CRITERION-RELATED VALIDITY OF PERSONALITY AND APTITUDE SCALES - A COMPARISON OF VALIDATION RESULTS UNDER VOLUNTARY AND ACTUAL TEST CONDITIONS (FROM POLICE SELECTION AND EVALUATION, BY SPIELBERGER - SEE NCJ-59208)

• NCJ Number: 59219
• Author(s): N D HENDERSON
• Date Published: 1979
• Length: 17 pages
• Grant Number(s): 75-DF-99-0037
• Sponsoring Agency: US Dept of Justice
• Corporate Author: Hemisphere Publishing Corp
• Publication Format: Document
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

THE CLEVELAND POLICE VALIDATION PROJECT IS DESCRIBED FOR THE PURPOSE OF IDENTIFYING THE ACCURACY AND USEFULNESS OF PERSONALITY AND APTITUDE MEASURES IN POLICE ENTRANCE EXAMINATIONS.

Abstract

THE CLEVELAND POLICE VALIDATION PROJECT PROVIDED AN OPPORTUNITY TO EXAMINE THE USEFULNESS AND VALIDITY OF PERSONALITY AND APTITUDE TESTS IN POLICE CANDIDATE SELECTION. THE PROJECT BEGAN IN 1973, AND INVOLVED CARRYING OUT A TRADITIONAL CONCURRENT VALIDATION STUDY. FOUR HUNDRED PATROL OFFICERS WERE RATED BY BOTH THEIR SUPERVISORS AND FELLOW OFFICERS. ONE HUNDRED WHITE OFFICERS AND FIFTY BLACK OFFICERS COMPLETED A BATTERY OF TESTS MEASURING A NUMBER OF APTITUDES, INTERSTS, SKILLS, AND PERSONALITY DIMENSIONS. SCORES WERE CORRELATED WITH VARIOUS CRITERION MEASURES AND MULTIPLE-REGRESSION EQUATIONS. THE BATTERY OF TESTS WAS THEN ADMINISTERED TO 160 ROOKIES JUST COMPLETING POLICE ACADEMY TRAINING, TO PERMIT A PREDICTIVE VALIDATION STUDY. IN 1974 APPROXIMATELY 950 OFFICERS TOOK THE PROMOTIONAL EXAMINATION FOR SERGEANT; SINCE SEVERAL OF THESE APTITUDE AND PERSONALITY SCALES REPRESENTED PARALLEL FORMS OF SCALES USED IN THE 1973 TESTING, RESEARCHERS HAD THE OPPORTUNITY TO COMPARE VALIDITY COEFFICIENTS OBTAINED UNDER BOTH ORIGINAL NO-STAKES PROCEDURES AND UNDER TRUE EXAMINATION CONDITIONS. ANALYSIS OF THE STATISTICAL DATA OBTAINED REVEALED THAT EMPIRICALLY VALIDATING SELECTION DEVICES FOR POLICE WORK IS EXTREMELY COMPLICATED. THE PROBLEMS OF SAMPLE SIZE, RANGE RESTRICTION, AND TEST SCORE CRITERION CONTAMINATION ARE SEVERE FOR EVEN THE MOST BASIC VALIDATION PROCEDURES. IN MOST SITUATIONS COMPLEX PROCEDURES, SUCH AS DIFFERENTIAL VALIDATION, MAY LOSE ALL INTERNAL VALIDITY BECAUSE OF THESE PROBLEMS. SPECIFICALLY, THE CLEVELAND PROJECT INDICATED THAT THE RELATIONSHIPS BETWEEN VARIOUS PERSONALITY DIMENSIONS AND JOB PERFORMANCE OBTAINED IN A CONCURRENT VALIDATION SITUATION WITH VOLUNTEER SUBJECTS DISAPPEARED UNDER ACTUAL TEST CONDITIONS, WHILE RELATIONSHIPS WITH APTITUDE MEASURES INCREASED SLIGHTLY DURING TRUE TEST CONDITIONS. FURTHER ANALYSIS OF PERSONALITY PROFILES OF HIGHLY-RATED OFFICERS SUGGESTED THAT BOTH SUPERVISOR AND PEER RATERS MAY BE CONTAMINATING THEIR JUDGMENTS WITH RACIAL STEREOTYPES OR DIFFERENTIAL EXPECTATIONS FOR BLACKS AND WHITES. IT IS CONCLUDED THAT CAREFUL EXAMINATIONS OF VALIDATION STUDIES IN THIS AREA IS NECESSARY FROM THE POINT OF VIEW OF BOTH INTERNAL AND EXTERNAL VALIDITY. REFERENCES, FOOTNOTES, TABULAR DATA, AND CHARTS PROVIDED. (LWM)