A Validity and Multiple Regression Study of Performance in the GEPI as a Predictor for Performance on the GED

Abstract

This study attempts to determine the relationship between the subtests of the General Education Performance Index (GEPI) and the revised version of the General Educational Development Test ( GED) , and reports a technique whereby performance on the GEPI may be used to predict performance on the GED. Participants were men and women, aged 16 to 62. who were given the GEPI and then the GED. Multiple regression equations were generated to predict GED scores using age and each GEPI score. No significant correlation was found between age and GED scores or between age and GEPI scores, but a highly significant correlation ( 9 < .001) was found between each GEPI subtest score and the subsequent score attained on the GED. Regression equations for each subtest and cross validation that demonstrated high predictive validity for each regression equation are reported.