Validity argument for language assessment: The framework is simple

• Published in: Language testing Vol. 29; no. 1; pp. 19 - 27
• Main Author: Chapelle, Carol A.
• Format: Journal Article
• Language: English
• Published: London, England SAGE Publications 01.01.2012; Sage Publications Ltd
• Subjects: Educational evaluation; English as a second language; Expected values; Language; Language assessment; Language Tests; Researchers; Scoring; Test Interpretation; Test Use; Validity; New York; United States > US; Iowa; interpretive argument; argument; validity argument; validity; Kane; validation
• ISSN: 0265-5322; 1477-0946
• DOI: 10.1177/0265532211417211

According to Kane (2006), the argument-based framework is quite simple and involves two steps. First, specify the proposed interpretations and uses of the scores in some detail. Second, evaluate the overall plausibility of the proposed interpretations and uses. Based on experience gained in developing that validity argument, Chapelle, Enright, and Jamieson (2010) outlined four important advances realized in a Kane-style argument. First, the approach provides conceptual tools for outlining an interpretive argument to express the multifaceted meaning of test scores. Second, the essential validation research is defined through a systematic process of examining the inferences in the interpretive argument. Third, the research results are integrated in a presentation which shows how they provide support for the interpretive argument. Fourth, the manner in which the interpretive argument is specified makes clear how the validity argument can be questioned, weakened, limited or refuted by research that supports rebuttals. As Chapelle, Enright, and Jamieson (2010) demonstrate, however, Kane's approach advances professional knowledge by offering specific guidance and conceptual infrastructure for doing so. Such conceptual tools for developing validity arguments (i.e. inferences, assumptions, and claims) help researchers to see where research is needed. At the same time, the opportunity to include more delicate levels of specification, evidence and rationales clarifies the challenges applied linguists face in understanding substantive issues about language assessment. In this paper, the author points out how some of the characteristics of Kane's approach intersect with issues in language assessment. (Contains 1 table and 1 figure.)