Some Applications of Logistic Latent Trait Models with Linear Constraints on the Parameters

摘要

The linear logistic test model (LLTM), a Raschmodel with linear constraints on the item parameters,is described. Three methods of parameter estimationare dealt with, giving special considerationto the conditional maximum likelihood approach,which provides a basis for the testing of structuralhypotheses regarding item difficulty. Standardareas of application of the LLTM are surveyed, includingmany references to empirical studies initem analysis, item bias, and test construction; anda novel type of application to response-contingentdynamic processes is presented. Finally, the linearlogistic model with relaxed assumptions (LLRA) formeasuring change is introduced as a special case ofan LLTM; it allows the characterization of individualsin a multidimensional latent space and thetesting of hypotheses regarding effects of treatments.