Classical test theory, item response theory, reliability, parameter estimation
Abstract
This article discusses basic concepts in Classical Test Theory and Item Response Theory. In the context of Classical Test Theory the concepts of observed and true score, reliability of observed scores, and item indices are discussed. Some common rules of thumb to interpret numeric values of these indices are also presented in line with some caveats. Some problems with Classical Test Theory are also summarized. In the context of Item Response Theory, basically, the logistic models are discussed highlighting the importance of the concept to be measured. Practical guidelines in taking the decision to accept the IRT model are discussed.
