U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Evaluation of the Written Selection Test: Effectiveness and Alternatives

NCJ Number
173264
Journal
Journal of Criminal Justice Volume: 26 Issue: 3 Dated: May/June 1998 Pages: 175-183
Author(s)
L K Gaines; S Falkenberg
Date Published
1998
Length
9 pages
Annotation

This article reviews the literature regarding the use of written selection tests and examines the particular application a written test to a group of police department applicants.

Abstract

The literature review indicates that police departments have long used written selection tests and that many early police selection practices were implemented to thwart the political spoils system. Psychometric properties of a typical written police selection test were examined using a sample of 419 subjects applying for the position of police officer with a medium-sized police agency; 61 or 14.6 percent were female and 358 or 85.4 percent were male. The test under consideration was a commercially available test. The test format consisted of a study booklet that was reviewed by test takers immediately prior to completing the examination. Job-relevant content was presented in the study guide to equalize the knowledge base of candidates and to provide materials to test learning, memory, and perceptual recognition skills. The test itself consisted of subscales, including ability to learn and apply information, ability to remember details, verbal ability, ability to accurately complete forms, spatial ability, and ability to use judgment and logic. Statistical analysis revealed the written selection test fell substantially short of expectations. Adverse impact was not avoided, an item analysis of the test revealed it was improperly constructed because extremely easy items were overrepresented, and none of the factors to be tested corresponded with any of the test subscales. The high correlation between test scores and applicant educational level suggested an education requirement may be a more effective way of selecting a pool of qualified applicants. 29 references and 6 tables