There are three volumes in the series on this topic: item response theory (IRT). What is IRT? Based on scores in a battery of questions in Psychometrics, IRT is a paradigm for designing, analysing, and interpreting the individual’s abilities. The volume one describes models that are used in IRT. This book is a collection of articles about IRT versus statistical tools. This is volume two covers mainly probability distributions, models with intentional and nuisance parameters, information criteria, and model identification issues among others. The volume three provides the applications of IRT. The IRT ideas and methods are highly utilised in economics, sociology, education, marketing research, political science, criminology, demography, and psychology.
Basic knowledge of calculus and statistical theory are necessary to understand the contents of the volume three. The volume three illustrates test item calibration, model fit, test score interpretation, adaptive testing, forensic responses, health measurement, computer programs, parameter linking, dimensionality analysis, differential item functioning, calibrating technology enhanced items in Section I, person fit, score reporting, IRT observed score equating in Section II, optimal test design, adaptive testing, standard setting, test spreededness and time limits, item and test security in Section III, large scale group score assessments, psychological testing, cognitive diagnostic assessment, health measurement, marketing research, and measuring change using Rasch models in Section IV, IRT packages in R, Bayesian inference using Gibbs sampling (BUGS) for IRT models, BILOG-MG, FARSCALE, IRTPRO, Xcalibre 4, EQSIRT, ACER conquest, Mplus, gllamm, latent GOLD, WinGen, firestar, and jMetrik in Section V.
The authors of the 20 chapters in this volume are well written by experts around the world. I enjoyed reading this book. I recommend this book to all applied, statistics and computing professionals.