[IPAC-List] Article on test bias in press at J. of Applied Psychology

[Aguinis, Herman]

Dear Colleagues,

In a posting from a few months ago (see below), Winfred Arthur referred to a manuscript that was then under review. That manuscript has now been accepted for publication in Journal of Applied Psychology and it challenges established conclusions regarding test bias in preemployment testing. If you are interested, feel free to download it from http://mypage.iu.edu/~haguinis/ [click on the "refereed journal articles" link]. The abstract is below.

Happy New Year!

--Herman.
*****************************************************
Herman Aguinis, Ph.D.
Dean's Research Professor &
Professor of Organizational Behavior and Human Resources
Department of Management and Entrepreneurship
Kelley School of Business, Indiana University
http://mypage.iu.edu/~haguinis/
****************************************************

Revival of Test Bias Research in Preemployment Testing

Herman Aguinis
Department of Management and Entrepreneurship
Kelley School of Business
Indiana University

Steven A. Culpepper
Department of Mathematical and Statistical Sciences
University of Colorado Denver

Charles A. Pierce
Department of Management
Fogelman College of Business & Economics
University of Memphis

Abstract
The authors developed a new analytic proof and conducted Monte Carlo simulations to assess the effects of methodological and statistical artifacts on the relative accuracy of intercept- and slope-based test bias assessment. The main simulation design included 3,185,000 unique combinations of a wide range of values for true intercept- and slope-based test bias, total sample size, proportion of minority group sample size to total sample size, predictor (i.e., preemployment test scores) and criterion (i.e., job performance) reliability, predictor range restriction, correlation between predictor scores and the dummy-coded grouping variable (e.g., ethnicity), and mean difference between predictor scores across groups. Results based on 15 billion 925 million individual samples of scores and more than 8 trillion 662 million individual scores raise questions about the established conclusion that test bias in preemployment testing is non-existent and, if it exists, it only occurs regarding intercept-based differences that favor minority group members. Because of the prominence of test fairness in the popular media, legislation, and litigation, our results point to the need to revive test bias research in preemployment testing.








>>> "Winfred Arthur, Jr." <wea at psyc.tamu.edu> 3/17/2009 10:31 AM >>>

. . . so, if you feel up to some leisure reading on this, Herman (Aguinis) recently shared with me a differential validity paper that he currently has under review.  amongst its strengths, i think the introduction provides a fairly good review and tutorial on tests for differential prediction.  i have taken the liberty of pasting the title and abstract below.  if you are interested in a copy of the paper, then pls contact Herman directly (Herman.Aguinis at ucdenver.edu<mailto:Herman.Aguinis at ucdenver.edu>); i am sure he will be glad to share a copy with you.

enjoy :)

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Aguinis, Culpepper, Pierce, (2009), Test Bias Revived: Challenging 40 Years of  Research on Preemployment Testing.

Abstract
We conducted a Monte Carlo simulation to assess the effects of methodological and statistical artifacts on the relative accuracy of intercept- and slope-based test bias assessment. Our simulation design included 3,185,000 unique combinations of a wide range of values for intercept- and slope-based test bias, total sample size, proportion of minority group sample size to total sample size, predictor (i.e., preemployment test scores) and criterion (i.e., job performance) reliability, predictor range restriction, correlation between predictor scores and the dummy-coded grouping variable (e.g., ethnicity, gender), and mean difference between predictor scores across groups. Results based on 15 billion 925 million individual samples of scores suggest that previous test bias assessment is likely to have made one or both of the following incorrect conclusions: (a) there is no slope-based test bias and (b) there is intercept-based biased favoring minority group members.  Thus, our results challenge the established conclusions that test bias in preemployment testing is non-existent and, if it exists, it occurs regarding intercept-based differences only that favor minority group members. The results provide an alternative explanation for an accumulated body of 40 years of research on test bias in preemployment testing, particularly in the context of general mental abilities testing, and have important implications for personnel selection research and practice. Also, these results have important implications for organizations and society given the prominence of the topic of test fairness in the popular media, legislation, and litigation.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

- winfred

Patrick McCoy wrote:

Hello Joel:



The following article might be of relevance to your question.:



Sackett, P.R., Laczo, R.M., & Lippe, Z. (2003). Differential prediction

and the use of multiple predictors: The omitted variables problem. JAP,

88(6), 1046-1056.



Patrick McCoy, Ph.D.

Ottawa', Canada







Joel Wiesen <wiesen at personnelselection.com><mailto:wiesen at personnelselection.com> 16/03/2009 8:26 am >>>



What do you think of this apparent support for differential validity

(from a recent Personnel Psych article)?



Might it imply that education is problematic as selection tool when the



applicant group includes minorities and/or women?  Would it imply that



an education requirement may be unfair to minorities and/or women?



Joel





"Finally, Hypothesis 6 predicted that the education-performance

relationship will be stronger for men (vs. women; Hypothesis 6a) and

for

Caucasians (vs. non-Caucasians; Hypothesis 6b). With respect to the

relationship between education level and task performance, we found

that

the relationship was more positive for Caucasians than for other racial



groups and for men than for women. Further, we found that the

relationship between education and OCB was more positive for Caucasians



than for other racial groups. These results provide some support for

Hypothesis 6a and Hypothesis 6b."



From:

PERSONNEL PSYCHOLOGY, 2009, 62, 89-134

HOW BROADLY DOES EDUCATION CONTRIBUTE TO JOB PERFORMANCE?

THOMAS W. H. NG; University of Hong Kong

DANIEL C. FELDMAN; University of Georgia



_______________________________________________________

IPAC-List

IPAC-List at ipacweb.org<mailto:IPAC-List at ipacweb.org>

http://www.ipacweb.org/mailman/listinfo/ipac-list

_______________________________________________________

IPAC-List

IPAC-List at ipacweb.org<mailto:IPAC-List at ipacweb.org>

http://www.ipacweb.org/mailman/listinfo/ipac-list




CONFIDENTIALITY NOTICE: This communication with its contents may contain confidential and/or legally privileged information. It is solely for the use of the intended recipient(s). Unauthorized interception, review, use or disclosure is prohibited and may violate applicable laws including the Electronic Communications Privacy Act. If you are not the intended recipient, please contact the sender and destroy all copies of the communication.