Latent class models for cognitive diagnosis have been developed to classify examinees into one of the 2<Superscript> K </Superscript> attribute profiles arising from a K-dimensional vector of binary skill indicators. These models recognize that response patterns tend to deviate from the ideal responses that would arise if...</superscript>

A statistical test of the unidimensionality of the parameter underlying independent non-identically distributed Bernoulli trials is proposed and large sample results are derived. The procedure has applications in psychometry and hopefully other applied disciplines.