Journal of Educational Measurement, Evaluation and Research

ABSTRACT

Measurement and evaluation play a central role in educational research, assessment design, and psychological testing. Among the most widely used frameworks for test analysis are Classical Test Theory (CTT) and Item Response Theory (IRT). While CTT has been historically dominant due to its simplicity and ease of application, IRT has emerged as a more sophisticated and flexible approach, offering deeper insights into item-level performance and test-taker ability. This paper presents a comprehensive and practical comparison between CTT and IRT, focusing on their theoretical foundations, assumptions, advantages, limitations, and real-world applicability. The discussion highlights how CTT relies on total test scores and assumes equal measurement precision across all ability levels, whereas IRT models the interaction between item characteristics and individual ability. A comparative analysis is supported through tables and a conceptual figure to demonstrate key differences. The paper further explores scenarios where each theory is more appropriate, particularly in modern large-scale assessments and adaptive testing environments. The findings suggest that while CTT remains relevant for small-scale assessments, IRT provides superior accuracy and flexibility in contemporary educational measurement.

KEYWORDS: Item Response Theory, Classical Test Theory, Educational Measurement, Test Reliability, Psychometrics, Assessment Models