Page 49 - Psychological Management of Individual Performance

28 ability and non-ability predictors of job performance
INTRODUCTION

The prediction of job performance involves the application of psychological theory
from basic domains of psychology (e.g., intelligence, personality, motivation, emotion)
to knowledge about job conditions and requirements. In conducting research on the
predictive validity of various ability and non-ability factors, industrial/organizational
(I/O) psychologists seek to answer two fundamental questions: (1) What role do person
factors play in determining job performance? (2) How may theories and tests be used
to improve predictions of person–job fit? Answers to these questions have theoretical
implications for the development of theories of work behavior and job performance,
as well as practical implications for the development of effective personnel selection,
training, and placement systems in organizational settings.
The purpose of this chapter is to review significant advances in theory and research on
general cognitive ability (also referred to in the literature as intelligence or general mental
ability) and non-ability predictors of individual differences in job performance over the
past 15 years. A comprehensive, historical account of progress in the cognitive and non-
ability domains is beyond the scope of the present chapter (for historical reviews, see,
e.g., Kanfer, Ackerman, Murtha, & Goff, 1995; Murphy, 1996) in this chapter we focus
on recent developments in order to provide a more complete account of the progress and
research trends that have occurred in the field since the mid-1980s.
Prior to the mid-1980s, studies of ability and non-ability predictors of job perfor-
mance proceeded largely independently. In the ability domain, theoretical and empirical
advances in the construction of tests of intelligence (e.g., Binet & Simon, 1905) provided
the foundation for a substantial number of studies demonstrating the validity of general
intelligence tests for predicting applicant selection and occupational success (cf. Viteles,
1930). In addition, researchers (e.g., Kelly, 1928; Thomson, 1916) refined a second
“group factors” approach. This approach utilized specialized aptitude batteries to test
group factors (e.g., spatial, verbal ability) shown to be job related, culminating in the
kinds of multiple-aptitude batteries used by the US Army Air Forces during World War II
(Flanagan, 1948). Although evidence demonstrating the validity of these two approaches
in predicting job performance mounted through the mid-1980s, concerns about the lack
of evidence demonstrating the broad validity of general intelligence tests for predicting
job performance precluded their widespread use in civilian settings.
In contrast, in the non-ability domain the absence of a broad organizing framework for
non-cognitive constructs led to disparate research programs on the role of non-ability fac-
tors. During the early part of the twentieth century, personality tests were frequently used
to identify and reject maladjusted individuals (Ellis & Conrad, 1948). However, reviews
of the empirical literature investigating the predictive validity of personality measures
were pessimistic and suggested a general failure to achieve consensus on the valid-
ity of such measures for predicting job performance (e.g., Crutchfield, Woodworth, &
Albrecht, 1958; Guion & Gottier, 1965).
During the past 15 years, however, a number of studies have been conducted using va-
lidity generalization methods (e.g., meta-analysis, Hunter, Schmidt, & Jackson, 1982). In
these studies, researchers aggregate prior empirical findings to determine the estimated
true-score relations between ability and performance after correcting for statistical arti-
facts (e.g., predictor unreliability, sampling error). Results of meta-analytic studies show
non-trivial relations between general cognitive ability and job performance, and have led