논증 기반의 언어평가 타당화 연구의 이해

Understanding argument based validation research in language testing: from the high-stakes decision-making contexts

The purpose of this paper is to discuss the applicability of the assessment use argument (AUA: Bachman & Palmer, 2010) to validation in the Test of Proficiency in Korean with foci of value implications and consequences. Ever since Messick (1989) articulated value implications and consequences in the conceptual model of test validity, the social dimensions of testing have been increasingly recognized (McNamara, 2012), and various approaches have been proposed to include the social dimensions into validation research. However, few empirical studies have investigated the social dimensions of testing nor do validation frameworks comprehensively provide procedures for evaluating value implications and consequences. Nevertheless, AUA addresses the limitations in the existing validation frameworks by specifying two claims of decisions for evaluating value and consequences. In this discussion paper, a description of the AUA and a review of empirical validation studies which used the AUA are provided, followed by discussions of the usefulness and applicability of the AUA to the TOPIK (Test of Proficiency in Korean) for evaluating issues of values and consequences. This paper also suggests that analyzing policy documents and media reportage as well as listening to multiple stakeholders’ voices be considered for further research in test validation.