How is HE differentiated in the UK? Is this
changing?
David Raffe and Linda CroxfordUniversity of Edinburgh
Changing Inequalities and Access to a Differentiated HE SystemSeminar at the University of Edinburgh, 14 June 2013
How is HE differentiated? Is this changing?
• Differentiation as identified through (or linked to) student characteristics
• Among HE institutions and ‘faculties’• Three perspectives:
1. Segregation indices• One student characteristic at a time • Uneven distribution across institutions/faculties
2. Mapping dimensions of variation • Based on combinations of student characteristics• Vertical v horizontal differentiation
3. Relate to ‘established’ hierarchy of institutional sectors• First examine stability and validity of sectors• Then look at uneven distribution across sectors
0
10
20
30
40
50
60
1996 2000 2004 2006 2008 2010
Levels of segregation of independent-school and working-class entrants: England, Scotland and Wales "faculties"
Independent sch: England
Independent sch: Scotland
Independent sch: Wales
Working class: England
Working class: Scotland
Working class: Wales
0
10
20
30
40
50
60
1996 2000 2004 2006 2008 2010
Levels of segregation of black and asian entrants: England, Scotland and Wales 'faculties'
Black: England
Asian: England
Black: Wales
Black: Scotland
Asian: Wales
Asian: Scotland
1. Levels of segregation (summary)
• Higher for ethnic minorities than social class– But downward trend for some ethnic minorities
• Higher for independent school background than social class– And increasing for independent school
• Higher in England than in other home countries• Analysis of institutions rather than ‘faculties’ gives similar
results (and trends) for class and ethnicity – But shows lower segregation for gender, over-21s, non-UK domicile
2. What dimensions can be inferred from student characteristics?
• Dimensions of variation (principal components) based on characteristics of entrants – Eg % high social class, % under 21, mean quals score
• Analyses of differentiation among ‘faculties’ and among institutions give similar results (institutions not subjects are the main basis of differentiation?)
• Analyses based on entrants and applicants give very similar results
• UK-wide, then repeated for England, Wales and Scotland, each cohort, etc
Four main dimensions (components) of variation among UK HE ‘faculties’
Dimension 1
‘Status’2
‘Black and Bangladeshi’
3‘Asian’
4‘Male, non-UK,
Chinese’Main factors (+ve)
QualificationsHigh social classIndep schoolNon-localUnder 21
Black AfricanBlack Carib.Mixed/other ethnicityBangladeshiOther Asian
IndianPakistaniOther AsianBangladeshiChinese
EU (non-UK)Outside EUChinese (UK domicile)
Main factors (-ve)
Low social classFE college background
Female
% of variance
25 14 13 10
UK HE : a hierarchical, stratified system?
• Status dimension (component 1)– Combines social and educational status– But only weakly associated with ethnic background– Very stable over time– Very similar across England, Wales, Scotland– Correlated with hierarchy of sectors (Russell Group/other pre-
1992/post-1992)– Not strongly correlated with subject area (except medicine
and vet medicine)
Horizontal differentiation
• Ethnicity (components 2 and 3) – Different dimensions associated with black and Asian
background students (all UK domicile)– Fairly stable over time– Not strongly correlated with 3-sector hierarchy– Not strongly correlated with subject area– Strong geographical aspect (London)– Similar dimensions in England and Wales, single dimension in
Scotland
• Male, non-UK, Chinese (component 4) – Associated with engineering and technology
3. Differentiation related to ‘established’ hierarchy of university sectors
• Scotland:– Ancient (pre-1600)– Old (other pre-1992)– New (post-1992)
• England:– ‘Golden triangle’ (Oxbridge/London Russell Group as at 1996)– Other Russell Group (as at 1996)– Other pre-1992– Post-1992
• Is the hierarchy (still) valid? Have status differences between sectors weakened?
• How are sectors associated with students’ social & ethnic backgrounds, and has this changed?
Have status differences between sectors become weaker? (as inferred from applicant preferences)
4 institution-level indicators of status …1. Entries as % of applications 2. % of entries through clearing3. Average quals of applicants4. Average quals of entrants
… reveal stable hierarchy:– Gaps between sectors don’t narrow – widen slightly at top in England after 2006– Not explained by subject areas – Variation within each sector doesn’t increase (in England)– Substantial stability in rank-ordering within sectors (in England)– Partial exception: former Scottish CIs in 1990s
2 system-level indicators … preferences of applicants with un/conditional offers from pre- & post-1992 univs… show no decline in preference for pre-1992 Universities
0
20
40
60
80
100
1996 2000 2004 2006 2008 2010
Average % of under-21 entrants from independent schools & professional/managerial class by university sector (Scotland)
Ancient: professional/ managerial
other pre-92: professional/managerial
Post-92: professional/ managerial
Ancient: independent schools
other pre-92: independent schools
Post-92: independent schools
0
5
10
15
20
25
1996 2000 2004 2006 2008 2010
Average % of entrants from ethnic minorities by university sector (Scotland)
Post-1992
Other pre-1992
Ancient
0
20
40
60
80
100
1996 2000 2004 2006 2010
How are sectors associated with student characteristics? Average % of under-21 entrants from professional and managerial
class, by university sector (England)
Golden Triangle
Other Russell Group
Other pre-1992
Post-1992
0
10
20
30
40
50
1996 2000 2004 2006 2008 2010
How are sectors associated with student characteristics? Average % of under-21 entrants from independent schools,
by university sector (England)
Golden Triangle
Other Russell Group
Other pre-1992
Post-1992
0
5
10
15
20
25
1996 2000 2004 2006 2008 2010
How are sectors associated with student characteristics?Average % of Black entrants, by university sector (England)
Post-1992
Other Pre-92
Other Russell group
Golden Triangle
0
5
10
15
20
25
1996 2000 2004 2006 2008 2010
How are sectors associated with student characteristics? Average % of Asian entrants, by university sector (England)
Golden Triangle
Other Pre-92
Post-1992
Other Russell group
How are institutional sectors associated with student characteristics? (Summary)
1. Associated with social class – Little change over time
2. More strongly associated with independent school background– Strong association – Strengthening in England – flight from post-1992s
3. Associated with ethnicity but variable and changing– Very weak association in ‘expected’ direction in Scotland – Strong association for blacks in England – but slight improvement– Asians most strongly represented in ‘golden triangle’ and least
strongly in other Russell Group in England– Declining proportion of Asians in post-1992s in England
Discussion points
• Hierarchy, not just diversity– Strong, stable and monolithic– Associated with institutions more than subjects– Implications for market model• Differentiation by ethnicity as well as social class
– But more variable and changeable• Key variables and processes
– Independent schools– Geography– Subjects?
• Home international comparisons– Differentiation (somewhat) weaker in smaller devolved systems– But same hierarchical structure …– … result of UK-wide system? Shared culture of HE? Universal pressures?
Top Related