The Effects of Test Facets on the Construct Validity of the Tests in Iranian EFL Students

Abstract

Language testing as a main device in assessing the learners` knowledge and language abilities plays a key role in training programs. Generally, the goal of language testing is to assure the extent to which learners have achieved the instructional goals during a course. The main objective of many studies in language testing has been to investigate whether test facets affect construct validity of the test or not. Therefore, in this study, we investigated whether the EFL Iranian participants` performances were different with respect to the different test facets and if these performances had some effects on the construct validity of the tests. In this investigation, the students were selected of 50 Iranian EFL students aged between 21 to 30 years, from two branches of Islamic Azad University, Dezful and Andimeshk, Iran. The 17 participants, placed at the low level in the Nelson proficiency test, received a test. The test facets included the integrative forms such as cloze-test, c-test, and discrete test items such as multiple choice and true/false. By statistics analyses, the significant differences were assessed in the test facets. Our results revealed that significant differences existed in the test facets among the performances of Iranian EFL students. Because of the integrity of the several abilities and mental strategies, the cloze-test was the most difficult form of testing.

Keywords: Test facets; Construct validity; Integrative test items; Discrete test items