ABSTRACT
Over the years language assessment researchers have provided reflective reviews of the current thinking and practices in the use of quantitative methods in language test validation research (e.g., Palmer, Groot, & Trosper, 1981; Anivan, 1991; Douglas & Chapelle, 1993; Alderson, 1991, 1994; Cumming, 1996; Bachman & Eignor, 1997; Hamp-Lyons & Lynch, 1998; Kunnan, 1998, 1999; Chapelle, 1999; Bachman, 2000; Alderson & Banerjee, 2001, 2002; Lumley & Brown, 2005; Chalhoub-Deville & Deville, 2005, 2008). These reviews have summarized the state of the field and initiated deliberation on where the field might be headed or, in some cases, should be headed. They have also described quantitative research methods used in published research, explained refinements to the current methods along with examples of their applications, and revealed advances in educational measurement with reflections of how new research methods might be applied to questions in language assessment and practice. Finally, some of these reviews (e.g., Palmer et al., 1981; Cumming, 1996; Bachman & Eignor, 1997; Kunnan, 1998; Xi, 2008) have related the use of quantitative methods to the types of validity evidence needed to support or refute claims of particular meanings or uses of language assessments. In this regard, as theories of validation and understandings of language constructs have evolved, so have the methods and associated technologies used to provide evidence of validity claims.
