The role of education and skills in
driving social mobility
Claire Crawford
© Institute for Fiscal Studies
Social mobility and the role of education
• Social mobility: link between SES of parents and children
• Role of education as a transmission mechanism:
– Children’s educational attainment is strongly related to parents’ SES
– There are large wage returns to educational attainment
• Focus on the first today:
– Evidence on the link between educational attainment and family
background in the UK, including over time and relative to other
nations
– What drives these relationships? What types of policies are likely to
be effective at reducing SES gaps in educational attainment?
© Institute for Fiscal Studies
Summary
• Strong relationship between children’s educational attainment
and family background in England
– Differences emerge early and widen as children get older
– Stronger than elsewhere and strengthening over time (until recently)
• There has been some success at encouraging children from
lower SES backgrounds to reach “expected” levels of
achievement
– But children from higher SES backgrounds continue to improve too,
highlighting difficulty of targeting relative measure of social mobility
• Large SES gaps persist in HE participation
– But mainly driven by differences in attainment at 16 and 18, so
earlier intervention is key to widening participation
– Most ethnic minority groups are more likely to go to HE than White
British students, although less likely to attend high status institutions
• Some thoughts on which policies are most likely to improve
education/skills amongst those from the lowest SES
backgrounds
© Institute for Fiscal
Studies
Percentile of the test score distribution
Differences emerge early and widen over time
80
Socio-economic gaps in cognitive test scores
70
60
50
40
30
20
Age 3
Highest
Age 5
Quintile 4
Age 7
Age 11
Quintile 3
Age 14
Quintile 2
Source: Goodman & Gregg (2010), Poorer children’s educational attainment: how important are attitudes and
behaviours?, Report to the JRF, available at: http://www.jrf.org.uk/publications/educational-attainment-poor-children
© Institute for Fiscal Studies
Age 16
Lowest
Stronger relationship between SES and
educational attainment in England than
elsewhere
France
Canada
Portugal
Belgium, Flemish
Iceland
Turkey
Japan
Finland
Denmark
Spain
Switzerland
Belgium, French
Netherlands
Italy
Greece
Norway
Austria
Sweden
Australia
Czech Republic
New Zealand
Ireland
USA
Slovak Republic
Korea
Germany
Hungary
Scotland
England
0
5
10
15
20
25
Source: Hanushek & Woessmann (2011), The Economics of International Differences in Educational Achievement, in
Hanushek, Machin & Woessmann (eds), Handbooks in Economics, Vol. 3, pp. 89-200, North-Holland, Netherlands.
© Institute for Fiscal Studies
30
35
Some evidence that this relationship has
strengthened over time . . .
Degree acquisition by age 23 by parental income
50%
46%
45%
40%
37%
37ppts
35%
30ppts
30%
25%
20%
20%
14ppts
15%
10%
9%
7%
6%
5%
0%
NCDS 1982
Bottom 20%
BCS 1993
Middle 60%
Top 20%
BHPS 1999
Difference
Source: Blanden & Machin (2004), Educational inequality and the expansion of UK higher education, Scottish Journal
of Political Economy, Special Issue on the Economics of Education, Vol. 51, pp. 230-249.
© Institute for Fiscal Studies
. . . at least until recently: gap in terms of %
getting 5 A*-C grades in GCSEs and equivalents
has fallen
% pupils getting 5 A*-C grades in GCSEs and equivalents
90%
35%
80%
30%
70%
25%
60%
50%
20%
40%
15%
30%
10%
20%
5%
10%
0%
0%
2004
2005
FSM
2006
2007
Non-FSM
2008
2009
2010
Difference (RH axis)
2010-2011 figures based on SFR 03/2012: GCSE and Equivalent Attainment by Pupil Characteristics in England.
2006-2009 figures based on SFR 37/2010: GCSE and Equivalent Attainment by Pupil Characteristics in England.
2004-2005 figures based on authors’ calculations using Key Stage 4 and PLASC data.
© Institute for Fiscal Studies
2011
But not when focusing on GCSEs including
English and Maths: equivalents are key
% pupils getting 5 A*-C grades including English and Maths
70%
35%
60%
30%
50%
25%
40%
20%
30%
15%
20%
10%
10%
5%
0%
0%
2004
2005
FSM
2006
2007
Non-FSM
2008
2009
2010
Difference (RH axis)
2010-2011 figures based on SFR 03/2012: GCSE and Equivalent Attainment by Pupil Characteristics in England.
2006-2009 figures based on SFR 37/2010: GCSE and Equivalent Attainment by Pupil Characteristics in England.
2004-2005 figures based on authors’ calculations using Key Stage 4 and PLASC data.
© Institute for Fiscal Studies
2011
Sizeable SES gaps in HE participation remain,
including by type of HE institution attended
% students going to university at age 18/19
60%
52%
50%
37%
40%
28%
30%
21%
19%
20%
12%
12%
10%
7%
4%
2%
0%
Lowest
2
Any university
3
4
"High status" university
Source: authors’ calculations based on linked schools and universities administrative data for the cohorts first eligible
to start university in 2004-05 and 2005-06 (who sat their GCSEs in 2001-02 and 2002-03)
© Institute for Fiscal Studies
Highest
But these differences are almost entirely
explained by differences in prior attainment
% students going to university at age 18/19
100%
25% of
highest SES
group get
top A-levels
90%
80%
Only 3% of
lowest SES
group get top
A-levels
70%
60%
50%
40%
30%
20%
10%
0%
Overall
No A-level
points
1-180 points
(up to DDD)
Highest SES quintile
181-300 points
301+ points
(DDD-BBB) (BBB or above)
Lowest SES quintile
Source: based on work in Chowdry, Crawford, Dearden, Goodman & Vignoles (2010), Widening participation
in higher education: analysis using linked administrative data, IFS Working Paper No. W10/04.
© Institute for Fiscal Studies
What about amongst groups of concern?
(Poor White men, Black men and Pakistani & Bangladeshi women)
•
Ethnic differences in cognitive skills are evident early (Dearden & Sibieta, 2010)
but narrow throughout secondary school (Wilson, Burgess & Briggs, 2006)
•
Most ethnic minority groups outperform whites in terms of HE participation . . .
80%
HE participation by gender and ethnic background
70%
60%
46%
50%
40%
35%
33%
40%
34%
32%
28%
30% 27%
19%
20%
10%
7%
19%
9%
0%
Males
© Institute for Fiscal Studies
Females
38%
30%
What about amongst groups of concern?
(Poor White men, Black men and Pakistani & Bangladeshi women)
•
But things are not so rosy in terms of HE participation at high status institutions
(though most of the difference can be explained by differences in prior
attainment)
35%
HE participation at high status institutions by gender and
ethnicity
30%
25%
20%
15%
10%
5%
11%
10%
9%
7%
1%1%
9%
9%
6%
5%
3%
6%
4%
0%
Males
Females
Source: authors’ calculations based on linked schools and universities administrative data for the cohorts first eligible
to start university in 2004-05 and 2005-06 (who sat their GCSEs in 2001-02 and 2002-03)
© Institute for Fiscal Studies
7%
What drives the relationship between family
background and educational attainment?
• Decompose the gaps in attainment between rich and poor pupils
• Factors will explain a large proportion of the gap if they:
– Are highly correlated with socio-economic status
– Have a large effect on attainment conditional on other characteristics
• Goodman & Gregg (2010) and accompanying studies find that
these factors seem to play an important role in perpetuating SES
gaps:
– Early home learning environment
– Expectations/aspirations for education
– Locus of control (belief that own actions make a difference)
– Behaviour
– Material factors (e.g. access to internet/computer)
• Suggests potentially important role for policy assuming that these
relationships are causal and that these factors are malleable
© Institute for Fiscal Studies
A note of caution: HE expectations amongst
young people are high across the board
90%
78%
80%
70%
60%
50%
64%
58%
50%
51%
40%
30%
20%
10%
0%
Lowest
2
HE participation at age 18/19
3
4
Expectations at age 14
Source: Chowdry, Crawford & Goodman (2011), The role of attitudes and behaviours in explaining socio-economic
differences in attainment at age 16, Longitudinal and Life Course Studies, Vol. 2, No. 1, pp. 59-76
© Institute for Fiscal Studies
Highest
Which types of policies are most likely to
improve attainment/skills amongst the poorest?
• Early interventions have the potential to be more productive than
later interventions
– Strongest evidence is for high intensity interventions, e.g. FamilyNurse Partnership; mixed evidence on lower intensity interventions
– But cannot just intervene once and then sit back; early interventions
are most productive if followed up: consistency matters
• Basic skills (literacy/numeracy) are highly valued in the UK
labour market, suggesting a shortage of such skills
– Very difficult to improve in adulthood
– Good evidence on (cost) effective literacy strategies, e.g.
• The Literacy Hour: structured teaching methods affecting all children
• Every Child a Reader: intensive 1:1 intervention for very lowest achievers
– Improve outcomes in short run but uncertain how long benefits last
© Institute for Fiscal Studies
Which types of policies are most likely to
improve attainment/skills amongst the poorest?
• Teachers matter:
– Recruiting and retaining high quality teachers, and helping them to
pass on their skills to other teachers is vital
– But identifying who will become “good” teachers is difficult; degree
class and experience are not good proxies; more evidence needed
– Also important to remember that schools are only part of the story;
parents/families have at least as great an influence on attainment
• Students need to be supported to make the right decisions
– Choice of GCSE and A-level subjects and what to do at 16
• Later interventions may be better targeted non-cognitive skills
(e.g. leadership and time management) than cognitive skills
– Though evidence remains weak; more is needed
© Institute for Fiscal Studies
Summary
• Strong relationship between children’s educational attainment
and family background in England
– Differences emerge early and widen as children get older
– Stronger than elsewhere and strengthening over time (until recently)
• There has been some success at encouraging children from
lower SES backgrounds to reach “expected” levels of
achievement
– But children from higher SES backgrounds continue to improve too,
highlighting difficulty of targeting relative measure of social mobility
• Large SES gaps persist in HE participation
– But mainly driven by differences in attainment at 16 and 18, so
earlier intervention is key to widening participation
– Most ethnic minority groups are more likely to go to HE than White
British students, although less likely to attend high status institutions
• Some thoughts on which policies are most likely to improve
education/skills amongst those from the lowest SES
backgrounds
© Institute for Fiscal
Studies