This study was designed to determine the predictive validity of a police officer selection program and identify the contribution made by each major selection device to the total program. Police officers employed by the Portland Police Bureau who had completed three years of post-probationary employment were randomly assigned to a validation group and a cross-validation group on a two for one basis respectively. Beta weights were computed for the written test, interview and psychological scores in the validation group. The regression formula was then applied to the data in the cross-validation group. A cross-validation R of 0.12 was obtained between the predicted performance criterion scores and the actual performance scores. When the interview variable was removed from the equation the cross-validation R increased to 0.16. Neither validity coefficient reaches statistical significance. Reasons were offered for believing that the low magnitude of the coefficients was attributable to restriction in range in the predictor variables and the unreliability of the criterion variable.
| Identifer | oai:union.ndltd.org:pdx.edu/oai:pdxscholar.library.pdx.edu:open_access_etds-3278 |
| Date | 29 July 1975 |
| Creators | Davidson, Neil Bingham |
| Publisher | PDXScholar |
| Source Sets | Portland State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Dissertations and Theses |
Page generated in 0.0021 seconds