Evaluating When Subscores Add Value in Psychological and Health Applications.

Abstract

Many scales used in psychological and health research are designed to yield subscores, yet it is common to see total scores reported instead. One challenge of using subscores is they can lack adequate reliability due to their shortened length. However, methods originally developed for educational measurement have shown that augmenting subscores can improve reliability estimates. Augmented subscores blend the individual score with other sources of information. The present study sought to understand (a) the costs of ignoring subscores in favor of total scores and (b) the extent to which augmentation can help alleviate challenges encountered when using subscores. Data were simulated to examine when subscores should be preferred to total scores and the magnitude of improvement from using augmented subscores over non-augmented subscores. Results suggested that when a scale is designed to yield subscores, there is practical benefit to using them. In situations where subscore reliability is low, we recommend using augmentation.