Description: The recent White Paper Higher Standards, Better Schools for All (2005) has re-ignited debates about selection. In the past selection has often been seen to be about admission policies. However, we argue in this paper that how systems of selection are understood will be related to the model of schooling that proponents in the debate espouse. We, therefore, outline two ideal type models of schooling and then show how these different models determine how the intended outcomes of selection are derived from them. While these ideal typical models are useful for heuristic purposes we need to evaluate the unintended or in the case of a new model of schooling that has been advanced in the White Paper, the likely consequences of these models in the real world. This involves the development of a typology of different kinds of schools and their associated selection systems. In particular, grammar schools, comprehensive schools where zones or catchment areas determine selection and comprehensive schools in a market driven by parental choice and schools which have autonomy over finance, management, pedagogy and selection – the type of Trust school outlined in the White Paper. The interest in a return to a modern version of the grammar school has been rekindled because of the perceived failure of comprehensive schools to promote working class students through to universities and especially the elite universities in the UK. We, therefore, review the kinds of evidence that may enable us to judge which system of selection is likely to be the fairest. Here we should note that when ‘fairness’ is applied to outcomes it has various meanings. For example, different systems of selection will involve trade-offs between particular types of outcome. Outcomes here may refer to the exam results attributable to different types of schooling or the proportion of upwardly mobile working class students entering university or indeed the elite universities such as Oxford or Cambridge. For example, it may be that the grammar school system of selection enabled a larger proportion of working class students to enter the elite universities but that they had a polarising effect on educational outcomes with those attending secondary modern schools having depressed educational results. So systems of selection may also tell us something about the policy priorities and trade-offs that policy makers are prepared to entertain. In reviewing the evidence for different kinds of schooling and their systems of selection we identify where there are gaps in the evidence and in the final section of the paper we outline how one particular research project that we are engaged in can address some of the key outstanding research questions. Methodology: Meta-Analysis and mixed quantitative and qualitative methods Conclusions: That what constitutes the fairest type of selection will be context dependent and that there are always trade-offs in relation to outcomes. However, in the case of England, a comprehensive system with zoning or catchment areas is likely to produce the fairest outcomes in terms of exam results.