Abstract
This research attempted to determine the effect of redundant and consistent stems in multiple-choice vocabulary tests on pre-university students’ performance. To this purpose, a sample English language proficiency tests was administered to a total of 130 pre-university students. Sixty homogeneous students were randomly selected and divided into two groups, one consistent group that took the test consisting of consistent stems and the other redundant group that took the test containing redundant stems. Next, three versions of the same test were provided in which the items had the same alternatives but different stems. The first and second versions of the consistent test were assigned to the consistent group, and the first two versions of the redundant tests to the redundant group. In the last versions of the two tests, the groups were reversed. This time, the consistent group took the test containing redundant stems while the redundant group took the test consisting of consistent stems. All the tests were teacher-made summative achievement tests and their reliability coefficients, the statistical independent t-test, paired samples t-test, and Pearson correlation formula were calculated. The results revealed that there was no significant difference between the students’ performance on two kinds of tests (p<0.05). The empirical findings of the present study suggest that pre-university students’ competence in vocabulary can be measured through both tests with redundant and tests with consistent stems.
