THE CORE SELF-EVALUATIONS SCALE: DEVELOPMENT OF A MEASURE
PERSONNEL PSYCHOLOGY
2003,56,303-331
THE CORE SELF-EVALUATIONS SCALE:
DEVELOPMENT OF A MEASURE
TIMOTHY A. JUDGE, AMIR EREZ
Department of Management
University of Florida
JOYCE E. BONO
Department of Psychology
University of Minnesota
CARL J. THORESEN
Department of Psychology
Tulane University
Despite an emerging body of research on a personality trait termed
core self-evaluations, the trait continues to be measured indirectly.
The present study reported the results of a series of studies that devel-
oped and tested the validity of the Core Self-Evaluations Scale (CSES),
a direct and relatively
ief measure of the trait. Results indicated that
the 12-item CSES was reliable, displayed a unitary factor structure,
co
elated significantly with job satisfaction, job performance, and life
satisfaction, and had validity equal to that of an optimal weighting of
the 4 specific core traits (self-esteem, generalized self-efficacy, neuroti-
cism, and locus of control), and incremental validity over the 5-factor
model. Overall, results suggest that the CSES is a valid measure that
should prove useful in applied psychology research.
A line of research has developed that suggests that a
oad person-
ality trait, termed core self-evaluations, is a significant predictor of job
satisfaction and job performance. Judge, Locke, and Durham XXXXXXXXXXin-
troduced the concept of core self-evaluations in an effort to provide a
trait that would be a useful predictor of job satisfaction, as well as, per-
haps, other applied criteria. According to Judge et al. (1997), core self-
evaluations is a
oad, latent, higher-order trait indicated by four well
established traits in the personality literature: (a) self-esteem, the over-
all value that one places on oneself as a person (Harter, 1990); (b) gen-
eralized selfeficacy, an evaluation of how well one can perform across
a variety of situations (Locke, McClear, & Knight, 1996); (c) Neuroti-
cism, the tendency to have a negativistic cognitive/explanatory style and
to focus on negative aspects of the self (Watson, 2000); and (d) locus of
Co
espondence and requests for reprints should be addressed to Timothy A. Judge,
Department of Management, Wa
ington College of Business, 21 1D Stuzin Hall, Univer-
sity of Florida, Gainesville, FL 32611; XXXXXXXXXX.
COPYRIGHT 8 2W3 PERSONNEL PSYCHOLOGY, INC.
303
304 PERSONNEL PSYCHOLOGY
control, beliefs about the causes of events in one’s life-locus is internal
when individuals see events as being contingent on their own behavior
(Rotter, XXXXXXXXXXAs one can gather from the commonality among these
traits, core self-evaluations is a basic, fundamental appraisal of one’s
worthiness, effectiveness, and capability as a person.
The four core self-evaluations traits are some of the more promi-
nent in psychology. Cumulatively, they have been referenced in more
than 50,000 publications (PsycINFO search, October 20,2001). Despite
the salience of these traits, and some strong similarities between them, it
has been relatively uncommon for researchers to study the traits together
(Judge & Bono, 2001a). Even in the relatively rare case when the traits
are studied together in personality research, generally they are treated as
entirely separate variables with no discussion of their inte
elationships
or possible common core (e.g., Abouserie, 1994; Hojat, 1983; Horner,
1996). Similarly, in industrial-organizational(1-0) psychology research,
various pairs of the core traits have been related separately to a variety
of outcomes, such as job performance (Bhagat & Chassie, 1978), career
decision making (Kishor, 19Sl), unemployment (Tiggemann & Wine-
field, 1984), or attributions (Hesketh, XXXXXXXXXXIn a number of studies,
however, Judge and colleagues have found that the four core traits load
on a single factor (Erez & Judge, 2001; Judge, Bono, & Locke, 2000;
Judge, Erez, & Bono, 1998; Judge, Locke, Durham, & Kluger, 1998),
suggesting that it may be appropriate to consider the traits as indicators
of a higher-order latent concept.
Beyond the empirical associations, support for the core self-
evaluations concept can be derived from clear conceptual similarities
among the traits. Pair by pair, the traits share conceptual similarities
(see Judge & Bono, 2001a). It is our argument that the reason these
surface traits share similarities is because they are indicators of a com-
mon core. Because core self-evaluations is a
oad, latent trait that is the
common source of the four (and perhaps other) specific traits, it is the
psychological mechanism that causes these individual traits to be co
e-
lated. Because an individual who scores high on core self-evaluations is
someone who is well adjusted, positive, self-confident, efficacious, and
elieves in his or her own agency, it is this
oad core that is then mani-
fested in high levels of self-esteem, emotional stability, and general self-
efficacy, and an internal locus of control. In short, we believe that high
self-esteem and the other core traits result from a
oad, general, posi-
tive self-regard.
By the same token, we do not contend that the four core traits are
completely redundant. There may be parts of each that are unique
and important. What we are arguing, though, is that there is consider-
able redundancy, and the latent concept of core self-evaluations explains
TIMOTHY A. JUDGE ET AL. 305
this conceptual and empirical redundancy. Latent constructs exist at a
deeper level than their indicators and, in fact, causally influence the indi-
cators or dimensions (Bollen & Lennox, XXXXXXXXXXThus, when an individual
has a positive self-concept, measures of the four core traits are manifes-
tations or indicators of this inner self-concept or core self-evaluation,
and this explains why the four traits are conceptually and empirically re-
lated. Thus, rather than being a multidimensional aggregate construct,
where a composite factor is comprised of dimensions that may or may
not be related, core self-evaluations is a latent psychological construct
ecause it is the “latent commonality underlying the dimensions” (Law,
Wong, & Mobley, 1998, p XXXXXXXXXXIn that way, self-esteem, generalized
self-efficacy, and the other core traits are different ways in which core
self-evaluations is realized.
Not only do the core traits appear to indicate a common factor, it ap-
pears that the concept is related to important work criteria. In two stud-
ies, research by Judge and colleagues has linked core self-evaluations to
job satisfaction and further showed that intrinsic job characteristics me-
diated the relationship (Judge, Locke, et al., 1998; Judge et al., 2000).
Judge et al XXXXXXXXXXfound that individuals with positive self-evaluations
not only perceived their jobs as providing more intrinsic characteristics,
they actually attained more challenging jobs (i.e., jobs with higher lev-
els of job complexity). In a meta-analysis of 169 co
elations, Judge and
Bono (2001b) showed that the relation of four core traits to job satis-
faction generalized across studies. In addition to meta-analyzing the re-
lationship between core self-evaluations and job satisfaction, Judge and
Bono (2001b) also investigated the co
elation between the core traits
and job performance. Analyzing 105 co
elations, these authors found
that the validity of three of the four core traits generalized across stud-
ies. The average validity was identical (ave. p = .23) to the validity of
Conscientiousness ( p = .23; Ba
ick & Mount, XXXXXXXXXXIn addition, Erez
and Judge XXXXXXXXXXshowed that core self-evaluations was related to moti-
vation and performance in two separate studies. In the first study, these
authors demonstrated that the core self-evaluations factor was related
to task motivation and performance in a laboratory setting. In the sec-
ond study, they showed that the core trait was related to task activity,
productivity as measured by sales volume, and the rated performance of
insurance agents.
Despite impressive support for core self-evaluations, one issue that
may interfere with future research is the measurement of the trait. In
contemporary personality research, most traits are measured with rel-
atively short, direct scales. For example, the best known measures of
Conscientiousness measure the trait with scales that include 9 (Benet-
Martinez & John, 1998), 10 (Goldberg, 1999), or 12 (Costa & McCrae,
306 PERSONNEL PSYCHOLOGY
1992) items. In contrast, at present core self-evaluations is measured
indirectly. In such studies (e.g., Judge, Locke, et al., 1998; Judge et al.,
ZOOO), the trait is not measured manifestly but is infe
ed to exist on the
asis of its indicators. As a practical matter, this measurement strategy
has several limitations.
First, the measures are indirect. This means that the core self-
evaluations trait must be extracted by factor analyzing the four scales
that indicate the trait (e.g., Judge, Erez, et al., XXXXXXXXXXA direct measure,
ecause it is designed to precisely measure the underlying concept itself
ather than the indicators of the concept, may be more valid. The indi-
ect measurement approach of past research also leads to confusion over
whether the trait is a latent or aggregate construct (see below). Second,
ecause of this indirect measurement from existing scales, the measure
of core self-evaluations is relatively long. Judge, Locke, et al XXXXXXXXXXand
Judge et al XXXXXXXXXXmeasured core self-evaluations with four scales that
total 38 items. Given the relative
evity of measures of other traits, it
would seem unnecessary to measure core self-evaluations with a com-
ination of scales that, cumulatively, are relatively long. The length of
the indirect measure may limit its usefulness, especially in organizational
settings. Rather than utilizing a lengthy measure, some researchers may
choose to measure only a single indicator (e.g., Neuroticism or Emo-
tional Stability), and thereby miss a substantial amount of valid variance.
A final possible limitation is that of empiricalvalidity. The core traits dis-
play slightly differential relations with criterion variables (e.g., in Judge
& Bono’s [2001b] meta-analysis, Emotional Stability predicted the crite-
ia less well than the other core traits, and the self-esteem-performance
co
elations were highly