Competition, cream-skimming and department performance within secondary schools 1 Peter Davies *a , Shqiponje Telhaj b , David Hutton c , Nick Adnett a and Robert Coe c a Institute for Education Policy Research, Staffordshire University, UK; b Centre for Economics of Education, London School of Economics, UK; c Curriculum, Evaluation and Management Centre, Durham University, UK The performance of departments has been largely neglected in previous studies of subject choice in secondary schools. This is a significant omission because analysis at departmental level enables a fuller assessment of the effects of competition and specialisation on pupil performance. This paper examines relationships between both absolute and value-added measures of departmental performance and the likelihood of students being entered for examination in a subject. It examines these relationships with reference to four option subjects: French, German, Geography and History. It utilises data from an Economic and Social Research Council-funded study which examined trends and patterns in departmental effectiveness using a sample of 664 schools which participated in the Yellis monitoring system for a minimum of five years during the period 1995–2002. Introduction Recent education policy in England has relied upon competition and specialisation between schools to improve pupils’ performance. All state schools are being encouraged to ‘establish distinctive identities through their chosen (subject) specialisms and achieve their targets to raise standards’ (Department for Education and Skills [DfES], 2004). Whilst there is evidence of gains from these policies (Bradley et al., 2001; Jesson, 2001, Burgess et al., 2005), these are mitigated by the association between school competition and greater segregation (Burgess et al., 2004) and difficulties in the estimation of a value-added premium to specialist schools (Schagen & Goldstein, 2002). Moreover, the emphasis on differences between schools is called into question by evidence that variation in the effects of *Corresponding author. Institute for Education Policy Research, Staffordshire University Business School, Leek Road, Stoke-on-Trent, ST4 2DF, UK. Email: p.i.davies@staffs.ac.uk British Educational Research Journal Vol. 35, No. 1, February 2009, pp. 65–81 ISSN 0141-1926 (print)/ISSN 1469-3518 (online)/09/010065-17 # 2009 British Educational Research Association DOI: 10.1080/01411920802041707