Test Construction & Validation
The B-PAD Testing Procedure and Scoring
1. The Validity of B-PAD as a Measure of Problem-Solving Skills
2. The Validity of B-PAD as a Measure of Interpersonal Skills
3. The Validity of B-PAD as a Measure of Expected Job Performance
4. B-PAD and Adverse Impact On Women & Minorities
5. The Effect of Job Experience On B-PAD Scores
6. The Effect of Practice On B-PAD Scores
7. B-PAD's Correlation With Observed Competencies in the Police Academy
8. The Validity of B-PAD as a Measure of Supervision & Management Skills
9. The Validity of B-PAD as a Measure of Ability to Function As Part of a Team
10. The Effect of Fire/EMS Job Experience on B-PAD Scores
The Validity of B-PAD As a Measure of Expected Job Performance
Purpose of the Study This study was conducted as part of the initial validation research carried out on B-PAD by Rand (1987). The purpose of this segment of the study was to determine the concurrent validity of B-PAD's Scale 3 (Overall Effectiveness) as a predictor of on-the-job interpersonal competence among police officers.
Methodology and Subjects 30 police officers from municipal police agencies in Marin County, California, were recruited as volunteers for participation in this study. All subjects were administered B-PAD in a manner consistent with pre-employment test administration outlined in B-PAD's Administrator & Proctor Manual (Second Edition). The subjects represented a diverse group in terms of demonstrated skill as measured by their B-PAD scores (subject scores on the Overall Effectiveness scale for 8 scenarios ranged from a low of 10 to a high of 29). Following the advice of Hunter (1984), the predictive criterion of interest was on-the-job interpersonal competence as rated by the subject's work supervisor. The raters who scored the subject's B-PAD performance were not informed of the subject's supervisory ratings. The supervisory ratings were for use in the study only and had no impact on the subjectsÕ personnel records.
The study's methodology adheres to the admonition of Wernimont and Campbell (1968), Burbeck and Furnham (1985), and Latham (1989) that predictors should represent realistic samples of behavior and that the criteria should be as similar as possible to the predictors. Supervisors in this study were asked to rate each subject's expected competency in handling the same interpersonal scenarios used as test stimuli in the B-PAD assessment.
Results Supervisory ratings of interpersonal competence correlated significantly with B-PAD ratings of interpersonal competence (r = .72, p > .01).
Conclusion Hunter & Hunter, in their meta-analysis of the validity and utility of alternative predictors of job performance, determined that the average validity of ability tests is .53 and concluded that "this very high level of validity is the standard against which all alternative predictors must be judged' (p. 81). RandÕs observed validity coefficient of .72 compares well with this high standard and supports the use of B-PAD as a valid predictor of on-the-job performance.