Our Team

Investigating the Quality of Language Tests; Calculating the Reliability Coefficient of Placement Test as a Case Study

Abstract

Language measurement tools should have reliability to obtain scores which help drawing meaningful inferences. The aim of the study was to find out the reliability of a placement test. Based on the Classical Test Theory, the reliability coefficient of tests is computed to find out the range of the measurement error. To this end, the data obtained from the University of Raparin Placement Test was analyzed. The split half method and the formulas – Kuder-Richardson 20 and Spearmen-Brown Prophecy – were used to calculate the reliability estimate. The results showed the test’s reliability is 0.81, and the number of the items need to be doubled to achieve the desired reliability of 0.9. Thus, it is recommended the test structure should be modified; test items are increased and the content is amended to cover the language skills. Additionally, other sources of unreliability need to be studied to find the source of measurement error.

Keywords: Placement Test, reliability coefficient, split-half method