Median Splits, Type II Errors, and False Positive Consumer Psychology: Don't Fight the Power

Abstract

Considerable prior statistical work has criticized replacing a continuously measured variable in a general linear model with a dichotomy based on a median split of that variable. Iacobucci, Posovac, Kardes, Schneider, and Popovich (this issue) defend the practice of “median splits�? using both conceptual arguments and simulations. We show that both their conceptual arguments and their simulations range from incomplete to incorrect. There are no real benefits to median splits, and there are real costs in increases in Type II errors through loss of power and increases in Type I errors through false positive consumer psychology. Median splits remain a bad idea.