Validity in educational and psychological assessment

My first engagement with the theoretical debates which underlie all educational assessments was a presentation delivered by Stuart Shaw, on this, his then forthcoming book. I was employed as a test designer at the time. Whether creating or grading tests, using exam results to make decisions, or writing essays or theses ourselves, practitioners and students all hold some interpretation of “assessment validity”. In an accessible, statistics-free treatment, Newton and Shaw first distinguish assessment validity from related concepts (such as research validity), before tracing its evolution through the primarily US-led assessment research tradition over the past century. As doctoral students we hope to tap into issues at the cutting edge of the field, and Newton and Shaw’s book enables this, providing the background and theory needed to contribute to contemporary assessment research.

In their final chapter, sidestepping the long-running “lexical dispute” (183) over how exactly validity relates to various types of consequence, the authors instead seek to accommodate all of these concerns within the broader concept of evaluation. Previous theory is synthesised into a “cogent, comprehensive and useful” framework (224) for judging the suitability of tests and test policies. The assignment of meaningful scores in relation to hypothesised constructs, the decisions made using those scores, and other effects not operating through score assignment (such as washback on learning) are each clearly distinguished, and then further divided between technical and social dimensions.

Researchers can sometimes talk at cross-purposes over assessment, but Newton and Shaw’s inclusive history and resulting framework should facilitate communication between many different perspectives – including psychometricians, policy-makers, and teachers, as well as those mounting a “passionate criticism of testing policy” (217). The book has been invaluable in my experience as an early career researcher, and its call for more “economic analysis of assessments” (195) has inspired a forthcoming article on marketability of test features.

Ricky Jeffrey
Faculty of Education, Northeast Normal University, Changchun, China
[email protected], @Ricky_Jeffrey
© 2017 Ricky Jeffrey
http://dx.doi.org/10.1080/00131911.2017.1291210