During the past few decades, documentation of test takers’ proficiency level has been accomplished through large-scale assessments most importantly Cognitive Diagnostic Assessment (CDA) in order to provide skills mastery profile of test takers in fine-grained detailed information. Accordingly, the present study attempts to scrutinize reading comprehension test items of a high-stakes test under CDA. To delve into this issue, DAF was used to detect the probability of mastery of attributes among test takers, and DIF was applied to show item performance among different candidates in terms of gender. Thus, the participants of this study were 3220 females and males attending PhD national admission test in Iran. Through adopting sequential exploratory mixed method design, GDINA model was run by the application of R-studio package. Results of the study revealed that test items suspected DIF against female. In the end, the findings of this study were discussed in light of their implications for language testing community to perceive potential social harm which derived from biased test items in PhD national admission exams.
- حق عضویت دریافتی صرف حمایت از نشریات عضو و نگهداری، تکمیل و توسعه مگیران میشود.
- پرداخت حق اشتراک و دانلود مقالات اجازه بازنشر آن در سایر رسانههای چاپی و دیجیتال را به کاربر نمیدهد.