Testing in Pragmatics Research

Abstract

AbstractIt may seem contradictory to talk about “testing” in the context of “research”: after all, testing in a narrow sense is usually done to collect evidence for real‐world decisions and involves stakes for all parties involved in the process, and tests therefore need to be based on a carefully designed interpretive validity argument (Messick, 1989; Kane, 2006; Bachman & Palmer, 2010), which in itself requires a fairly extensive research program. However, testing in a simpler sense is the measurement of the strength of an attribute, which does not necessarily have to be the basis for a high‐stakes decision. Rather, it might be undertaken to answer a research question; for example, to eval..