Computational aspects of modelling item responses

Abstract

ZÁKLADNÍ ÚDAJE: IMPS 2023: Abstract Book: Talks. Psychometric Society, 2023. s. 202-202. [IMPS 2023: International Meeting of the Psychometric Society. 23.07.2023-28.07.2023, Washington DC]. ABSTRAKT: Item response theory models can be derived in factor analytic framework as well as in the framework of generalized linear and nonlinear mixed-effect models. In this work, we focus on the latter one. We first describe step-by-step development of IRT models via empirical characteristic curves and generalized linear and nonlinear models (GLNM) with emphasis on didactic value of such approach. In addition to that, we discuss wide usage possibilities of GLNM in terms of criterion-related item validity and we demonstrate these aspects with real data examples. Finally, we present some novel approaches to parameter estimation in this framework together with their challenges in practical implementation.