Select Committee on Education and Employment Appendices to the Minutes of Evidence


APPENDIX 9

Memorandum from the Sutton Trust (HE 78)

  The Sutton Trust gave evidence on access to higher education to the Education Select Committee on 20 June 2000. At the end of the session the Trust was asked to submit any further evidence which it considered relevant to the enquiry. The following submission contains information on five areas which the committee may wish to consider.

    1.  INDIVIDUAL UNIVERSITY ACCESS STATISTICS
    2.  LIFE CHANCES
    3.  RELATIVE PERFORMANCE OF INDEPENDENT AND STATE SCHOOLS
    4.  OPEN ACCESS ANALYSIS
    5.  DISCUSSION OF CRITERIA FOR SELECTION TO UNIVERSITIES

1.  INDIVIDUAL UNIVERSITY ACCESS STATISTICS

  Our analysis of leading universities (see table one) has demonstrated that nine of the top 13 universities have a deviation from benchmark (what numbers should be based on entry qualifications and subjects taught at the institution) of over 10 percentage points and intake ranges from 38 per cent independent and up. As a group these nine universities account for 18,000 students per year, 44 per cent of which are from independent schools and they are an average of 15 percentage points over benchmark.

  We have looked at the 165 higher education institutions in Great Britain surveyed by the Higher Education Funding Council. As can be seen from table 2, outside the top 13 there are only five institutions with an annual intake of more than 500 young entrants that have over 30 per cent independent students and a deviation from benchmark of over 10 percentage points. These are Oxford Brookes, King's College London, Royal Holloway London, Exeter and Newcastle. As a group they account for 9,000 students, 34 per cent of which are from independent schools and are on average 12 percentage points over benchmark.

  This shows that the access problem is largely confined to the top universities and this is where we feel resources should be focussed.

Table 1

TOP 13 UNIVERSITIES

(RANKED BY DEVIATION FROM BENCHMARK)

  
No young entrants
% Independent School Entrants
Deviation from
Benchmark*
  
  
Actual
Benchmark
  
Oxford
2,957
53
35
18
Imperial
1,278
47
29
18
UCL
2,134
43
26
17
Bristol
2,310
45
29
16
St Andrews
964
40
25
15
Edinburgh
2,979
39
26
13
LSE
493
43
30
13
Durham
2,418
38
26
12
Cambridge
2,712
48
37
11
Birmingham
3,314
30
25
5
Nottingham
2,768
32
27
5
Warwick
2,009
23
26
-3
York
1,264
21
24
-3
Total
27,600
39
28
11


  * Benchmark is what numbers should be based on entry qualifications and subjects taught at the institution.

  Data from "Performance Indicators in Higher Education, 1996-97, 1997-98" HEFCE, 1999.

Table 2

OTHER HIGHER EDUCATION INSTITUTIONS

(RANKED BY DEVIATION FROM BENCHMARK)

  
No young
entrants
% Independent School Entrants
Deviation from
Benchmark*
  
  
Actual
Benchmark
  
Royal Free Hospital
81
60
42
18
Oxford Brookes
1,670
33
15
18
SOAS
205
37
34
14
Kings College London
1,753
37
24
13
School of Pharmacy
98
31
18
13
City University
780
28
17
11
Exeter
1,957
34
23
11
Wye College, London
119
31
20
11
Royal Holloway
1,115
31
21
10
Guys & St Thomas's
488
52
42
10
Newcastle
2,869
34
24
10
St Georges Hospital
151
52
43
9
Central School of Speech and Drama
101
19
12
7
West of England, Bristol
3,557
22
15
7
Westminster College, Oxford
159
18
11
7
Leeds
4,243
30
24
6
Manchester
3,701
30
24
6
Reading
1,836
24
20
4
Total
24,900
  
  
  


  * Benchmark is what numbers should be based on entry qualifications and subjects taught at the institution.

  Data from "Performance Indicators in Higher Education, 1996-97, 1997-98" HEFCE, 1999.

2.  LIFE CHANCES

  If we look at life chances, as can be seen from table 3 below, based on A-level performance a student from an independent school should be 14 times more likely to go to a leading university than someone from a poorer area or a lower social class. This in itself is a matter of concern. However what is more shocking is that given the imbalance of entry to our leading universities a student from an independent school is actually 25 times more likely to go to a leading university than someone from a lower social class or a poorer area.

  The figures for entry from independent schools are not the final indicators of the degree of success or lack of success according to family background. We acknowledge that some students at independent schools who gain access to the leading universities are from poorer areas or less affluent social backgrounds. On the other hand, there are a number of affluent students who have been educated to GCSE level in the independent sector who switch to the state sector for A-Levels, particularly into high performing state sixth form colleges. The bottom two rows of table 3 below pinpoint the disparity between the chances of a student from an independent school getting to a top university and one from the bottom three social classes and poorer neighbourhoods. Taking the mid point between the last two rows they show that the probability of getting in to the top 13 universities from a private school is about 25 times greater than for a student from a lower social class or poorer area. If this figure were based on A-Level performance it should be about 14 times.

Table 3

  
Actual
  
Benchmark*
  
Top 13 Universities
No of young
entrants
%
No of young
entrants
%
Total Entrants
27,600
  
27,600
  
From independent schools (7 per cent of families)
10,690
39
7,830
28
From social classes III-V (50 per cent of families)
3,470
13
4,570
17
From low participation areas (33 per cent of families)
1,740
6
2,290
8
Probability of getting in from independent school compared to social classes III-V
22x
12x
Probability of getting in from independent school compared to low participation neighbourhoods
29x
16x


  *  Benchmark is what numbers should be based on entry qualifications and subjects taught at the institution.

3.  RELATIVE PERFORMANCE OF INDEPENDENT AND STATE SCHOOLS

  In 1998 a study was published by the Institute of Education analysing A-level performance of pupils on the Assisted Places Scheme in independent secondary schools and comparing them with pupils from comprehensive schools who were considered to have equivalent ability and family background. They found on average, the independent school pupils gained 1.2 points per A-level more than their comprehensive counterparts.

  This does not take into account the advantages of independent primary schools which may be even greater than the advantages found in secondary schools. Once this is taken into account we believe that independent education from age five results in at least two points per A-level or one grade per A-level performance advantage.

  We believe that the major reason for the difference in performance between independent and state schools is the much higher level of funding per pupil in independent schools. Analysis by the NFER has shown that in secondary schools spending per pupil is 2.3 times greater in independent day schools than in state schools. The majority of costs in schools are staff costs and the higher spending in independent schools results in far fewer pupils for each member of staff.

  Simon Szreter, a lecturer in modern social and economic history at Cambridge, has documented that there are twice as many staff for every pupil in independent schools as there are in state schools. This results in smaller classes, more individual attention for pupils and more non-contact time for teachers enabling them to prepare and mark work.

Table 4:

PUPIL/STAFF RATIOS

Year
State Sector
Independent Sector
1964
24
14
1978
20
13
1988
18
11
1998
20
10


  Primary and Secondary Schools combined

  Source: DfEE

  As can be seen from table 4, the ratio of pupils to staff in the state sector has worsened since 1988 whilst there has been continual improvement in the independent sector ie the gap between independent and state schools is widening. Szreter goes on to argue, "Britain's state education system now has staffing levels that are typical of non EU east European states and those of Central Asia rather than those of our peers in Western Europe and North America".

  The low staffing levels in Britain reflect the low proportion of funds devoted to education as a percentage of GDP. Out of the 28 OECD countries with figures available for 1997, Britain ranks 16th, alongside countries like Hungary, Ireland and the Czech Republic. Britain spends approximately 4.6 per cent of its GDP on education compared to an OECD average of 5.1 per cent.

  Secondary state schools have limited resources which can only be stretched so far. Increased expenditure is required specifically to bring down the pupil-staff ratio in state schools to bring them in line with other comparable countries and the independent sector.

  As far as access is concerned the question is, how much this should be taken into account?

  There is growing evidence that students can be taken from state schools with lower A-Level grades than their independent school counterparts and will achieve at the same level or higher in their finals. The Bristol study referred to in the Trust's previous evidence is one example of this. Arguments have also been made to take into account rank in class when making offers. Subsequent evidence sessions have covered these issues. In summary we are in favour of using a variety of complementary selection procedures backed up with extensive recruitment.

4.  OPEN ACCESS

  As many places at top universities are taken up by entrants from leading independent day schools we would like to make the benefits of attending them available for all who qualify. A study carried out by the Trust in 1998 reviewed by NFER which is currently being updated showed that entry to the top universities comes from very few schools: of the top 133 schools in the Times league table, 100 are independent day schools, 21 are predominantly boarding schools and 12 are state schools. These 100 independent day schools represent almost a quarter of entry to Oxbridge from UK schools and a high proportion of entry to other Russell Group universities.

  The principle of Open Access is that independent schools should be allowed to take any pupil based on merit, irrespective of their ability to pay. The Open Access concept has been inspired by the "needs blind" admissions system used by the Ivy League universities in the United States where students are admitted irrespective of their ability to pay. If accepted, it is the right of that child to go to the school. There is no limit on funded places, which avoids the charity aspect of the Assisted Places Scheme (APS).

  The concept relies on the active recruitment by an Outreach Officer of pupils from all primary schools in the catchment area, many of which may not be traditional "feeder" schools. The Outreach Officer also offers support to schools and families during the application process. Each student is means-tested on the basis of parental income, net assets and "significant other" income, if appropriate (unlike the Assisted Place Scheme, which took account of income only).

  Providing Open Access or "needs blind" admission to these schools would make them available to the most able children rather than just well-off children in the area. Total annual fees for these schools are around £375 million. Annual gross costs to provide Open Access would be around £220 million and, assuming savings to the state sector, net costs would be around £100 million compared to an education budget of £40 billion.

The Belvedere School, Liverpool

  The first Open Access school was launched in May 1999 at the Belvedere School, an independent girls' day school in Liverpool, in partnership with the Girls' Day School Trust who own the school. The school has 650 girls aged 3-18. The Open Access scheme is running in the Senior School, which has 460 girls and fees of £4,353 per year.

  The first Open Access cohort will enter the school in September 2000 and have been selected entirely on merit, regardless of their parents' ability to pay. The entrance procedure consists of a verbal and non-verbal reasoning test developed by NFER-Nelson and also includes a Maths and English test set by the school. The new entrance procedures are designed to evaluate academic potential and not just achievement to date.

  Around two thirds of the girls in the first cohort will be fully or partially funded and the cost, estimated at £200,000 in the first year and rising to £1.7 million annually after seven years, is to be matched equally by the two charities.

5.  CRITERIA FOR SELECTION TO UNIVERSITIES

  Most areas were discussed in the evidence given to the Committee but we would like to add the following points.

  The current university admissions system is premised on projected performance at A-Level. University success is defined as final degree classification. We would question these assumptions: that projected performance at A-Level is the appropriate method of selection and that the objective of selection is to pick students who will get the best final degree class.

A-Level Grades

  UCAS research referred to by Tony Higgins in his evidence to the Committee showed that A-Level predictions are wrong in 65 per cent of cases. Of those, 52 per cent of predictions were two grades too high, 26 per cent were one grade too high and 22 per cent had lower predictions than were achieved.

  More fundamentally, the reliability of A-Levels as the primary indicator of academic quality has been questioned by research at King's College London. This showed that of candidates who have a "true" score of three As (the score a candidate would get if they took the exam repeatedly) over half of the cohort will fail to achieve these results "on the day". This is because of candidates having a bad day, inconsistent marking and the unpredictability of questions coming up on A-Level papers.

  Universities are beginning to look at the usefulness of aptitude tests to complement A-Levels as a selection device. We have funded the NFER to pilot an SAT test in 70 schools this year, as well as funding research at Oxford University to develop aptitude tests for entry to Oxford.

Final Degree Classification

  John Randall, head of the Quality Assurance Agency, has argued that degree classification should be abolished since it is not consistent over time or between institutions. Yet this is what British universities are trying to select for.

  Such a narrow measure of success takes no account of the broader impact a university education provides. In assessing suitable candidates for entry to elite American universities the suitability of an entrant is not measured simply in terms of projected final degree result, but in how successful the entrant will be after graduation and what contribution they will make to society. In America selection is based on a broad definition of merit which includes academic performance, aptitude tests, rank in class, extra curricular activities and other outstanding achievements all in the context of where the student has come from.

  By contrast in this country whilst a range of factors, including extra curricular activities, are considered, the definition of merit is based primarily on performance at A-Level with the objective of picking students who will get the best final degree class.

  The relationship between performance at A-Level and final degree class and between degree class and future success is not at all clear. While we do not pretend to have the answers we think that the assumptions that we should only pick students who will be the most academically successful at graduation and using A-Level performance to select them, needs to be reviewed and examined.

July 2000


 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2001
Prepared 8 February 2001