In England, as in other countries (e.g. Finland, Australia, New Zealand, United States), recent policies on teacher recruitment have given a high priority to increasing the number of men in the profession, particularly in the primary sector (e.g. Carrington & Skelton, 2003; Lahelma, 2000; Mills, Martino & Lingard, 2004). A declared objective of the Teacher Training Agency's (TTA) current Corporate Plan (2003-2006) is, 'by November 2005, annually to achieve an increase of a further 20 per cent of male trainees on top of the previous year's baseline'. The importance given to this is signalled by a press release which states: 'the TTA has a strategic objective of recruiting more men into primary teacher training and has targeted male applicants through communications such as advertising, press and direct marketing material'. This policy arises from the widely held notion that pupils do better when there is a 'match' between characteristics of pupils and teachers in terms of gender. Policy-makers have generally taken the view that more male primary teachers are needed to provide disaffected, working class boys with exemplars of alternative (i.e. more compliant) forms of masculinity and, thereby, counter negative attitudes towards schooling. Hence initiatives to boost male recruitment have been presented by government officials as a solution to what David Blunkett, when Secretary of State for Education and Employment, termed boys' 'laddish culture' and consequent 'underachievement' (DfEE, Press Notice, 2000/0368). Although policy-makers tend to assume that there is a causal link between the 'gender gap' in attainment, the negativity of some boys towards school and the dearth of men teachers as male 'role models', there is scant research evidence to support this view. For example, Ehrenberg, Goldhaber & Brewer (1995), drawing on a large dataset in the United States (i.e. the National Education Longitudinal Study of 1988), conclude that matching teachers and students by gender (or ethnicity) has little effect on educational achievement. Similarly, Lahelma's (2000) qualitative research in Finland with 13- and 14-year-olds reveals that although students often comment on the lack of male teachers in schools, the issue of gender did not figure prominently in their observations about the quality of teaching that they had experienced. For the large part, the teacher's gender was perceived to be irrelevant. Leaving aside the issue as to whether boys might be alienated or marginalized in institutions dominated by women teachers (Pollock, 1998), we are not aware of any evidence to suggest that men and women teachers relate to pupils differently according to their gender. In this paper, we draw upon evidence from an ongoing ESRC funded study of 7- to 8-year-olds' (Year 3) perceptions of men and women primary teachers. Working with matched samples of pupils (150 boys, 150 girls), from schools in London and the North East of England, interview data are used to explore the extent to which the children relate better to, or draw inspiration from, teachers of the same gender. We also consider whether their teachers (25 men, 25 women) saw themselves as gender specific 'role models' for pupils. Aware of the problematical and contested nature of 'role models' (Carrington and Skelton, 2003), we examine teachers' perceptions of the extent to which children's gender impacts upon their classroom practices. We conclude by exploring the policy implications of the findings.