Going beyond traditional examinations of the relationship between high school grades and college performance.

Abstract

The overarching goal of this thesis is to conduct a more detailed examination of HSGPA as a predictor of college performance. Specifically, following a review of the pertinent literature, three issues were considered. First, variability in the predictive power of HSGPA across institution was assessed including the extent to which college/university characteristics can predict such variability. Results indicated that institutional characteristics can help explain variability in the relationship between HSGPA and FGPA. Such information can result in better estimates of validity at specific institutions. Second, the degree to which HSGPA displays non-linearity in the prediction of college grades across its range was examined. Predictive strength was found to decline for HSGPAs above 4.0. Finally, the impact of socioeconomic status on the predictive power of HSGPA was assessed. SES had little impact for all but the lowest levels of HSGPA. A weaker relationship between HSGPA and FGPA emerged for students who were both low in SES and HSGPA when compared to their higher SES counterparts.