A validation study of the Kansas civil service exam for office assistant, level 2.

Abstract

The use of testing in the selection of employees is a common practice. However, such testing must meet certain requirements set forth by the Uniform Guidelines for Employee Selection Procedures (1978). These guidelines require that all selection tests be periodically validated to ensure the test is a good predictor of how well someone will perform on the job. The purpose of this study was to validate a state civil service exam. A concurrent validity study was conducted using the selection test (Civil service exam), and the level of education as predictors and employee's initial performance appraisal as the dependent variable. Two regressional analysis were conducted to determine: 1) the predictive ability of the civil service exam on employee performance and 2) the predictive accuracy of the selection device and education on employee's performance. Results of the analysis revealed that neither the civil service exam or the level of education served as accurate predictors of employee performance. This researcher concluded that the state of Kansas should examine its selection process in order to determine if the discrepancies were due to the selection test, the performance appraisal, or both.