Abstract

In educational research that calls itself empirical, the relationship between validity and reliability is that of trade-off: the stronger the bases for validity, the weaker the bases for reliability. Validity and reliability are widely regarded as basic criteria for evaluating research; however, there are ethical implications of the trade-off between the two. The paper traces a brief history of the concepts, and then describes four ethical issues associated with the validity–reliability trade-off in educational research: bootstrapping, stereotyping, dehumanization, and determinism. The article closes by describing emerging trends in social science research that have the potential to displace the validity–reliability trade-off as a central concern for the evaluation of educational research: the introduction of translational sciences, a shift from significance to replicability, a move from inference to Big Data, and the increasing importance of consequential validity.