Evaluating the use of contextual data in undergraduate admissions

点击翻译按钮获取中文摘要

Universities are increasingly using 'contextual data' to help them identify prospective students from disadvantaged backgrounds with the potential to succeed in higher education. At present, however, it is not clear which kinds of contextual data best capture the socioeconomic circumstances of individuals or what the consequences of particular contextual admissions policies might be for widening participation and student achievement. This project sets out to meet an urgent need for evidence-based guidance on the effective use of contextual data in university admissions.The research questions are:1) Which contextual indicators (such as measures of poverty or learning challenges), singly and in combination, best capture the socioeconomic circumstances of individuals? How strongly associated are the individual-level, neighbourhood-level, and school-level contextual indicators typically available to universities, and how well do these indicators cover for more verified measures of individual socioeconomic circumstances that are typically unavailable to universities (such as parental education or occupation)?2) How well do different contextual indicators, singly and in combination, predict success at university degree level, for comparatively disadvantaged students? Which contextual indicators identify students who are (a) likely to outperform comparably qualified peers from more advantaged environments and who may therefore warrant lower offers (e.g. AAB instead of AAA at A-level)? (b) can be expected to perform as well as comparably qualified peers from more advantaged environments and may therefore warrant prioritisation for standard offers? (c) are likely to perform well in absolute but not relative terms and may therefore require additional support to realise their full potential? How does this vary by degree subject area and academic entry requirements, and what sector wide patterns can be identified?3) How much of an impact can contextual admissions policies be expected to have on the goal of widening participation in higher education? What would be the likely effects of implementing different contextual admissions models on the social composition of the student body, for particular degree subject areas, for courses with different academic entry requirements, and across the sector as a whole? Which particular indicators, used in which particular ways, would enable universities to make the most progress towards the widening participation targets set out in their Access Agreements without compromising student achievement?The research will make use of three major secondary data sources - the National Pupil Database, the Longitudinal Study of Young People in England, and student records data held by the Higher Education Statistics Agency - to establish the ways in which universities could use contextual data to widen participation in higher education whilst maintaining high standards of student achievement.