The classical test or item response measurement theory

Abstract

While the Examination Council of Lesotho (ECOL) is burdened with a huge workload of assessment tasks, their procedures for developing tests, analysing items, and compiling scores heavily rely on the classical test theory (CTT) measurement framework. The CTT has been criticised for its flaws, including being test-oriented, sample dependent, and assuming linear relationships between latent variables and observed scores. This article presents an overview of CTT and item response theory (IRT) and how they were applied to standard assessment questions in the ECOL. These theories have addressed measurement issues associated with commonly used assessments, such as multiple-choice, short response, and constructed response tests. Based on three search facets (Item response theory, classical test theory, and examination council of Lesotho), a comprehensive search was conducted across multiple databases (such as Google Scholar, Scopus, Web of Science, and PubMed). The paper was theoretically developed using the electronic databases, keywords, and references identified in the articles. Furthermore, the authors ensure that the keywords are used to identify relevant documents in a wide variety of sources. A general remark was made on the effective application of each model in practice with respect to test development and psychometric activities. In conclusion, the study recommends that ECOL switch from CTT to modern test theory for test development and item analysis, which offers multiple benefits