This paper explores the question of whether school culture are 'determined' bu the social class composition of stduents or by teachers and the management of schools. This is an important qiuestion with respect to the possibilities of school improovment across Europe (see e.g.,Wikeley, Stoll, Murillo and de Jong (2005). In this study the question is raised as to how British pupils with similar social class backgrounds and prior attainment on test scores view themselves and their classmates as learners in two schools of different social class composition. The paper seeks to make a contribution to the debate about the effects of social class composition of schools on pupils’ learning by examining how composition and context affect their view of the 'ideal learner'. How they view themselves as learners in relation to what they consider the ideal may have a significant impact on their educational motivation and aspirations. While this research takes place in primary schools in England, it is of relevance to educational systems in Europe and especially where either makrets in education and/or systems of testing by which to judge school performance have been or may be introduced (Youdell 2006; Au 2008). There are, therefore, implciations for educational policy in Europe. The paper employs Bernstein's (2000) notion of pedagogical discourse to examine the different cultures within the two schools and the role of stduent composition in the construction of these cultures. The analysis is undertaken against the background of the testing culture which both schools have to negotiate.

Using the above theoretical framework, this paper explores and compares the effects of social class composition on pupil learner identities in two contexts: a school with a low social class composition and another with a high social class composition, both in year four. One year group was focused on in order to obtain detailed data, although extensive interviews with school management indicates that this year group was typical of other classes in both schools. The data used for analysis is derived from a three year multi-method project which aimed to uncover the effect of social class on school processes and pupil attainment. A year was spent collecting data using ethnographic methods (a combination of observation, interviewing and collecting documents) in the two schools.

It was found that there were considerable differences in the culture of the two schools and that their construction of the ideal pupil in the two schools was quite different. This raises fundamental questions about the degree to which pupil composition 'determines' school culture and degree of freedom teachers have in changing school cultures and views of the ideal pupils. In turn addressing these issues will illuminate the extent that schools can be 'improved' through various improvement strategies.