methods, the archival
data were used to measure compensation. The compensation levels of
executives in this sample are lower than the total compensation levels
typically reported in articles on executive pay in the popular press and
executive compensation literature. This may be due to several factors,
such as this sample includes many small and privately-held firms, where
pay levels may be lower. This sample also includes executives up to five
levels below the CEO, whereas these other literatures often focus on top
executives. Finally, it is possible that equity-based aspects of pay are not
fully reflected due to difficulty in valuing equity rights. However, com-
pensation remains a key success measure, and thus this measure seems
appropriate.
Because incomes of executives are likely to be positively skewed (in
this study the skewness coefficient for salary was quite high [n= 6.88,
p < .OOl]), a natural logarithmic transformation is suggested as a means
of normalizing the distribution of pay (Gerhart & Milkovich, 1989).
Thus, consistent with standard practice in wage regressions, we trans-
formed the compensation variable by computing its natural log. Num-
ber of promotions was measured on the survey by asking executives to
indicate the total number of promotions (upward changes in job levels)
they received in their career.
Career satisfaction. Career satisfaction was measured with the five-
item scale developed by Greenhaus, Parasuraman, and Wormley (1990),
which appears to be the best measure available in the literature (Ober-
field, 1993). The five items are: (a) I am satisfied with the success I
have achieved in my career; (b) I am satisfied with the progress I have
made toward meeting my overall career goals; (c) I am satisfied with the
progress I have made toward meeting my goals for income; (d) I am satis-
fied with the progress I have made toward meeting my goals for advance-
ment; (e) I am satisfied with the progress I have made toward meeting
my goals for the development of new skills. Greenhaus et al. (1990) re-
ported an acceptable level of internal consistency for this scale (a = .88).
In the present study, the coefficient alpha reliability estimate was 37.
Overalljob satisfaction. Overall or general job satisfaction was mea-
sured with three items. First, the Gallup Poll measure of job satisfaction
was used, where the respondent circles a “yes” or “no” response to the
question, “All things considered, are you satisfied with your job?” Sec-
ond, the single item job-in-general scale was used, which was adapted
by Scarpello and Campbell (1983) from the G.M. Faces Scale, where
the respondent uses a 1 = very dissatisfied to 5 = very satisfed scale in
responding to the question, “How satisfied are you with your job in gen-
eral?” These two measures were used due to their favorable reviews
by Scarpello and Campbell (1983). Finally, an adapted version of the
Fordyce Percent Time Satisfied Item was used (Diener, 1984), in which
respondents are asked to report the percent time they are happy, neutral,
and unhappy with their job on average (only the percent happy figure is
used). To reduce consistency effects, the three job satisfaction measures
were placed in different parts of the survey. Because the three items had
different response formats, they were standardized before computation
of the composite measure. The alpha of this composite measure was 3 5 .
Education. Level of education was taken from the Paul Ray Berndt-
son database, which contained information on the highest degree re-
ceived (coded 0 = bachelor’s degree, 1 = master’s degree or higher).
The database also identified the universities the executives attended.
Thus, we created a variable representing whether the executive’s highest
degree was from an Ivy League school, coded 1 = yes, 0 = no. Dummy
variables were created from the database representing executives’major
fields of study, including business, law, and engineering (other degrees
served as the excluded group in the regressions).
The Gourman Report (Gourman, 1993) is the only guide to higher
education quality that assigns numerical scores measuring university
quality, and has consequently been used by a number of researchers
(e.g., Ehrenberg, 1989; Solmon, 1973). The Gourman Report rates vir-
tually every degree-grantinguniversity in the U.S. on the basis of 18crite-
ria (e.g., qualifications of the faculty members, admission requirements,
curriculum, quality of instruction). Ratings are based on archival data
and interviews or surveys of students, alumni, faculty members, and ad-
ministrators. Each university receives a continuous overall rating that
ranges from 1.00 to 5.00; this rating served as the measure of educational
quality (to take full advantage of the precision of the Gourman rating,
each rating was multiplied by 100 for the analyses). The Gourman rating
was applied to the university from which the executive’s highest degree
was granted, based on the rating of the major in which the degree was
earned.
Work centrality. Job importance/work centrality was assessed using a
measure developed by researchers involved in the Meaning of Working
(MOW) project (MOW International Research Team, 1987). Work cen-
trality is measured by asking the respondent to assign 100 points to five
different life domains (work, family, religion, leisure, and community).
Most of the research on this scale has been conducted cross-culturally,
and, due to its ipsativity, internal consistency estimates of reliability are
inappropriate in evaluating the measure. However, research on U.S.
samples has indicated that the measure has high test-retest reliabilities
(Claes & Quintanilla, 1992) and is correlated with related measures such
as job involvement (MOW International Research Team, 1987).
Other variables. Hours worked per week, hours spent on dependent
care, whether the executive’s spouse was currently employed (coded 1
= yes, 0 = no), number of evenings worked per month, and number of
hours per week the executive wished to work, were assessed with spe-
cific questions on the survey. Organizational success was measured by
asking executives to respond to the question “How successfulwould you
say your organization has been in reaching its strategic goals during the
last two years?” with a percentage estimate (0% to 100%). Consistent
with past research (Howard & Bray, 1988; Judge & Locke, 1993), ambi-
tion was defined as the number of levels executiveswished to advance in
their organization (“HOWmany levels do you want to move up from your
current position?”). The followingvariables were collected from infor-
mation contained in the search firm’s database: marital status (coded 1
= married, 0 = otherwise), age, race (coded 1 = White, 0 = other), sex
(coded 1 = male, 0 = female), whether the stock of the company for
which executive worked was publicly traded (coded 1 = yes, 0 = no), in-
dustry in which the executive worked, region of the country in which the
executive currently worked, whether the executive occupied a position
on an external board of directors (coded 1 = yes, 0 = no), years of job
and occupational tenure, and international experience (coded 1 = yes,
0 = no). Also, the data base contained information on number of em-
ployees working in the executive’s organization. Due to the large num-
ber cases with missing data on this variable, n = 80, missing values were
coded to the mean; dropping cases which had missing values instead of
recoding them had no effect on the coefficient estimates. Associates of
Paul Ray Berndtson, whose job is to evaluate and place executives in new
organizations,rated the level of accomplishment of the executive using a
single item three point scale (3 = marginal, 4 =good, 5 = excellent).This
rating was based on interviews of the candidate, which focused on their

past accomplishments,current skills, and future plans and potential.
Analyses
When multiple criterion variables are regressed on a single set of

predictors, the error terms associated with the different equations often
are correlated (Johnson & Wichern, 1992, p. 314). Because correlations
between the error terms violate an assumption of ordinary least squares
(OLS)regression (Greene, 1993, p. 143), it was important to ascertain
the level of error correlation before proceeding with OLS regression.
The Bartlett test of sphericity estimates the degree to which the error
terms are correlated; a significant coefficient suggests significant inter-
correlations among the error terms (Johnson & Wichern, 1992). In the
present study, the Bartlett coefficient was highly significant (p < .001),
indicating that the error terms were significantly correlated. To control
for the relationships among the error terms, and therefore to predict the
set of criterion variables more accurately and efficiently,we used multi-
variate multiple regression (Johnson & Wichern, 1992, p. 314), which is
a method of analysis that controls for the relationships among the error
terms of the dependent variables. In SPSS, this is accomplishedby using
the MANOVA multivariate command (SPSS Inc., 1990, p. 383). Results
from multivariate regression analysis are interpreted in the same way
as they are using ordinary least squares regression. Because hierarchi-
cal regression is not possible with the multivariate regression module in
SPSS, changes in R2values were computed using SPSS by removing each
bloc of variables from the full OLS equation, and testing the decrease in
R2 for significance. For the industry and region variables, those cases
with no industry or region specified were treated as the excluded groups
for the regression analyses.
Recently, &hen (1994) and Schmidt (1994) have persuasively ar-
gued against the use of statisticalsignificance testing in psychologicalre-
search. Their criticisms of statistical significance testing (many of which
are interrelated) include the fact that significance testing ignores effect
sizes, it leads to ignorance of actual (vs. presumed) error rates, and it
ignores (and thus leads to increases in) Type I1 error. Because of these
problems inherent in statistical significance testing, erroneous conclu-
sions are often reached in data interpretation. The alternative to sta-
tistical significance testing recommended by both Cohen and Schmidt is
to draw confidence intervals around point estimates. Accordingly, 90%
confidence intervals are drawn around the estimated effects of the inde-
pendent variables on career outcomes. Also reported are the lower and
upper Iimits of the confidence intervals.
Results
Table 1 contains the means, standard deviations, and intercorrela-

tions of the individual variables used in the analyses. The multivariate
regressions predicting objective career success (compensation, number
of promotions) are provided in Table 2. As the table indicates, each set
of hypothesized variables (demographic, human capital, motivational,
organizational, and industxyhegion) explained a significant amount of
variance in pay. For most of the specific variables within each bloc, the
confidence intervals around the effect sizes did not include zero.
The demographicvariables of age, gender, marital status, and spouse
employment all predicted compensation level; executives who were old-
er, male, married, and whose spouse did not work outside the home
earned higher salaries than other executives. For the human capital vari-
ables, executives who earned their degree in business or in law, who had
a graduate degree, and who earned their degree from an Ivy League or
high-quality university, and who were evaluated as high in job and ca-
reer accomplishments, earned more money than other executives. Each
additional point in educational quality as measured by the 1.00 to 5.00
scale of the Gourman Report was associated with a predicted increase
in cash compensation of $2,291 per year. This finding is consistent with
Ehrenberg (1989), who found that a one point increase in the Gour-
man ratings for law schools led to a $1,500 increase in starting salary
for lawyers. Executives who graduated from an Ivy League university
earned $30,929 more per year than other executives, controlling for the
quality of the university and the type of degree held. The pay advan-
tages for those with business and law degrees were $5,116 and $30,328,
respectively. Finally, the confidence intervals for all the motivational
variables excluded zero. More hours per week and evenings per month
worked was associatedwith higher levels of pay. Executiveswho desired
to work more hours per week, who had high ambitions for advancement,
and whose work was a central part of their lives, earned more than other
executives.
All of the organizational variables and several of the industryhegion
variables predicted executive pay. Executives who worked in small com-
panies, in organizations perceived as successful, or in companies whose
stock was publicly traded, earned higher salaries than executives who
worked in small or unsuccessful or private firms. Finally, executiveswho
worked in the consumer durable goods industry earned higher salaries
while those who worked in the health care industry and in the South
earned lower saIaries.
In order to illustrate the practical effects of the predictors of com-
pensation, Table 3 provides the estimated effects for realistic levels of
TABLE 1
Means (M), Standard Deviations (SD),and Intercorrelationsof Study Variables
Variable A4 SD 1 2 3 4 5 6 7 8 9 1011121314
1. Log cash comp. 11.62 SO -
2. # of promotions 6.39 3.52 261 -
3. Job satisfaction .02 2.63 062 052 -
4. Career satisfaction 23.72 6.03 192 150 482 -
5. Age 45.47 7.32 230 274 004-001 -
6. Male .93 .26 196 129 011-013 187 -
7. Married .91 .29 112 079 002 027 173 223 -
8. Spouse employed .43 SO-172-070 005-012-082-109 266 -
9. Time/child care 4.93 8.75 -095 -123 -019 -068 -260 -180 013 048 -
10. White .97 .16 053 060-062-078 120 102 030-080-032 -
11. Bd. of dir. mem. .01 .12 108 066 -03 045 103 029 035 -044 -022 019 -
12. Grad. degree .30 .46 105 029 009 020 119 042 007 063-080 013 063 -
13. Qual. highest deg. 261.08 216.92 139 041 016 067-028 024 016-052-013-010 063 073 -
14. Ivy League grad. .09 .28 164 007 053 018 -020 -063 -046 004 011 -040 078 064 254 -
15. Business degree SO S O 110 063 005-005-057 092 036 019 038 044 009 142 090 062
16. Law degree .02 .15 051 -053 052 025 OOO -029 -009 -03 -033 029 033 -051 -060-018
17. Engineering degree .16 .37 -037 029 -033 042 119 108 025 -052 -062 -013 -029 -018 186 -069
18. Job tenure 3.10 2.53 073-050 052 031 171 027-007 054-019 030-026 060-044 026
19. Occupationaltenure 19.92 8.06 180 278 049-003 514 169 120-064-180 089 023 050 015 018
20. Internat’l experience .44 .50 076 154 003 056 115 081 042-050-052-019 008 089 001 049
21. Accomp. rating 4.73 .46 139 063 036 081 -016 -038 -037 -024 -051 020 -012 003 014 023
22. Ambition 1.30 .98 192 109 -133 -205 -259 045 -029 -002 075 - 0 2 -127 -106 -043 -063
23. Eveningswrkedmo. 4.80 4.25 155 138 090 061 041 068 016 -081 -062 031 022-036 004 002
24. Hours w k e d w k . 55.67 8.87 031 061 -063 024 060 094 053 -062 -057 -035 -026 OOO 001 -060
25. Hours wrk. desired 49.02 8.05 162 095 034 034 -076 045 -009 -072 -047 -027 075 020 022 038
26. Work centralitv 38.53 14.72 166 073 086-008-060 101 -002-064 015 042 065 043 115 062
27. # of emps. 11690.00 20886.08 121 051 -018 -011 066 038 -172 -106 -160 -014 009 062 007 041
28. Org. success 65.67 26.08 -013 029 000 -027 016 020 034 -015 4HM -028 -013 M)1 019 035
29. Public firm .12 .33 113 063 325 180 041 054 019-048-030 022 028-028-053 012
30. Cons. dur. gds. ind. .04 .20 127 085 -014 -007 040 023 -019 -054 -058 -042 036 023 028 -027
31. Entertnmt ind. .07 .26 062 021 038 051 096 054 014-055-008-021 018 027 004-023
32. F o o ~ v ind.
. .15 .36 035 027-007-013-003 005 043-005 061 023 003-027 022-013
33. Health care ind. .17 .38 067 024 022 029 -024 -003 -070 -034 015 024 OOO 026 -051 -006
34. High tech. ind. .25 .43 -167 -046 035 007 039 -104 -077 051 -014 013 -050 -003 -003 018
35. Indus. mfg.ind. .05 .22 -051 -037 048 -033 -041 005 026 013 -026 -039 -026 -004 -007 -019
36. Non-profit ind. .14 .34 059-105-023-030-126-035 016 039 073 002 079 028 042 051
37. Petrol. ind. .07 .25 -018 014 -005 -021 -041 OOO 037 039 014 020 042 -036 -028 -019
38. M.W. region .13 .34 -035 -006 006 -033 -021 044 006 -033 048 008 -042 -084 -012 -022
39. N.E. region .16 .37 108 -028 -016 -032 -035 002 -014 018 428 011 021 -079 062 164
40. S. region .22 .41-144 -012 006 -032 077 021 058 -004 -025 -020 -019 032 -046 -078
41. W. region .28 .45 043 OOO 044 042 OOO -036 003 -016 029 -009 -013 033 -011 -030
Note: Decimals are omitted from correlations; N = 1,012.
the variables in Table 2 whose confidence interval did not include zero.
For the dummy variables, effect sizes were provided by the raw regres-
sion coefficient obtained from an equation predicting the unlogged mea-
sure of compensation. For the continuous variables, effect sizes were
computed by multiplying the coefficient estimate by the standard devi-
ation of the variable. The values presented in the table show that the
variables have appreciable effects on compensation earned per year.
These effects ranged from $3,855/year for hours worked per week to
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
-
-199 -
-539-079 -
012 021-033 -
002002090159 -
002-016 065 006 061 -
029-007-040 028-005 000 -
-030-010 056-054-192-020 003 -
014 012 OOO-008 001 040 036 004 -
-070 008 198-050 047 118-044-002 061 -
094 031-089-027-046 000 040 057 293 023 -
083 024-058 028-055-018 023 043 114 024 488 -
053 026-038 018 063-007 035 020 111 001 245 114 -
-045-010 014-011 029 014 028 039 027 017 015 008-012 -
-014 028 002 096 090-059-004-057 134-043 047 055 024-007 -
029 042 027 011 039-001 017 026 028 052 040 040 045 236-001 -
-130-036 091 020 071 080 035-022-049-124-057-031-033 028-026 036 -
108-023-095 061 025-014 008-052-028-159 016 007-019 001 030 002-058 -
062 025-076 041-034-053 006 036 045-249 045 021 019-034 090 077-091-116 -
-073 003-005-009-006-030-005-048 032-261-032-032 059422027-136-095-121-15€l -
-040 066 018 028-039 012-063 037-063-137-012 000-042 045 028 044-050-064-100-104 -
144-019-114 031-085-099 056 054-036-238-030 036 017-010-008-034-086-111-173-181495 -
001-020-077-005 000-060 004-042-002-152-006-051-050 001-016 048-055-070-110-115-061-105 -
-003 009 060-016 000-010-067 064 045-061-033 007-052 034 097-038-048 019 002 056 091-019-058 -
036 020-049 000-015 059 018 001 004-054027 025 067 010-032 027-093-003 049-052-043 104 040-162 -
-048 006 035 063 036 094 040 038-016-033-003 013-013-033 011-047 256-016-118 083 043-036-042-198-237 -
043 022-052 001 001-063-057-045-001-018 020 045 007 039-033 018-113 043 087 000 008-019-011-233-278-340
$54,19S/year for working in the consumer durable goods industry. In

part, the considerable pay advantage enjoyed by executives working in
the consumer durable goods industry is due to the fact that in the re-
gression, their salary (as with all the industry variables) is compared to
executives where no industry was specified. The average salary for ex-
ecutives in this latter group was only $109,434. Thus, executives in the
consumer durable goods industry enjoy twice the pay advantage over the
excluded group (no industry specified) as they do over the average of all
TABLE 2
Multivariate Regressions Predicting Objective Career Success
Log cash Number of
compensation promotions
Predictor B CIL ~
CIu B CIL CIu
Demomaphic variables
Age +.010* +.006 +.014 +.079* +.048 +.lo9
Male +.248* +.145 +.351 +.388 -.390 +1.165
Married +.218* +.128 +.307 +.306 -.368 +.981
Spouse employed -.156* -.205 -.lo7 -.143 -S15 +229
Time devoted to dependent care +.001 -.002 +.004 -.OW -.030 +.011
White -.007 -.146 +.132 +.491 -.557 +1.539
Change in R 2 .067* .024*
Human capital variables
Board of directors position +.062 -.133 +258 +.788 -.688 +2.264
Graduate degree +.062* +.012 +.112 -.133 -.510 +.244
Quality of highest degree +.001* +.om +.ooO +.OOo -.001 +.001
Ivy League graduate +.zoo* +.120 +.280 -.042 -.646 +.561
Business degree +.033 -.029 + .094 +.321 -.143 +.785
Law degree +.187* +.044 +.331 -.987 -2.071 +.o98
Engineering degree -.044 -.124 + .035 +.032 -.570 +.634
Job tenure +.006 -.005 +.016 -.166* -.244 -.OM
Occupational tenure +.002 -.001 +.006 +.080* +.054 +.lo5
International experience +.040 -.007 +.088 +.853* +.495 +1.211
Accomplishment rating +.158* +.lo7 + 209 +.528* +.144 +.912
Change in R2 .060* .052*
Motivational variables
Ambition +.068* +.043 +.092 +.126* +.059 +.311
Evenings worked per month +.011* +.005 +.016 +.078* +.034 +.121
Hours worked per week +.003* +.OW + .007 +.019 -.m +.044
Hours of work desired +.006* +.002 + .009 +.028* +.ME +.054
Work centrality +.002* +.001 + .004 +.ME -.010 +.015
Change in R Z .041* .021*
Organizkional variables
Number of employees in firm -.OW* -.OOo -.OOo +.OOO +.OOO +.OOO
Oreanization success +.001* +.001 +.002 +.005 -.002 +.012
Pubylic firm +.135* +.064 + 207 +.598* +.057 +1.139
Change in R 2 ,015. .004
Industdregion variables
Consumer durable goods industry +.230* +.080 +.381 -.476 -1.612 +.659
Entertainmentfleisure industrv +.049 -.082 +.181 +.083 -1.076 +.910
Food and beverage industry +.070 -.044 +.184 -.112 -.973 +.750
Health care industry -.177" -228 -.005 -.653 -1.495 +.189
High technology industry +.019 -.088 +.126 -.425 -1.233 +.383
Industrial manufacturing ind. +.230 -.168, +.lo9 -.571 -1.613, +.472
Non-profit industry +.112 -.003, +.228 -.941* -1.814, -B68
Petroleum industry -.008 -.141, +.125 -.080 -1.086, +.926
Midwest region -.050 -.136, +.036 -.451 -1.099, +.198
Northeast region +.051 -.027, +.129 -.680* -1.268, -.@2
South region -.203* -275, -.130 -.502 -1.050, +.046
West region -.008 -.075, +.059 -.418 -.925, +.OH
Change in RZ ,037' .012
Constant +9.327* +8.933, +9.722 -4.571* -7.550, -1.591
R .557* .425*
RZ .310* .181*
Adjusted RZ .B5 ,151
Note: Entries are unstandarized coefficients; CIL = Lower limit of 90% codidence
interval; CIu = Upper limit of 90% confidence interval; N = 1,057.
* Indicates coefficient estimates whose confidence interval does not include zero and
R/R2 values significantly different from zero; incremental RZ values were taken from
ordinary least squares(0LS) estimations.
TIMOTHY A. JUDGE ETAL. 505
TABLE 3
Effect Sizes for Significant Independent Variables
Predictkg Cornpenration Levels
Variable Change in level Effect size
DernomDhic variables
Age 7 years $10,262
Male no-yes $ 6,575
Married no-yes $27,845
Spouse employed nohyes -$22,011
Graduate degree no-yes $7,488
Quality of highest degree 2.17 points $4,581
Ivy League graduate no-yes $30,929
Law degree no-yes $30,328
Accomplishmentrating .5 rating $11,816
Ambition 1 level $9,238
Eveningsworked per month 4 evenings $3,855
Hours of work desired 8 hours $8,624
Work centrality 15 points $4,545
Organizationalvariables
Number of emps. in firm 20,886 emps. 43,046
Organization success 26% $ 5,306
Public firm no-yes $19,831
Industwhenion variables
Consumer durable goods industry no-yes $54,195
Health care industry no-yes $4,738
South region no-yes -$24,452
Note: Effect sizes are predicted changes in earningdyear as a result of specified change
in the independent variable.
executive salaries. The coefficients of many of the nonsignificant vari-

ables in the compensation equation revealed substantial effect sizes. For
example, serving on an external board of directors was associated with
a predicted pay increase of $41,772/year and the predicted pay advan-
tage enjoyed by Whites was $7,689/year. However, these and other dif-
ferences were not statistically reliable-the confidence interval around
these variables included zero.
As a whole, the variables predicted number of promotions similarly
to how they predicted compensation, although their effects were some-
what weaker in magnitude. This replicates findings from previous re-
search on lower-level managers (Whitely et al., 1991). Three of the five
individual sets of variables explained a significant amount of the vari-
ance in number of promotions: demographic, human capital, and moti-
vational. Within the set of demographic variables, only the confidence
interval around age did not include zero, indicating that older execu-
tives had achieved more promotions in their careers than had younger
executives. In terms of the human capital variables, executives who had
international experience, had a high degree of occupational tenure, or
were rated as high in their accomplishments earned more promotions.
Contrary to expectations,job tenure negatively predicted number of pro-
motions. Although nearly all the educational variables predicted cash
compensation, they only weakly predicted number of promotions. As
in predicting compensation, the confidence intervals around most of the
motivational variables in predicting number of promotions did not in-
clude zero. Promotion ambition, evenings worked per month, and de-
sired hours worked per week, were associated with more promotions.
Also similar to the variables predicting compensation, several organi-
zational and industryhegion variables predicted number of promotions.
Executives who worked in organizationswhose stock was publicly traded
earned more promotions than executives who worked in private compa-
nies. Executives who worked in the non-profit sector, and those who
worked in the Northeast region, earned fewer promotions in their ca-
reers.
Table 4 displays the results of the regressions predicting subjective ca-
reer success. As the table indicates, the variables that predicted job sat-
isfaction tended to be different from those that predicted career satisfac-
tion. Specifically,motivational and organizational variables explained a
significant amount of variance in job satisfaction whereas objective ca-
reer success, and demographic, human capital, motivational, and orga-
nizational variables explained a significant amount of variance in career
satisfaction. Overall, the total variance explained in the job and career
satisfaction equations was significant and comparable in magnitude.
In terms of the specific coefficient estimates in the job satisfaction
equation, neither objective career success nor any of the human capi-
tal characteristics predicted job satisfaction, with the exception of ed-
ucational quality, which positively predicted job satisfaction. In terms
of the demographic attributes, White executives were less satisfied with
their jobs than minority executives. Contrary to expectations, the moti-
vational variables eveningsworked per month and hours of work desired
positively predicted job satisfaction. Ambition negatively predicted job
satisfaction. Organization success was the only organizational variable
to predict (positively) job satisfaction. Finally, the confidence intervals
for all of the industryhegion variables included zero.
A number of variables predicted career satisfaction. Both pay and
promotions positively predicted career satisfaction. In terms of demo-
graphic variables, older and White executives, and those who devoted
more time to dependent care, reported lower levels of career satisfaction
TABLE 4
Multivariate Regressions Predicting Subjective Career Success
Job satisfaction Career satisfaction
Predictor B CIL CI,y B CIL CIu
Obiective successvariables
Log cash compensation +.019 -.020 +.058 +1.651* +.966 +2.336
Number of promotions -.087 -.382 +.208 +.208* +.118 + .298
Change in R~ ,001 .031
Demographicvariables
Age -.018 -.041 +.005 -.097* -.151 -.043
Male +.158 -.434 +.750 -1.024 -2.3% +.348
Married -.077 -591 +.437 +.649 -.543 + 1.842
Spouse employed +.091 -.194 +.377 +.056 -.607 +.719
Time devoted to dep. care 3 I@. - -.018 +.013 -.045* -.081 -.010
White -1.152* - 1.937 -.366 -2.825. -4.647 -1.002
Change in R~ ,008 .019*
Bd. of directors position -.637 -1.811 +.537 C.792 -1.930 +3.515
Graduate degree +.030 -.253 +.313 -.170 -.826 + .487
Quality of highest degree +.001* +.001 +.001 +.001* +.001 +.003
Ivy League degree +.269 -.190 +.729 -S32 -1.597 +.534
Business degree +.044 -.306 +.393 +.376 -.434 +1.187
Law degree +.741 -.076 1.559 +1.298 -.598 +3.194
Engineering degree -.129 -.582 +.324 +1.170* +.119 +2.220
Job tenure +.OH -.041 +.078 +.116 -.021 +.253
Occupational tenure +.008 -.011 +.028 -.057* -.lo2 -.012
International experience +.097 -.175 +.369 +.557 -.074 +1.187
Accomplishment rating +.227 -.067 +.520 +.718* +.038 +1.399
Change in fi2 ,007 .017*
Ambition -.318' -.459 -.177 -1.121. -1.448 -.793
Eveningsworked per mo. +.037* +.004 +.070 +.003 -.073 +.079
Hours worked per week -.008 -.027 +.011 +.019 -.025 +.063
Hours of work desired +.027* +.008 +.047 -.042 -.087 +.002
Work centrality -.007 -.017 +.003 -.013 -.035 +.010
Change in R~ .020* .031*
Organizationalvariables
Number of emp. in firm +.001* +.001 +.DO1 +.001* +.001 +.001
Organization success +.031* +.026 +.OM +.038* +.027 +.050
Public firm -.115 -.527 +.297 -.604 -1.559 +.352
Change in Rz ,093, .OB*
Industrylregion variables
Cons.durable goods ind. +.643 -.222 +1.507 +2.019* +.014 +4.023
Entertainmenfleisure ind. -322 -1.077 +.433 -.191 -1.942 + 1.560
Food and beverage ind. -.128 -.788 +.533 +.687 -.845 +2.220
Health care ind. +.155 -.491 +.a02 +1.045 -.455 +2.544
High technology ind. -.302 -.927 +.323 +.579 -.a70 +2.028
Industrial mfg, ind. +.393 -.411 +1.197 -254 -2.118 +1.611
Non-profit industry -.205 -.878 +.469 +.142 -1.420 +1.705
Petroleum industry -.090 -361 +.680 - . a 2 -1.869 +1.706
Midwest region -.070 -560 +.419 -.734 -1.869 +.401
Table 4 (continued)
Job satisfaction Career satisfaction

Predictor B CIL CI u B CIL CIU
Northeast region +.088 -.354 +.SO -.746 -1.771 +.279
South region +.025 -.390 +.441 -.553 -1.517 +.411
West region +.344 -.037 +.725 +.210 -.673 +1.094
Change in R2 .011 .010
Constant -1.132 -4.727 +2.462 +7.631 -.705 +15.968
R .394* .392*
R2 .155* .154*
Adjusted R2 .121 .120
Note: Entries are unstandardized coefficients; CIL = Lower limit of 90% confidence
interval; CI u = Upper limit of 90% confidence interval; N = 1,012.
* indicates coefficient estimates whose confidence interval does not include zero and
R/R2values significantly different from zero; incremental RZ values were taken from
ordinary least squares (OLS) estimations.
than other executives. Several of the human capital attributes predicted

career satisfaction; executives whose terminal degree was in engineer-
ing, who earned their degree from a high-quality university, and who
received a high accomplishment rating from the search firm reported
higher levels of career satisfaction while those with high levels of occu-
pational tenure reported lower levels of career satisfaction. In terms of
the motivationalvariables, ambition negatively predicted career satisfac-
tion. As with job satisfaction, organization success positively predicted
career satisfaction. Finally, executives in the consumer durable goods
industry reported higher levels of career satisfaction than the excluded
group (no industry specified).
Discussion
The overall goal of this study was to investigate more comprehen-

sively what predicts executive career success. Although various limita-
tions in the study (see below) prohibit definitive explanations, and rel-
atively small effect sizes for some of the variables circumscribe the im-
plications of some of the results, this study did reveal several interesting
insights into the predictors of executive career success. The conceptual
model of career success received general support from the results in that
most sets of variables contributed a unique amount of variance in pre-
dicting objective and subjective career success. Several aspects of the
findings deserve further discussion. We begin with the findings regard-
ing objective career success.
Demographic characteristics explained a significant amount of vari-
ance in both dimensions of objective success (particularly with respect
to pay, where demographics explained more variance than any other set
of predictors). This result is consistent with past research, which has re-
ported a similar finding on lower-level employees (Gattiker & Larwood,
1988,1989; Gould & Penley, 1984). After controlling for a wide range
of factors, women and minorities had lower levels of objective success
than White males (the gaps were $6,575 per year and 0.66 promotions
over a career for women and $7,689 per year and 0.60 promotions over
a career for minorities). Although the relative disadvantage in objec-
tive success experienced by minority and female executives is not trivial,
we cannot conclude that it represents discriminationbecause there were
relevant variables we could not include (e.g., personal choices, entry pat-
terns into the labor market), the representation of women and minorities
in our sample was relatively small, and computation of indirect effects
(e.g., the effects of gender and race on success mediated through vari-
ables such as education quantity and quality) also may affect the gaps.
Thus, due to low power and the likelihood of omittedvariables, consider-
able caution must be exercised in interpreting the race and gender gaps.
Without more complete data, the results cannot reasonably be inferred
to represent discrimination.
Motivational and human capital variables also explained a significant
amount of variance in objective career success. Executives who devel-
oped their human capital, and who displayed a desire to get ahead, were
substantially more likely to achieve objective success. The overall im-
portance of human capital and motivational variables, and the notewor-
thy effects of the specific variables within these categories of variables,
suggests how aspiring executives may be more extrinsically successful in
their careers.
An intriguing finding was the effect of promotion ambition on pay
and promotions. Ambitious executivesearned more pay and promotions
in their careers. Interestingly, promotion ambition is positively (but not
strongly) related to job level (T = +.18, p < .OOl). Thus, higher level ex-
ecutives display more promotion ambitions than lower level executives
even though their prospective opportunities may be more limited due to
their high position in the organization. In fact, the effect of promotion
ambition on objective success does not appear to be subject to ceiling ef-
fects-ambition was related to pay and promotions even when expected
advancement (i.e., number of levels executives thought they realistically
could advance) was taken into account.
Especially interesting is the role education played in financial suc-
cess. Quantity of education made a material difference in executive
earnings. Over the course of an average career in our sample (20 years),
the estimated cumulative earnings gap between executiveswith a gradu-
ate degree and those with an undergraduate degree was nearly $150,000.
Perhaps the most interesting and unique findings describe how univer-
sity quality and prestige relate to financial success. At the extremes,
the difference in earnings due to educational quality was substantial.
Executives who obtained their degrees from a university not recom-
mended by the Gourman Report (i.e., those scoring a rating of l),earned
$16,070 less per year than executives who obtained their degrees from
a highly recommended university (i.e., those scoring between 4 and 5
on the Gourman Report). Over the course of a 20-year career, this
could amount to a earnings disadvantage over $320,000. This repre-
sents a unique finding in this study, but it remains for future research
to investigate why educational quality affects compensation level. Some
possible explanations have been suggested before (Useem & Karabel,
1986): High-quality universities teach students more than lower qual-
ity institutions; high-quality educational institutions are more likely to
admit high-quality students in the first place; students are more likely
to make connections and plug into influential networks in high-quality
schools; high-quality schools provide important “signs” or credentials
that organizations use in selection and promotion decisions.
An intriguing result was the very large pay premium enjoyed by grad-
uates from Ivy League universities, particularly because this effect was
observed after controlling for educational quality. The predicted earn-
ings advantage for Ivy League graduates, over the course of a 20-year
career, is more than $600,000. One plausible interpretation of this find-
ing is Useem and Karabel’s (1986) hypothesis that prestigious univer-
sities, besides being more likely to bestow scholastic capital upon their
graduates (which should be captured by educational quality), also pro-
vide graduates with social and cultural capital in the form of personal
contacts, network ties, symbols of prestige, and perhaps even inculca-
tion of the motivation to succeed. Alternatively, this result may be due
to favoritism or bias in favor of prestigious schools (Thelin, 1976). For
whatever reason, the executive labor market attaches a premium to ma-
triculation from an Ivy League university, and this premium is higher
than the quality of the school would dictate. As shown in Table 1,univer-
sity quality and Ivy League status were positively correlated (T = +.25,
p < .Ol), but the correlation was far from unity. Again, it remains for
future research to investigate the relative validity of these different in-
terpretations.
Finally, executives who possessed law degrees earned substantially
higher salaries than executives who possessed degrees in other areas.
We had expected business and engineering degrees to positively predict
salary. In fact, business school graduates did earn a compensation ad-
vantage ($5,116) but this effect size was not distinguishable from zero.
A law degree may put an executive in a different labor market (i.e., cor-
porate attorney market), and because few executives have law degrees
(only 32 in our sample), it is a scarce resource that apparently yields sig-
nificant dividends.
Organization size negatively predicted compensation, but the effect
size was quite small (a 10,000 employee increase in firm size corre-
sponded to a predicted increase in pay of $1,458). This finding is con-
trary to other findings in the literature ( e g , Brown & Medoff, 1989).
One possible explanation of these findings is that the firms in this study
were relatively small by Brown and Medoff's (1989) standards (although
only 12% of executives in our sample worked in firms with less than 500
employees). However, the negative firm size-earnings effect persisted
even when the small organizations (those with less than 500 employees)
were removed from the analysis. Part of the explanation for these in-
congruous findings may be the differences in the samples (executives vs.
broader employee groups). Another explanation may be the extensive
use of control variables in this study (in fact, with no controls, a 10,000
employee increase in firm size corresponded to a slight increase [$840]in
predicted pay). Clearly, these issues should be investigated more thor-
oughly in future research.
Consistent with past research (Dreher & Ash, 1990; Whitely et al.,
1991), financial success was easier to predict than hierarchical success.
Still, few coefficients of particular variables were in opposite directions
across the two equations. Perhaps one reason for the decrement in the
promotions equation is that compensation is a better measure of objec-
tive success than number of promotions because the latter variable is
partly confounded with organizational structure and unmeasured mo-
bility patterns. Also, promotions were measured in terms of job level
changes but not other criteria (raises, increases in responsibilities, etc.),
which may have limited its variation and thus its covariance with other
variables. Because the relationship between pay and promotions is posi-
tive but not overwhelmingly strong (Judge & Bretz, 1994; Whitely et al.,
1991; see also Table l), variables that predict one may not predict the
other. Because past research has found results similar to ours, it would
be interesting for future research to investigate the circumstancesunder
which variables that predict compensation do not predict promotions.
However, several variables did not predict pay, but did predict as-
cendancy. In particular, three types of,experience (international expe-
rience, job, and occupational tenure) predicted promotions but not pay.
The positive relationship between occupational tenure and number of
promotions is not surprising because promotions accrue over time; the
longer an executive is in a career, the more chances for promotion. In
fact, this is what has lead some researchers to construct age or tenure
normed measures of career success (e.g., Judge & Bretz, 1994). In ef-
fect, this is what we have done because age and occupational tenure are
partialled out in estimating the other coefficients. The negative effect of
job tenure on promotions is probably a reflection of being plateaued in
one’s position. It may reflect that one of the important ingredients of as-
cendance is velocity (Stewman, 1988), or how quickly one moves up the
corporate ladder; the longer the job tenure, the slower the movement.
It also may reflect job hopping behavior (Judge & Watanabe, 1995). Fi-
nally, the relationship between international experience and promotions
suggests that global assignments may help aspiring executives climb the
corporate ladder (Kets de Vries & Mead, 1992).
Interestingly, the blocs of variables that explained variance in objec-
tive career success explained similar amounts of variance in career sat-
isfaction. Furthermore, both pay and promotions positively predicted
career satisfaction. These results suggest that, to some degree, the stan-
dards society uses to judge the success of an individual’s career are also
those that executives use to evaluate the success of their own careers.
Thus, career satisfaction of the executives in this sample appears to be
a function of their level of objective success and several frame of refer-
ence variables. Results suggested that the frame of reference variables
operated largely (but not totally) as expected. After controlling for ob-
jective success, frames of reference like age, ambition, time devoted to
dependent care, occupational tenure, and nonminority status negatively
predicted career satisfaction. We have argued that these variables index
career expectations in the sense that older, more ambitious, more senior,
and nonminority executives, who have achieved a particular level of ob-
jective success are more likely to be dissatisfied with their careers than
a younger, less senior or less ambitious, or minority executives achiev-
ing the same level of objective success. However, we should note that
several of the hypothesized variables did not predict career satisfaction,
although a few others (e.g., educational quality, accomplishmentrating)
were positiveb related to career satisfaction. We can only speculate why
we observed these latter results; one explanation is that high-qualityedu-
cational institutionsand career accomplishments bestow enriching qual-
ities on executives ( e g , personal growth) that reach beyond the qualities
which predict objective success.
With respect to job satisfaction, the results suggest that the variables
that predict objective success, and even career satisfaction, are different
from those that predict executives’job satisfaction. Demographic and
human capital variables, which explained more variance in objective ca-
reer success and career satisfaction than the other sets of variables, did
not account for a unique amount of the variance in job satisfaction. Con-
versely, the organizational variables explained more variance in job sat-
isfaction than in any of the other equations. One interpretation of these
unexpected findings is that job and career satisfaction are related but dis-
tinct attitudes subject somewhat different psychological processes. Be-
cause extrinsic success predicted career but not job satisfaction, perhaps
for executives career satisfaction may be more outcome or achievement
oriented whereas job satisfaction is more process oriented. Past accom-
plishments may be more relevant to career satisfactionwhile current or-
ganizational characteristics are more important to executive judgments
of job satisfaction. Thus, althoughjob satisfaction and career satisfaction
are related (as shown in Table 1, the correlation between the two vari-
ables is +.48, p < .Ol), they are associated with different variables, at
least for this sample of executives. Of course, it is possible to overinter-
pret the job satisfaction results. Many of the effect sizes in the job satis-
faction equation could not be distinguished from zero, and the strongest
predictor, organization success, was measured with a common method.
Perhaps the most prudent interpretation of these results is that we have
only begun to model the career-related predictors ofjob and career satis-
faction. Given the modest degree to which the variables predicted these
attitudes, clearly further work is needed, particularly in the area of ca-
reer satisfaction.
Limitations
This study has several limitations. Because executives are pressed

for time, we were forced to limit the length of the survey. This may
have caused several problems. First, some of the predictor variables
were measured with single items (e.g., executive accomplishment, or-
ganization success, ambition), which have unknown reliability and va-
lidity. Because the rating of executive accomplishments was measured
with a single item that was provided by the search firm, we cannot have
strong confidence in the psychometric qualities of this measure. Sec-
ond, limitations on survey length forced us to exclude some potentially
relevant variables such as mentoring and socioeconomicstatus (although
other studies have found little biasing effects from excluding mentoring
[Dreher & Ash, 19901 and home backgrounds [Wolfle, 1973; Solmon,
19731).
Although the present paper examined the main effects of individual
and environmental characteristics on career success, a developing lit-
erature suggests that interactive effects between environmental and in-
dividual characteristics are important to consider (e.g., Olian & Rynes,
1984). Relevant to the present paper, research has suggested that certain
executive characteristics (e.g., functional area, tenure, education) may

be considered more or less valuable to organizations depending on their
strategy (Hitt, Ireland, & Palia, 1982; Olian & Rynes, 1984), past exec-
utives (Smith & White, 1987), and size (Olian & Rynes, 1984). Thus,
the effects of the variables in the present study may be influenced by
the possible interactions between executives and organizations. Future
research could clarify these relationships.
Another potential limitation is that some relationships may be bi-
ased. The survey data were collected after the archival data had been
compiled, so the causal nature of some of the relationships might be
called into question. For example, although we posited that motivation
leads to objective success, it also is possible that success leads to moti-
vation. Furthermore, because some of the variables used to predict job
and career satisfaction were collected from the same survey, it is possi-
ble that common methodvariance inflated some of the relationships. Fi-
nally, it is possible that collection of the objective success data influenced
responses to the survey. Several factors partly mitigate these concerns.
First, priming is not likely with many relationships in the model either
because both variables were collected from archival data (e.g., age and
objective success), or because the archival data were collectedwith a dif-
ferent method (interview) some time before the survey was distributed.
Second, the effects reported in Table 2 do not vary according to whether
the data were collected from the same source. In fact, with respect
to the correlations between the criteria and the predictors, a t test re-
vealed no significant difference between common-method correlations
and different-methodcorrelations (t = -0.56, ns). Third, there were no
significant differences in correlations between two variables measured
at the same point in time versus variables measured at different points
in time (t = +0.38, ns). These factors suggest that the nature of the data
collection has not biased the results, although the possibility cannot be
fully dismissed.
Some discussion of the advantages and limitations of the sample is in
order. It is likely that most executives have a relationship with an exec-
utive search firm; surveys reveal that this is the method through which
most executive-level staffing decisions are made in the U.S. (Magnus,
1989) and abroad (Rock, 1990). Thus, there is no reason to believe
that the source of this sample makes it unrepresentative of the larger
population of executives. Also, the executives in this sample worked in
many different types and sizes of organizations, in many different indus-
tries and regions throughout the U.S. Still, there is very little normative
data on the characteristicsof executives, so the representativenessof our
sample is unknown. Thus, it must be acknowledged that some charac-
teristics of the sample (e.g., compensation figures are lower than those
for most high-level executive positions, 12% of executives in our sam-

ple worked in publicly-held organizations although 4% of corporations
earning more than $l,000,000/year are publicly-held [U.S. Bureau of the
Census, 19941, a higher than expected number of executives worked in
privately-held organizations, executives with few accomplishmentsmay
be less likely to have contacts with a search firm in the first place) may
have influenced the results.
PracticaI Implications
The results suggest a profile of a successful executive. The most ob-

jectively successful executive appears to be one who is a married, middle-
aged, White man whose spouse does not work outside the home, who has
impressive (high quality and prestigious) educational credentials, and
who displays a high commitment to his work. From the perspective of
an individual who aspires to be a “successful” executive, it appears that
educational credentials and high commitment to work pay off. Previ-
ous research has shown that executives report only average levels of life
satisfaction and high levels of stress and work-family conflict (Judge et
al., 1994). Thus, for some executives, objective success is achieved at
some cost. On the other hand, given the comparability in results between
the objective career success and career satisfaction equations, many of
the factors that make executives objectively successful also contribute to
their career satisfaction (including objective success itself). A compara-
ble conclusion, however, cannot be drawn with respect to executive job
satisfaction.
Finally, we should note that although variance explained by the blocs
of variables was relatively small, the practical effect sizes are substantial.
Specifically, with respect to the human capital attributes, an executive
who earned a master’s degree in business from an Ivy League school with
international experience is projected to earn $54,434 more per year than
an executive with no international experience and with an undergradu-
ate degree from a non-Ivy League school. Similarly, an executive whose
accomplishment rating was “5,” with 20 years occupational tenure, and
with international experience, is projected to have earned nearly three
more promotions than an executive who was rated as a “3,”with 10years
of occupational tenure, and with no international experience. Compa-
rable effect sizes can be demonstrated with respect to the demographic,
motivational, and organizational variables. Thus, although incremental
R2is an informative measure of effect sizes, it does have limitations in
estimating practical effects (Champoux & Peters, 1980) which in some
cases were substantial. In interpreting the practical effects of the spe-
cific variables, it is interesting to note that the variables that contributed
to one definition of success are not necessarily the same as those that
contributed to another definition of career success. Thus, these results
suggest that the career preparation strategies of aspiring executives may
depend on the career outcome(s) that is most important to them. Per-
haps an even more fundamental conclusion suggested by these results is
that the attainment of executive career success is a complexphenomenon
that defies simple prescriptions and, due to the importance of the topic,
is deserving of further research that would extend the results presented
in this paper.
REFERENCES
American Psychological Association. (1994). Publication manual of the American Psycho-

logical Association (4th ed.). Washington,D C Author.
Becker G. (1964). Human capital: A theoretical and empirical analysk with special refmnce
to education. New York Columbia University Press.
Bielby DD, Bielby WT. (1988). Women’s and men’s commitment to paid work and family.
In Stromberg AH, Larwood L, Gutek B (Eds.), Women and work (pp. 249-263).
Beverly Hills, CA: Sage.
Bloch FE, Kuskin MS. (1978). Wage determination in the union and nonunion sectors.
Industnizl and Labor Relations Review, 31,183-192.
Bray DW, Howard A. (1980). Career success and life satisfactions of middle-aged man-
agers. In Bond LA, Rosen JC (Eds.), Competence and coping during adulthood
(pp. 258-287). Hanover, NH: University Press of New England.
Bretz RD, Boudreau JW, Judge TA. (1994). Job search behavior of employed managers.
PERSONNEL PSYCHOLOGY, 47,275-301.
Brown C, Medoff J. (1989). The employer size-wage effect. Journal of Political Economy,
97,1027-1059.
Campbell JP, Dunnette MD, Lawler EE 111, Weick KE Jr. (1970). Managerial behuvio&
performance, and eflectiveness. New York McGraw-Hill.
Cannings K, Montmarquette C. (1991). Managerial momentum: A simultaneous model
of the career progress of male and female managers. IndwtrialandLzzborRefations
Review, 44,212-228.
Carlson LA, Swartz C. (1988). The earnings of women and ethnic minorities, 1959-1979.
Industrial and Labor Relations Review, 41,530-546.
Cava A, Mayer D. (1993). Gender discrimination abroad. Business and Economic Review,
40,1516.
Champow JE, Peters WS. (1980). Applications of moderated regression in job design
research. PERSONNEL PSYCHOLOGY, 33,759-783.
Claes R, Quintanilla S A R . (1992). Beginning careers in two occupatwnal groups. Gent,
Belgium: University Gent Press.
Cohen J. (1994). The earth is round (p < .05). American Aychologist, 49,997-1003.
Cox CJ,Cooper CL. (1989). The making of the British CEO. Childhood, work experience,
personality, and management style. Academy of Management Executive, 3,241-245.
Cox TH,Harquail CV.(1991). Career paths and career success in the early career stages
of male and female MBAs. Joumal of Vocational Behavim 39,5675.
Cox TH, Nkomo SM. (1991). A race and gender-group analysis of the early career expe-
rience of MBAs. Workand Occupations, 18,431-446.
Dalton DR, Kesner IF. (1985). Organizational performance as an antecedent of in-

side/outsidechief executivesuccession: An empirical assessment. Academy ofMan-
agement Journal, 28,749-762.
Diener E. (1984). Subjectivewell-being. Psychological Bulletin, 95,542-575.
Dreher GF, Ash RA. (1990). A comparative study of mentoring among men and women
in managerial, professional, and technical positions. Joumaf of Applied Psychoh,
75,539-545.
Dumhoff GW. (1967). Wio rules America? Englewood Cliffs, NJ: Prentice-Hall.
Edstrom A, Galbraith JR. (1977). Transfer of managers as a coordination and control
strategy in multinational organizations. Adminktnative Science Quarter& 22,248-
263.
Ehrenberg RG. (1989). An economic analysis of the market for law school students.
Journalof Legal Education, 39,628-654.
England GW, Whitely W. (1990). Cross-national meanings of working. In Brief M,Nord
W R (Eds.),Meaningsof occupationalwork (pp. 65-106). Lexington,MA: Lexington
Books.
Gattiker UE, Larwood L. (1986). Subjective career success: A study of managers of
support personnel. Journal of Business and Psychology, I , 78-94.
Gattiker UE, Lanvood L. (1988). Predictors for managers' career mobility, success, and
satisfaction. Human Rehtions, 41,569-591.
Gattiker UE, Larwood L (1989). Career success, mobility and extrinsic satisfaction of
corporate managers. Social Science Journal, 26,75-92.
Gerhart BA, Milkovich m.(1989). Salaries, salary growth, and promotions of men and
women in a large, private firm. In Michael RT Hartmann HI, O'Farrell B (Eds.),
?by equity Empirical inquiries (pp, 2343). Washington: National Academy Press.
Gomez-Mejia LR, Balkin DB. (1992). Compensation, organizational sfratw, and @
performance Cincinnati: South-WesternPublishing Co.
Gomez-Mejia LR, Welbourne TM. (1989). Strategic design of executive compensation
programs. In Gomez-Mejia LR, Olsen R, Milkovich G'T (Eds.), Compensationand
benefit3 (pp. 216- 269). Washington: Bureau of National Affairs.
Gould S, Penley LE. (1984). Career strategies and salary progression: A study of their
relationships in a municipal bureaucracy. Organizational Behavior and Human
Performancq 34,244-265.
Gourman J. (1993).The Cowman report. Los Angeles: National Education Standards.
Greenberg J, McCarty CL.(1990).Comparable worth A matter of justice. In Fems GR,
Rowland KM (Eds.),Research in personnel and human resources management (Vol.
8,pp. 265-301). Greenwich, C T JAI Press.
Greene WH. (1993). Econometric analysis. New York Macmillan.
Greenhaus JH, Parasuraman S, Wormley WM.(1990). Effects of race on organizational
experiences,job performance evaluations, and career outcomes. Academy ofMan-
Gutteridge TG. (1973). Predicting career success of graduate business school alumni.
Academy of Management Journal, 16,129-137.
Harrell Tw.(1969). The personality of high earning MBA's in big business. PERSONNEL
psycH0LoGY, 22,457-463.
Haunschild PR. (1993). Interorganizationalimitation: The impact of interlockson corpo-
rate acquisition activity. Adm'nisfrativeScience Quarter&, 38,564-592.
Hitt MA,Ireland RD, Palia KA.(1982). Industrial firms' grand strategy and functional
importance: Moderating effects of technology and uncertainty. Academy of Man-
Hough LM. (1984). Development and evaluation of the “accomplishment record” method
of selecting and promoting professionals. Journal of Applied mchology, 69, 135-
146.
Howard A, Bray D. (1988). Managerial fives in tmnsition: Advancing age and changing
times. New York Guilford Press.
Hulin CL. (1991). Adaptation, persistence, and commitment in organizations. In Dun-
nette MD, Hough LM (Eds.), Handbookof industrial andorganizationalpsycholw:
Vof.2. (2nd ed., pp. 445-505). Palo Alto, C A Consulting Psychologist Press.
Jaskolka G, Beyer JM, Trice HM. (1985). Measuring and predicting managerial success.
Journal of Vocational Behavior; 26, 189-205.
Johnson RA,Wichern DW. (1992). Applied multivariate statistical analysk. Englewood
Cliffs, NJ: Prentice-Hall.
Judge TA, Bretz RD. (1994). Political influence behavior and career success. Journal of
Management, 20,4345.
Judge TA, Boudreau JW, Bretz RD. (1994). Job and life attitudes of male executives.
Journal of Applied Psycholoa, 79 ,767-782.
Judge TA, Locke EA. (1993). Effect of dysfunctionalthought processes on subjectivewell-
being and job satisfaction. Journal ofApplied Psychology, 78,475-490.
Judge TA, Watanabe S. (1995). Is the past prologue?: A test of Ghiselli’shobo syndrome.
Journal of Management, 21,211-229.
Kerr JL, Bettis RA. (1987). Boards of directors, top management compensation and
shareholder returns. Academy of Management Journal,30,745-764.
Kets de Vries MFR, Mead C. (1992). The development of the global leader within the
multinational corporation. In Pucik V, Tichy Nh4, Barnett CK (Eds.), Globalizing
management: Creating and leading the competitive organization (pp. 187-205). New
York Wiley.
Konda SL, Stewman S.(1980). An opportunity labor demand model and Markovianlabor
supply models: Comparativetests in an organization. American SociologicalRevim,
45,276-301.
Korman AK,Wittig-Berman U, Lang D. (1981). Career success and personal failure:
Alienation in professionals and managers. Acudeny of Management Journal, 24,
342-360.
Kotter JP. (1982). The general managers. New York The Free Press.
Locke EA. (1976). The nature and causes of job satisfaction. In Dunnette MD (Ed.),
Handbook of industrial and organizational psychology (pp. 1297-1343). Chicago:
Rand McNally.
London M, Stumpf SA. (1982). Managing careers. Reading, UA: Addison-Wesley.
Magnus M. (1989). Hire spending. Personnel Journal, 68,73-76.
MOW International Research Team. (1987). The meaning of working. London: Academic
Press.
Murphy KJ. (1985). Corporate performance and managerial remuneration. Journal of
Accounting and Statistics, 7,1142.
Oberfield A. (1993,August). A proposed model of career satisfaction. Paper presented at
the 1993 Academy of Management Annual Meetings, Atlanta, GA.
Olian JD, Rynes SL. (1984). Organizational staffing: Integrating practice with strategy.
Industrial Relations, 23,170-183.
Pfeffer J, Ross J. (1982). The effects of marriage and a working wife on occupational and
wage attainment. AdminisfrativeScience Quarter&, 27,66-80.
Pfeffer J. (1983). Organizational demography. In Cummings LL, Staw BM (Eds.), Re-
search in organizationalbehavior (pp. 299-357). Greenwich, C E JAI Press.
Pfeffer J. (1991). Organization theory and structural perspectiveson management. J o m l
of Management, 17,789403.
Psacharopoulos G. (1985). Returns to education: A further international update and

implications. Journal of Human Resources, 20,583-604.
Rock S. (1990).The hunt for the best hunter. Director; 44,7679.
Scarpello V, Campbell JP. (1983). Job satisfaction: Are all the parts there? PERSONNEL
PSYCHOLOGY,36,577400.
Schmidt FL. (1994,August). Quantitativemethods and cumulativeknowledge inpvchology:
Implicationsfor the training of researchers. Paper presented at the Annual Conven-
tion of the American Psychological Association, Los Angeles, CA.
Smith M, White MC. (1987). Strategy, CEO specialization, and succession.Adminktmtive
Science Quarter&, 32,263-280.
Solmon LC. (1973). Schooling and subsequent success: The influence of ability, back-
ground, and formal education. In Solmon LC, nubman PJ (Eds.), Does college
mutter? New York Academic Press.
SPSS Inc. (1990).SPSS reference guide. Chicago: Author.
Stewman S. (1988). Organizational demography. Annual Review of Sociology, 14,173-202.
Stewman S , Konda SL. (1983). Careers and organizational labor markets: Demographic
models of organizational behavior. American Jownal of Sociolw, 88,637485.
Swinyard AW, Bond FA. (1980). Who gets promoted? Ha~vardBusinessReview, (Septem-
ber-October), 6-18.
Thelin JR. (1976). The cultivationof Ivy: A saga of the college in America. Cambridge, MA:
Schenkman.
Tsui AS, Gutek BA. (1984). A role analysis of gender differences in performance, affec-
tive relationships, and career success of industrial middle managers. Academy of
Management Journal, 27,619435.
U.S. Bureau of the Census. (1994). Statistical abstract of the United States (114th ed.).
Washington, D C Author.
Useem M, Karabel J. (1986). Pathways to top corporate management. American Socio-
logical Review, 51,186200.
Warner WL, Abegglen JC. (1955).Big business leaders in America. New York Harper and
Row.
Whitely W, Dougherty TW,Dreher GE (1991). Relationship of career mentoring and so-
cioeconomicorigin to Managers’and professionals’early career progress. Acudewy
of Management Journal, 34,331-351.
Wolfle D. (1973). To what extent do monetary returns to education vary with family
background, mental ability, and school quality? In Solmon LC, Taubman PJ (Eds.),
Does college mutter? New York Academic Press.

An Empirical Investigation Career Success PDF

Uploaded by

Copyright:

Available Formats

An Empirical Investigation Career Success PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

An Empirical Investigation Career Success PDF

Uploaded by

Copyright:

Available Formats

PERSONNEL PSYCHOLOGY

AN EMPIRICAL INVESTIGATION OF THE

This study examined the degree to which demographic, human capital,

What factors lead some executives to be more successful in their ca-

can be enhanced in several ways. First, researchers have predicted ca-

Conceptual Model of Executive Career Success

Objective Career Success Subjective Career Success

from educational attainment in terms of pay and promotions are sig-

human capital acquired (Thelin, 1976), we predict that controlling for

Sample and Procedures

Subjects were executives contained in the database of Paul Ray

salary, promotions, experience, marital status, age, race, sex, industry,

Objective career success. Information on annual salary-as well as

rating was based on interviews of the candidate, which focused on their

When multiple criterion variables are regressed on a single set of

Table 1 contains the means, standard deviations, and intercorrela-

$54,19S/year for working in the consumer durable goods industry. In

Variable Change in level Effect size

executive salaries. The coefficients of many of the nonsignificant vari-

Job satisfaction Career satisfaction

than other executives. Several of the human capital attributes predicted

The overall goal of this study was to investigate more comprehen-

This study has several limitations. Because executives are pressed

executive characteristics (e.g., functional area, tenure, education) may

for most high-level executive positions, 12% of executives in our sam-

The results suggest a profile of a successful executive. The most ob-

American Psychological Association. (1994). Publication manual of the American Psycho-

Dalton DR, Kesner IF. (1985). Organizational performance as an antecedent of in-

Psacharopoulos G. (1985). Returns to education: A further international update and

You might also like