Abstract
In recent years, we see many examples of applying item response theory to psychological questionnaires. There are various conditions under which psychological tests are administered. Often, we are interested in the subjects' latent traits in the context of experimental design. In that case, we might want to know the effects of the experimental factors on the item parameters. In this study, we proposed an item response model in which items could be classified by some qualitative factors as in the situations where the items are administered under several different experimental conditions. The model's item parameters were linearly structured as in the general linear model. Then, we showed an example of analysis using an actual data set which was collected under four conditions, where the latent trait in question was “tension.”
