Investigating the Reliability of Language Tests Using Classical Test Theory and Item Response Theory

Abstract

Aim of the Study: The study aims to investigate the concerns about the reliability of language tests. It uses both Classical Test Theory (CTT) and Item Response Theory (IRT) to examine the reliability of language tests with high-stake consequences.

Methodology: The study uses a mixed method approach using both qualitative and quantitative data. It conducts interviews with language experts and participants to examine the reliability of language tests. Furthermore, the study collects quantitative data through a survey with language professionals.

Findings & Conclusion: Using 15 questions, the value of Cronbach’s Alpha was 0.939. It is above 0.70, which means it is good. This shows a great level of reliability. It means the language tests prepared were valuable and helpful for the participants. The tests of synonyms, reported speech, and analytical questions are a part of aptitude tests. CTT and IRT approaches were used, and which marks obtained from the participants helped to know the level of reliability.