Abstract
Weber (1977) contends that the use of Rasch analysis, principal components analysis, and classical test analysis shows that an in strument designed to measure a "bilevel dimensionality" in proba bility achievement (Levin, 1975) in fact measures a single latent trait. This article questions that interpretation and the use of Rasch and classical analysis to establish unidimensionality.

This publication has 2 references indexed in Scilit: