Abstract:
: Most universities across Iran tend to develop English tests for placement, exit, achievement, and other purposes. Examining the various features of such tests is imperative for making informed decisions about learners’ achievement level. The present study examined the features of a university-wide administered English language achievement test at Iran University of Science and Technology (IUST). The characteristics of the test including reliability, item facility/difficulty, and item discrimination were examined. The reliability estimates showed that the test had a relatively acceptable level of reliability. However, it was relatively easy, and the majority of the items had a rather low discrimination power. Moreover, all the subsections of the test were found to contribute significantly to the overall achievement of the participants. It was concluded that more attention needs to be paid to the difficulty and discrimination characteristics of such achievement tests assigned on a large scale across universities. Suggestions are made about how more accountable decisions can be made to have fair tests for language achievement purposes.
Machine summary:
"The present study examined the features of a university-wide administered English language achievement test at Iran University of Science and Technology (IUST).
The present study, using classical testing theory measures, tries to examine the English language achievement test administered to the undergraduate engineering university students at Iran University of Science and Technology.
The motivation behind this study was that since this test is used to make judgments about the success or failure of a large number of students at the end of an English course, an assessment of the characteristics of this test in terms of its reliability, item difficulty, item discrimination, and of the performance on its subsections seems warranted.
For example, reading rate studies and its effect on comprehension, level of L2 language proficiency required to perform effectively in reading comprehension, the role of L1 reading strategies in L2 reading comprehension, the role of background knowledge or schemata in reading comprehension, the complex interactions between background knowledge and other variables like interest and text structure, knowledge of metacognitive strategies and how to use those strategies in conjunction with other strategies as an index of successful reading (Anderson, 1991), as well as the role of extensive reading in L2 reader progress.
Rubin (1994) highlights some important factors affecting listening comprehension: (1) text characteristics, such as temporal/acoustic variables like speech rate, pause, and hesitation; (2) interlocutor characteristics like interlocutors’ gender, culture, place of origin; (3) task characteristics, such as multiple-choice, Wh-questions, open-ended questions; (4) listener characteristics, which refer to language proficiency level, memory, attention, gender, affective variables, background knowledge, and learning disabilities, and (5) process characteristics, which refer to how listeners interpret input in terms of what they know or identify what they do not know."