Abstract
The information function is an important statistic in item response theory (IRT) applications. Although the information function is often described as the IRT version of reliability, it differs from the classical notion of reliability from a critical perspective: replication. This article first explores the information function for the one-parameter model in detail and suggests an alternative method for its computation. Second, the difference between the IRT and classical test theory standard errors of measurement is discussed.
Keywords
Get full access to this article
View all access options for this article.
