Publication Date

Fall 2020

Advisor(s) - Committee Chair

Dr. Reagan Brown (Director), Dr. Elizabeth Shoenfelt, and Dr. Katrina Burch

Degree Program

Department of Psychological Sciences

Degree Type

Master of Science

Abstract

Equal weights are an alternative weighting procedure to the optimal weights offered by ordinary least squares regression analysis. Also called units weights, equal weights are formed by standardizing scores on the predictor variables and averaging these standardized scores to create a composite score. Research is limited regarding the conditions under which equal weights result in cross-validated š¯‘…š¯‘…2 values that meet or exceed optimal weights. In this study, I explored the effect of various predictor-criterion correlations, predictor intercorrelations, and sample sizes to determine the relative performance of equal and optimal weighting schemes upon cross-validation. Results indicated that optimally weighted predictors explained more criterion variance upon cross-validation as the variability in predictor-criterion correlations increased. Similarly, it appears that as predictor intercorrelations and sample size increase, optimally weighted predictors cross-validate to explain more criterion variance than equally weighted predictors. Implications and directions for future research are discussed.

Disciplines

Applied Statistics | Industrial and Organizational Psychology | Psychology

Download

Share

COinS