Background European Union is undergoing considerable demographic change with diverse language groups moving into Europe and also across national borders within the Union. This study investigates in what way the instructional practise in mathematics education is challenged by this increasing diversity of language in the classrooms. The study is aiming to explore a model for description and analysis of instructional practise with the potential to clarify how instruction does effect mathematics achievements for different groups of students. It will also be investigated how this model corresponds with data from TIMSS 2003, 8th grade. Sweden is used as an example. The traditional division in teacher- versus student-centred modes of instruction is challenged by a perspective focusing teacher’s responsibility for knowledge generation and how this responsibility will be expressed in the instructional practise. Results from this study support the possibility to adopt this alternative perspective. Important underlying pedagogical principles to be considered in mathematics education in the multilingual classroom will constitute the theoretical framing of the model. Talking and interactivity between participants are examples of such principles, since it is of importance for student’s mathematical progress if and how they can make use of their languages, experiences and reasoning about the content (Barwell, 2003; Cummins, 1984; Moschkovich, 2002; Setati & Adler, 2000). Another principle concerns how teachers take responsibility for emphasising and preparing the mathematics content. A significant relation between student’s language skills and attained achievements in different academic subjects is stated (Collier & Thomas, 2002), but if students are getting explicit explanations and rules for words and concepts when learning mathematics, they are favoured (Elmeroth, 2006; Jamieson, Chapelle, & Preiss, 2004). Finally a dimension concerning in what way instruction encourage conceptual instead of procedural mathematics appears as crucial (Lester, 2007; Lithner, 2006). When conceptualising and defining class-room practise, a widespread way is to distinguish between teacher- and student-centred modes of instruction. This approach for categorisation of instructional modes has however been challenged by claiming that such one-dimensional perspective on mathematics education could hide other existing important dimensions (Clarke & Xu, 2008; Mok, 2003). This approach could also sketch dimensions not existent in the practise, as e.g. in Sweden where instructional modes due to a huge frequency of “student’s independent work” often is defined as student-centred (Vinterek, 2006), but evaluations show that this instruction is impressed by lack of teachers taking responsibility for knowledge generation (Skolverkets Rapport nr. 323, 2008).

Method The data source for the empirical study was obtained from the TIMSS 2003 study, focusing on mathematics for Swedish students in 8th grade. A latent variable analysis was conducted in order to identify descriptive dimensions of instructional modes which may support students’ mathematical progress. Because of the design effect in survey research it was necessary to account for the hierarchical structure of the population (Hox, 2002), hence a Multilevel Confirmatory Factor Analysis (MCFA) was conducted. A two-level structural equation model approach with three latent factors indicated by items from both the student- and the teacher questionnaire was adopted for the measurement model, and the Mplus and STREAMS software was applied in the analyses. These factors were founded in theories concerning instruction and learning in classrooms depicted of diversity of language, and they were labelled Teacher Responsibility for Content, Student Responsibility for Learning and Challenging Mathematics Content.

Results Results from this study show that the explored model for description and analysis of instructional modes do correspond well with data from TIMSS 2003, 8th grade. In challenging the traditional way of defining class-room practises as either student- or teacher-centred, it has in this study been explored an alternative model concerning teacher’s responsibility both for emphasising and preparing the mathematics content and for making student’s experiences and reasoning about the content explicitly exposed. The model also concerns mathematics content being challenging for the students. The over all model fit shows the potential for the model to further clarify how instruction does effect mathematics achievements for different groups of students.