Recently, there has been an increasing level of interest in subscores for their potential diagnostic value. As a result, there is a constant demand from test users for subscores. Haberman (2005) and Haberman, Sinharay, and Puhan (2006) suggested methods based on classical test theory to examine whether subscores provide any added value over total scores. This paper applied the above mentioned methods to recent data sets from a variety of operational tests. The results indicate that subscores provide added value for only a handful of tests.
