A Review of Court Decisions on Cognitive Ability Testing, 1992-2004

• Author: Elizabeth L. Shoenfelt,Leslie C. Pedigo
• Published date: 01 September 2005
• Date: 01 September 2005
• DOI: http://doi.org/10.1177/0734371X05276800
• Subject Matter: Articles

10.1177/0734371X05276800
REVIEW OF PUBLIC PERSONNEL ADMINISTRATION / Sept. 2005
Shoenfelt, Pedigo / COGNITIVE ABILITY TESTING
A Review of Court Decisions on
Cognitive Ability Testing, 1992-2004
ELIZABETH L. SHOENFELT
Western Kentucky University
LESLIE C. PEDIGO
State of Tennessee
General cognitive ability is likely the single best predictor of job performance,
although it typically results in race-based adverse impact. The majority of the 22
cognitive ability testing cases in appellate and district courts from 1992 to 2004
involved class action plaintiffs and civil service jobs. Organizations that used
professionally developed tests that were validated and that set cutoff scores sup-
ported by the validity study fared well in court. The validation study must be
conducted according to professional standards, and the results should be used
properly when setting cutoff scores. Plaintiffs in cognitive ability test–related
lawsuits were likely to be members of minority groups; the majority of cases were
race-based claims. Utilizing a consultant to develop and validate selection tests
may ensure the appropriate expertise and a professional y developed and vali-
dated test; however, it does not alleviate the responsibility of the employer for
adverse impact.
Keywords: employment testing; cognitive ability testing; adverse impact; EEO
court decisions; selection
Duringtheselectionprocess,organizationsattempttopredictfuturejob
performance, often using cognitive ability testing. Many have argued
that general cognitive ability is the single best predictor of job performance
across a wide range of occupational settings (e.g., Gottfredson, 2000, 2002;
Kuncel, Hezlett, & Ones, 2004; Murphy, 2002; Outtz, 2002; Schmidt &
Hunter, 1998, 2004; Wagner, 1997). There are, however, potential draw-
backs to using cognitive ability tests as predictors. The most obvious negative
consequence of cognitive ability testing is that it is likely to result in substan-
tial adverse impact against protected racial and ethnic groups (Murphy, 2002;
Outtz, 2002). Typically, African Americans score one standard deviation
lower than Whites on tests of cognitive ability (e.g., Ceci, 2000; Hunt, 1996;
Review of Public Personnel Administration, Vol. 25, No. 3 September 2005 271-287
DOI: 10.1177/0734371X05276800
© 2005 Sage Publications
271

---

272
REVIEW OF PUBLIC PERSONNEL ADMINISTRATION / Sept. 2005
Hunter & Hunter, 1984; Murphy 2002). Murphy (2002) and Outtz (2002)
noted substantial differences between scores for Hispanics and Whites on
cognitive ability tests, although the differences are not as great as those
between White and African American scores. Consequently, very few mem-
bers of the lower scoring minority groups are selected when organizations uti-
lize cognitive ability tests as predictors. There are gender differences in perfor-
mance on cognitive ability tests. However, Hunt (1996) reported that
differences between male and female scores on cognitive ability tests are
smaller than racial differences and typically are more prevalent when using
tests of mathematical ability, mechanical reasoning, and spatial visual reason-
ing content. The underlying reason for race-based differences in cognitive
ability (i.e., whether the gap is due to heredity, environment, or the interac-
tion of the two) continues to be vigorously debated by theorists and research-
ers (Weiten, 2004). Current researchers (e.g., Bouchard, 1998; Scarr, 1997)
believe heredity sets the upper and lower bounds on intelligence, and
environment determines where one falls within these boundaries.
Hence, the dilemma in deciding whether to use cognitive ability tests for
selection boils down to a choice between using what is likely the single best
predictor of job performance and incurring adverse impact or utilizing
other, less effective predictors and potentially increasing minority represen-
tation in the workforce (Murphy, 2002; Schmitt, Rogers, Chan, Sheppard,
& Jennings, 1997). This dilemma may be even more chal enging for public
sector employers who are often under legal and societal mandates to create a
more diverse, representative workforce. Furthermore, a demographical y
diverse workforce may enhance job performance in some functions, such as
policing urban areas with substantial minority populations.
The discrimination against minority groups resulting from the use of
cognitive ability tests is often legally defensible and fair because test perfor-
mance is likely to be related to job performance. Although other predictors,
such as integrity tests, personality tests, or structured interviews, may result
in less adverse impact, they also typically result in less utility in predicting
job performance. Individuals selected using noncognitive predictors will
likely exhibit different strengths and weaknesses on the job than those
selected using cognitive ability tests (Murphy, 2002). Where cognitive skills
are critical to effective job performance, those selected using alternative
tests will likely not perform as well as those selected based on cognitive abil-
ity. Hattrup, Rock, and Scalia (1997) indicated that increasing the weight
of noncognitive factors in selection may reduce adverse impact but will
likely reduce job performance as well. In many civil service jobs where pub-
lic safety and welfare are a direct consequence of effective job performance,

---

Shoenfelt, Pedigo / COGNITIVE ABILITY TESTING
273
a greater premium may be placed on better prediction of job performance
using cognitive ability tests than on risking less potential utility gained from
noncognitive predictors.
The landmark case of Griggs v. Duke Power Company (1971) found that
if a selection device results in adverse impact, that test must be job related.
Furthermore, the Uniform Guidelines on Employment Selection Procedures
(Equal Employment Opportunity Commission [EEOC], 1978), which are
given great deference by the courts, and subsequent case law indicate an
organization using a test resulting in adverse impact should ensure that the
test has demonstrated job-related validity. Cognitive ability is related to
performance across a wide range of jobs and occupational settings (e.g.,
Hunter & Hunter, 1984; Murphy, 2002; Schmidt & Hunter, 1998).
Nonetheless, because the use of cognitive ability testing frequently results in
race-based adverse impact, organizations using cognitive ability testing in
their selection process are likely to be challenged (Terpstra, Mohamed, &
Kethley, 1999).
Little published research has examined the outcome of the litigation
associated with different types of selection procedures. In one of the few
such studies, Terpstra et al. (1999) reviewed cases involving nine selection
devices, including cognitive ability testing. However, relatively little infor-
mation was reported regarding cognitive ability testing other than that 28
discrimination charges were identified, and 67% of these charges were
found to be not discriminatory. A review of the literature yielded no other
studies analyzing legal conclusions regarding cognitive ability testing in
selection.
The current study examined factors determining the outcome in cogni-
tive ability testing court cases from 1992 through 2004. Only cases that
took place after 1991 were included because in November 1991 a new Civil
Rights Act was passed that resulted in significant changes to court rulings.
The 1991 Civil Rights Act returned the burden of demonstrating the job
relatedness and business necessity of the test from the plaintiff to the defen-
dant. Prior to 1989 (Wards Cove Packing v. Atonio, 1989), as established in
Griggs (1971), the burden was on the employer to demonstrate that the test
was job related. The judicial scenario was such that the initial burden was
on the plaintiff to establish a prima facie case of Title VII discrimination,
typically by demonstrating disparate impact on members of a protected
group. With a successful prima facie case by the plaintiff, the burden of pro-
duction shifted to the defendant, who was required to establish that the test
in question was job related and consistent with business necessity. If the
defendant was successful, the burden fell back on the plaintiff to establish

---

274
REVIEW OF PUBLIC PERSONNEL ADMINISTRATION / Sept. 2005
either that the articulated reason was a pretext for discrimination or that an
equally valid yet less discriminatory alternative test existed. In Wards Cove
(1989), the Court held that the burden was on the plaintiff to demonstrate
that the defendant’s test was not job related. During the period between
1989 and 1991, plaintiffs had the burden of persuasion that the test did not
serve the legitimate business needs of the employer. Plaintiffs were fre-
quently at a disadvantage because they often lacked the knowledge of test
validity, the access to selection tests, and the financial resources needed for
this demonstration. The Civil Rights Act of 1991 enacted into law the judi-
cial scenario set forth in Griggs.
In addition, the 1991 Civil Rights Act added jury trials. Prior to 1991,
the Supreme Court in Curtis v. Loether (1974) and later in Lorillard v. Pons
(1978) indicated that jury trials are not available under Title VII. The
Court found the discretionary nature of the equitable relief award of back
pay under Title...