Abstract
The fairness and criterion-related validity of the Correctional Officers' Interest Blank as a predictor of rated job performance was investigated for three county-level corrections-related jobs: correction officers, juvenile counselors, and probation officers. Validity for corrections officers was considerably below that reported in the test manual (Gough, 1982). Criterion unreliability and range restriction in COIB scores were ruled out as explanations of the lower validity. Combining all available data yields an estimated validity of .27 for the COIB as a predictor of the job performance of corrections officers. The COIB is only weakly related to the job performance of juvenile counselors and has no relationship to the performance of probation officers. Analyses based upon the Cleary model (1968) indicate that the COIB fairly predicts the performance of minority and majority corrections officers and juvenile counselors. Further analyses indicate, however, that to avoid adverse impact the passing score for the COIB has to be set so low as to seriously reduce its effectiveness as a selection tool.
Get full access to this article
View all access options for this article.
