Introduction to item response theory (IRT) and computerized adaptive testing (CAT) with the R software

Item response theory (IRT) has become an important field of research for psychology and educational assessment. Recently, with the increase of computational power, several IRT-related topics have emerged, among others, computerized adaptive testing (CAT). The main aim of CAT is to provide a framework for individualized assessment by means of optimal item selection and administration to the test takers. CAT has several assets to linear (non-adaptive) testing: individualized assessment, limited risk of cheating or fraud, shorter tests providing the same amount of information as longer linear tests, automatic scoring and reporting at the end of the test. Practical use of CAT, however, remains limited so far due to several factors (lack of available large item banks, content validity and security, lack of suitable software for practical CAT assessment, ethical issues in administering different tests to estimate the same ability, etc.).
The purpose of this workshop is threefold:
(a) to provide a general overview of IRT and CAT,
(b) to introduce the R software in a user-oriented way, as well as several IRT tools (including the package catR for CAT simulations),
(c) to perform practical training sessions with the participants.
The workshop will be a mix of oral presentations, demonstrations related to the R software, and practical sessions where participants will be invited to train with R and catR.
The R software is an open-source platform for statistical inference and testing, graphical display and data visualization. It also holds several add-on packages for specific IRT purposes (item calibration, ability estimation, multidimensional scaling, equating, differential item functioning etc.). The R community is worldwide and proposes free exchanges of shared R packages through the CRAN (comprehensive R archive network). In this workshop, the R package catR will be examined and used in the practical sessions.