Predicting Optimal Training Group Assignment

Abstract

Multiple regression equations were used to assign 40 students to fixed-difficulty or adaptive training based upon the shorter predicted time-to-train score. In addition, 40 students were randomly assigned to the two training conditions, and 40 students were mismatched to training based upon the longer predicted time to train. Using predicted scores to match students to training alternatives resulted in a 47% savings in training time over random assignment and a 53% savings over mismatched assignment. The assignment effect was reliable at the 0.0001 level. Future research will examine different categories of predictors, additional training alternatives, and more complex training tasks.