My previous four posts have focused on basic item response models1. These models are commonly found in an educational testing context but are less often seen in clinical research settings where questionnaires are used for the measuring and assessment of individuals’ conditions. Due to the higher level of complexity in the study design of clinical studies, IRT models are often discarded for the analyses of questionnaires. Simple sum scores of the questionnaires, instead of the more robust estimates derived from IRT models, are taken directly to represent the quantity of the latent …