نویسندگان | Parvis Ajideh, Massoud Yaghoubi-Notash, H..R. babaiee Bormanki |
---|---|
نشریه | Journal of English Language Teaching and Learning (JELTL) |
ارائه به نام دانشگاه | University of Tabriz |
شماره صفحات | 39-56 |
شماره سریال | 14 |
شماره مجلد | 30 |
نوع مقاله | Full Paper |
تاریخ انتشار | 2022-08-07 |
رتبه نشریه | علمی - پژوهشی |
نوع نشریه | چاپی |
کشور محل چاپ | ایران |
نمایه نشریه | ISC Q1 |
چکیده مقاله
This paper reports on an investigation of native language-based differential item functioning (DIF) across the subtests of Iranian Undergraduate University Entrance Special English Exam
IUUESEE). Fourteen thousand one hundred seventy two foreign-
language test takers (including four groups of Azeri, Persian
Kurdish, and Luri test takers) were chosen for the study. Uniform DIF (UDIF) and Non-uniform DIF (NUDIF) analyses were conducted on data from the four versions of IUUESEE. After
establishing the unidimensionality and local independence of the data, DIF findings showed that Luri test takers were more advantaged than other native language groups across the subtests.
NUDIF analysis uncovered that almost all subtests functioned in favor of low-ability test takers who haven’t been expected to outperform high-ability test takers. A probable explanation for
native language-ability DIF was that Luri and low-ablity test takers were more likely to venture lucky guesses. Thoughtless errors and guessing, test-wiseness, overconfidence, stem length, unappealing distractors, and time were proposed as possible causes of DIF in IUUESEE. It was also found that the reading subtest included the large number of items with significant DIF.
tags: Iranian Undergraduate University Entrance Special English Exam (IUUESEE), Rasch Analysis, Differential Item Functioning, Native language.