Exploring Test Takers’ Perceptions and Characteristics in Predicting TOEIC Validity Arguments

Abstract

TOEIC is one of the most impactful English proficiency tests in South Korea. This study examined test takers’ perceptions of TOEIC validity following the Assessment Use Argument (AUA) framework and further explored validity arguments’ interconnectedness and effects of test takers’ characteristics. A total of 106 employees of South Korean companies completed an online survey developed for this study, rating their opinions about TOEIC’s beneficence towards society and individuals’ English proficiency improvement, its uses in employment decisions, and score interpretation. In multiple hierarchical regression analyses, this study used the predictors of validity arguments and individual characteristics to predict three criterion variables: social beneficence (Model 1), individual beneficence (Model 2), and employment decision (Model 3). The validity predictors were entered first in step 1, followed by individual characteristics (age, gender, company size, and business partner types) in step 2. The results revealed diverse patterns and interconnectedness of significant validity predictors in three models. The prediction power of individual characteristics differed by models. Business partner types were a significant predictor of individual beneficence and employment decision, while gender was a significant predictor of employment decisions. Further discussions are presented from the perspectives of gender differences and the proxy of English proficiency.