Fairest of them all? : Assessment identity development among Swedish student and novice teachers of English as a foreign language

Sammanfattning: This thesis explores language students and novice teachers’ assumptions and beliefs about assessment and grading, how they view themselves as assessors, and whether their opinions change over time with increased teaching experience. Besides exploring questions such as how student and novice teachers think about quality assessment, there is the question of what challenges teachers face in assessing one language (English), among many others known or spoken in the linguistically heterogeneous classroom. As in many other European countries, the multilingual classroom has brought changes to the Swedish educational context, where the predominantly monolingual classroom was previously the norm. This research project builds on the assumption that there is something that can be called assessment identity, and that competence is gained through dialogical processes between the individual and the context. The research context is focused on Swedish middle school, where the subject of English is taught as a foreign language, and where students are usually 10–12 years old.A mixed methods methodology was used in a sequential explanatory design. Different data collection methods were used in two partly overlapping phases over a period of 2.5 years. In the first phase, an anonymous online survey was answered by 128 students. In the second phase, 17 novice teachers participated in recurrent focus-group interviews and occasional in-depth interviews. The total amount of data analysed comprised 158 surveys and 39 hours of interview recordings where both types of data were analysed mainly with qualitative content analysis. The results display diverse beliefs and assumptions about assessment among the participants; two main themes of fairness and accountability are highlighted. Based on the findings, a four-field model is introduced, where assessment identities are positioned according to their technical/formal or pedagogical approach, and their compliance to the accountability system. This model of language assessment identity positions, LAIP, is seen as the major contribution of the study.Findings visualised using the LAIP model, reveal a need to address assessment identity formation during teacher education before teachers are confronted with all the complexities of assessment. Moreover, the study provides a nuanced understanding of assessment identity formation and accentuates the importance of the immediate context for classroom assessment practices, high- and low-stakes alike. Being able to problematise different teacher roles and assessment identities as part of the learning process of how to become a teacher may counteract feelings of stress and inadequacy. It would also help novices to connect what is learnt at university to what is encountered at work.The thesis aims at problematising different assessment identity positions and how they relate to the aims of equality and quality assessment by which a recognition of different positions would be acknowledged and opened up for discussion. In this way, different assessment practices or assessment identities could be discussed on a system level, away from a more opinionated level. This would mean to accept different positions as legitimate expressions of different ideals and contexts. This broad outlook on assessment identity development could be used to strengthen all types of teachers.