Rating and Reflecting: Displaying Rater Identities in Collegial L2 English Oral Assessment

In this chapter, we adopt a qualitative, interactional approach to raters’ discussions on L2 speaking in situated assessment talk. With a conversation analytic approach, we examine how teachers participating in training for the assessment of L2 oral proficiency and interaction assess and discuss their assessments of learner productions in the high-stakes, National English Speaking Test (NEST) in Sweden. We focus on the interactional management of ‘rater identities’, displayed as raters’ orientations to severity and leniency when delivering and accounting assessments of learner productions in groups of teachers-as-raters. Thus, we examine raters’ reflection-in-action, as different rater identity positionings are claimed, mitigated, negotiated, and linked to assessment claims. Data is drawn from two datasets collected during rater training for the year 6 and 9 NEST respectively. Rater interactions in small groups were video-recorded in both datasets. The rater video-recordings constitute the primary data; individual and group assessments, survey responses, and rater profiles generated through Many-Faceted Rasch analysis (Linacre, 2017) constitute secondary material.