Beyond measurement: speaking assessment as social judgment in high-stakes English testing

Speaking proficiency assessment is often treated as a technical enterprise concerned with the reliable measurement of individual ability. In many Asian contexts, however, speaking scores are embedded in high-stakes educational, professional, and mobility regimes, intensifying the social consequences of assessment decisions. This article argues that dominant models of speaking assessment insufficiently account for the interactional and institutional work through which speaking performances are rendered meaningful. Drawing on scholarship in language testing, interactional sociolinguistics, and validity theory, the article reframes speaking assessment as a socially situated practice of judgment rather than the neutral measurement of an underlying competence. It shows how rating scales, task designs, and validity frameworks function not only as technical instruments but also as institutional mechanisms that stabilize particular understandings of speaking, while marginalizing variability and interactional contingency. An illustrative vignette from a high-stakes IELTS speaking encounter is used to foreground how such judgments are enacted within Asian assessment ecologies. The article concludes by considering implications for speaking assessment research and practice in Asia, calling for greater reflexivity about the values embedded in speaking constructs and evaluative decisions.