چکیده

There has been a plethora of research and discussion on test fairness during the last couple of decades. The scope of fairness analyses range from a mere comparison of mean performance across groups of participants to large scale (qualitative and qualitative) studies of both psychometric features of the test and the uses to which test scores have been put. Differential item functioning (DIF) examines test fairness at the item level. Various methods have been suggested for DIF analysis most of which are based on categorization of the respondents based on observed variables. This study examines the applicability of a latent class analysis approach to DIF detection proposed by Zumbo, Liu, Wu, Shear, Olvera Astivia, and Ark (2015). In this approach, latent class analysis (LCA) is applied first to identify the underlying groupings that exist among the participants. Various covariates are then entered into the model to predict class membership. This way, an objective mechanism for identifying sources of DIF is created. The data for the present study comes from an administration of the nationwide university entrance exam for foreign languages. The implications of the results for both research on DIF analysis in general and the fairness of the test itself are discussed accordingly.

کليدواژه ها

Test fairness, Differential item functioning, Latent class analysis

کد مقاله / لینک ثابت به این مقاله

برای لینک دهی به این مقاله، می توانید از لینک زیر استفاده نمایید. این لینک همیشه ثابت است :

نحوه استناد به مقاله

در صورتی که می خواهید در اثر پژوهشی خود به این مقاله ارجاع دهید، به سادگی می توانید از عبارت زیر در بخش منابع و مراجع استفاده نمایید:
Hossein Karami , 1398 , A latent class approach to differential item functioning , اولين کنفرانس بين المللي مطالعات زبان انگليسي