ABSTRACT
This chapter focuses on the properties and the possibilities of the random item effects model for the analysis of cross-national survey data. The random item effects approach supports the use of country-specific item characteristics and a common measurement scale. Item response theory (IRT) methods are standard tools for the analysis of large-scale assessments of student's performance. IRT methods provide a set of techniques for estimating individual ability levels and item characteristics from observed discrete multivariate response data. Common IRT models assume a priori independence between individual abilities. The common IRT model with a multilevel population model for the ability parameters is called a multilevel IRT model. The estimation method for the multilevel item response theory model with random item effects is evaluated by investigating convergence properties and by comparing true and estimated parameters for a simulated data set.
