We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

School of Education

Staff Profile

Publication details for Prof Stephen Gorard

Gorard, S., Siddiqui, N. & Boliver, V. (2017). An analysis of contextual and other indicators of HE students for possible use in widening participation. (Working Paper 3).

Author(s) from Durham


This paper looks at the HESA statistics for all students at university at HEIs in England from 2008/9
to 2011/12, linked to the National Pupil Database (NPD), with records for all pupils in England who
ended Key Stage 4 (KS4) in 2006.This limits the analysis only to those eventually continuing to HE
but not among the 7% who attended a private school (and these two limitations must be borne in mind
throughout). Where a young person is obviously disadvantaged, this can be taken into account in
contextualised admissions to HE. Of the indicators currently available in the generic HESA data,
which give the most accurate assessment of that context – singly or in combination? This paper looks
at missing data, and what is known about students for whom data is missing. It looks at changes in
indicators of potential disadvantage over a student’s lifetime at school and beyond. And it looks at the
relationship between all indicators and student attainment and progress at school and beyond.
Most of the potentially useful or interesting variables in this HESA dataset are so incomplete as to be
useless for the purposes of contextualised admissions. Only occupational class, parental education and
POLAR quintile of residence have sufficient coverage to consider in detail here. Other indicators such
as sex and ethnicity are best used via the linked NPD dataset. Occupational class and parental
education are only known for that subset of students in each age cohort applying for HE, and this
distorts the picture. For example, of students attending HE, Black students are most likely to have
parents with HE, and so any attempt to use parental education as a contextual indicator could be
counter-productive. Around 20% or more of HE students have unknown occupational class or
parental education, and this also distorts the picture. In general, NPD data on all students is fuller and
better quality. The only indicator in the HEAS dataset used here that is worth pursuing further is
POLAR, which could be available for all of each school cohort, and has the least missing data.
However, there is not much difference in the qualified HE participation rates by POLAR (or
occupational group, or parental education). Ethnic minority (of all categories) and EAL students are
considerably over-represented in HE, while males and those living in care are under-represented.
Poor students and those living in poorer areas appear to participate in almost direct proportion to their
earlier qualifications. Students are then gaining HE results in proportion to their prior attainment, and
largely unrelated to their background once these qualifications are taken into account. Finer-grained
analysis by individual HEI and subject is not really possible because of the small cell sizes. Some
aggregation is necessary. There are clear differences in the prior attainment levels of students
attending different types of HEI, and this difference in attainment is reflected in the type of students
attending – with fewer poor students or those living in care in Russell Group Universities.