This study investigated teacher cognition and behavior in a high-stakes, English as a Foreign Language (EFL) school context where no common rating scale exists. 24 EFL teachers at the upper secondary level in Norway were asked to rate the performance of a student taking her oral English exam and to give an account of what kind of performance aspects they pay attention to in the rating process. The study showed that while the raters had the same general ideas of the constructs to be assessed, there were differences in how they perceived the relative importance of these constructs, particularly as regards topical knowledge. The study has implications for language teaching and assessment practices at the intermediate to upperintermediate levels (Common European Framework of Reference, level B1/B2), particularly with regard to the role of topical knowledge.

In order to gain a deeper understanding of what language assessment literacy (LAL) is and how it can be developed, it is relevant to investigate it from the point of view of different stakeholders (Harding & Kremmel, 2016). As for teacher LAL, this phenomenon has largely been studied from the teachers’ own perspectives, leaving other stakeholders’ understanding of the construct underresearched. The current study used an exploratory, qualitative research design to investigate teacher educators’ conceptions of teacher LAL in the Norwegian educational context. Teacher educators are an influential stakeholder group in education, whose knowledge of teaching and assessment can inform our understanding of teacher LAL in important ways. Taking Taylor’s (2013) LAL model as a starting point, this study explored five English teacher educators’ general understanding of teacher LAL, as well as the relevance of the components of Taylor’s model in particular. The study provides important empirical evidence of how teacher LAL may be understood and how Taylor’s model can be further developed.

No abstract