49. Memorandum submitted by the Higher
Education Policy Institute
WIDENING PARTICIPATION AND FAIR ACCESS: AN
OVERVIEW OF THE EVIDENCE
EXECUTIVE SUMMARY
1. Widening Participation is largely in
the hands of schools and although universities have a part to
play they are relatively minor actors. By and large, pupils of
all social classes with similar levels of achievement at school
behave in a similar way when it comes to higher education (HE)
entry.
2. We need to recognise that there has been
a lot of progress. Although an oft quoted statistic is that the
gap between social classes in HE has not narrowed, the reality
is that it has narrowed enormously: from the higher social groups
being six times more likely to participate three decades ago to
three times now. This is still a large gap, but much less than
has been suggested, and because the number of people in the lowest
social groups has declined very substantially, even this overstates
the issue.
3. There is a clear perception of a hierarchy
of universities—manifested in the average A-level points
of their entrants. Because school performance and social class
are closely related, the academically most selective institutions
are bound to be socially unbalanced. This hierarchy is well known
both to students and to employers among others, and the institution
attended makes a difference to your life chances. That is why
the Government is right to be concerned with which university
poorer students go to (fair access) as well as ensuring that they
go to university at all (widening participation).
4. Universities need to be more explicit
than they are at present about their basis for selection. Generally,
they admit those students they think are likely to do best in
their institution. That is why they should not simply base their
acceptances on crude A-level scores, although A-level scores remain
important. In order to make a reality of fair access, Universities
urgently need better selection and diagnostic tools than they
have at present.
INTRODUCTION
5. In many ways the issues of widening participation
and fair access are quite distinct, and the measures required
to address them are different. Nevertheless, in some important
respects they are closely related, and in political discourse
they are often conflated. Broadly speaking, widening participation
is a sector-wide issue whereas fair access is one that concerns
individual institutions.
6. Widening participation broadly refers
to the widening of the social groups that benefit from higher
education. Fair access is less straightforward. Most obviously
the term refers to the fairness, or otherwise, of the admissions
processes of institutions (and even this provides room for argument:
for example, whether it is past performance or future potential
that should be considered). But it has also come to refer to the
mix of students in individual institutions. On the one hand, it
is quite possible to widen participation without having fair access
in either sense of the term. On the other hand, it is possible
to concentrate on fair access in a way that detracts from a broader
effort to widen participation.
Widening Participation
7. It is a well rehearsed fact that children
from better-off backgrounds are far more likely to participate
in higher education than children from poorer backgrounds, who
also tend to enter later and are more likely to study part-time.
Also well known is the fact that the absolute gap in participation
between the higher and lower social groups has not narrowed over
the past few decades. Chart 1 shows this clearly. If anything,
the absolute gap has widened: in 1970, 32% of the higher social
groups and 5% of the lower groups participated—a gap of
27 percentage points. In 2000, 48% of the higher social groups
participated compared with 18% of the lower groups—a gap
of 30 percentage points.
CHART 1
8. There is another way of viewing this,
however, which suggests that there has been quite a lot of progress.
Chart 2 shows that in 1970 the higher social groups were more
than six times more likely to participate in higher education
than the lower groups, and this ratio had reduced to just less
than three times by 2000.
CHART 2
9. What is more, the significance of this
gap has reduced enormously, and on a like-for-like basis, even
the 3:1 ratio overstates the gap, relative to the earlier differences.
This is because the number of people in those groups defined as
lower social groups has declined very considerably over the years.
As recently as 1970, nearly 90%[14]
of the population was defined as being in those groups, but that
has now declined to about 40%. As a consequence, the scale of
the problem is very much smaller than it was previously. Whereas
after the war and right up to the 1960s the number of students
from the lower groups would have had to increase more than four-fold
to achieve parity with the higher groups, only a 40% increase
would be required today.
10. So, whether you believe that things
have got better or not depends on whether you wish to show that
the glass is half empty or half full. However, better or not,
it is still apparent that there is a large gap between the different
social groups and it is this that widening participation needs
to bridge.
11. In order to widen participation, we
need to understand the reasons for the current differences. One
welcome development over the past two years has been the recognition
that the disparity in participation is by and large not a problem
caused by higher education nor one which higher education is able
to resolve. Higher education has an important role, of course,
in raising the duration of a pupil's time in education through
initiatives such as summer schools and Saturday schools. However,
it is very much a subsidiary one. Chart 3 shows the disparity
in staying-on rates at school for ages 16plus between different
social groups.
CHART 3
12. The chart shows that at every age, 16,
17 and then 18plus, there is a greater falling away out of education
by pupils from the lower social groups. If this were addressed,
and students from lower social groups were to stay on in greater
numbers and achieve the level three qualifications required for
higher education, then participation would be widened. This is
illustrated in Chart 4, which shows that once they have achieved
the relevant qualifications, students from all social groups are
equally likely to participate in higher education. And there is
other evidence that, once in higher education, students of all
social groups with similar school achievements behave in a similar
way[15].
How well they do in school (and at A level in particular) is the
key determinant of whether they go on to higher education, and
how well they do there.
CHART 4
13. It could be, of course, that higher
education institutions are putting potential students off by their
character or by what they offer. It could also be the prospect
of student debt that is putting them off. The evidence to support
these conclusions is weak. As far as student debt is concerned,
the introduction of fees and maintenance loans in 1998 caused
no change in the pattern of student enrolment in higher education,
either in absolute terms or differentially by social group[16].
It is still too soon after these policy changes to be able to
draw definitive conclusions, but other evidence seems to support
the view that demand for HE is relatively insensitive to price.
In New Zealand, for example, participation by the most disadvantaged
groups (Maoris and Pacific Islanders) increased after prices went
up, as did demand in this country after maintenance grants were
replaced by loans in the late 1980s. Australia and Ireland provide
other examples where changes in the cost of HE (both up and down)
do not seem to have affected demand[17].
14. It is sometimes suggested that young
people from poor backgrounds—especially those without an
academic bent—are discouraged from entering higher education
because the courses that are offered are unattractive to them
and irrelevant to their needs and inclinations. However, because
non-participation in HE follows non-participation in school, it
would be necessary to show that decisions to leave school at 16
were taken because of a view about the nature and offering of
higher education—it is there that the main differences occur.
It seems implausible to think that it should be so. And the evidence
from focus groups and other surveys suggests that the strongest
single factor which leads young people from poorer backgrounds
to decide against staying on in school and going on to higher
education is the desire to earn money, followed by a feeling that
higher education is just not the sort of activity that people
of their class would engage in[18].
It is these attitudes that we need to understand better and to
tackle.
15. One particularly illuminating finding
that has emerged from the work done by Louise Archer is that young
people from poorer backgrounds believe strongly that the rates
of return they will secure from higher education depend to a large
extent on which institution they go to. Many are convinced that
they would not be able to get into a sufficiently good institution
to achieve a high rate of return. This provides a link between
concern with widening participation and fair access. If it were
demonstrable that access to high-prestige institutions was not
closed to young people from poor backgrounds, then a key to widening
participation would be to provide education and information. This
would ensure that pupils are aware of the benefits of higher education
and that it is in their interests to achieve well at school and
to be ambitious in the institutions they apply for.
16. There is one footnote on the question
of widening participation. If we are to widen participation, this
implies necessarily that we must expand the system at the same
time. Especially with a growing population, the alternative would
be to deny higher education to appropriately qualified people
from better-off backgrounds in order to make way for those from
poorer backgrounds. Even if this were the right thing to do, it
would be politically so brave as to be unlikely.
Fair Access
17. What do we mean by fair access? A reasonable
definition would be that no institution should exclude applicants
on anything other than academic grounds, and in particular that
extraneous matters like family circumstances, social class or
ethnic origin should not enter into decisions about admission.
Most obviously, fair access refers to admissions processes. Institutions
must ensure that there is nothing in these that unfairly prejudices
the ability of any applicant to be selected on merit[19].
However, this might be regarded as too narrow a definition. For
example, an institution might have a perfectly fair access policy
in these terms, but be so unwelcoming that it deters people from
some social groups or of some ethnic origins. And even if the
institution does not put them off, some people may be put off
for other reasons[20].
It could be argued that in order to achieve fair access institutions
should be making more active efforts to attract such people. The
social mix of students actually achieved, and therefore their
success in attracting applications, as well as institutions' admissions
processes, are elements in fair access.
18. There is no doubt that our higher education
institutions are very different in their social composition. But
then our higher education institutions are very different in almost
all respects.
19. Chart 5 shows the spread of the lower
social groups between higher education institutions. Some have
nearly 50% of students from such groups, while others have less
than 5%.
CHART 5
20. All other things being equal, this would
appear to provide evidence that access is not fair. Those institutions
at the left-hand end of the scale are much more socially exclusive
than those at the right-hand end. But all else is not equal. Over
the years, in this country, we have developed a clear hierarchy
of institutions in terms of the academic demands they make for
entry. Chart 6 shows the average A-level points achieved by students
attending each institution, and it will be seen how different
they are in this respect. There is a continuum: from institutions
that demand and get 30 points on average down to some whose students
have just 10 points.
21. This situation arises because in this
country students are entitled to apply to any university they
wish—and they take advantage of this to a greater extent
than students in other countries—and institutions are entitled
to select for admission whichever students they wish. This has
resulted in the clear hierarchy demonstrated in Chart 6. By and
large students apply for the most demanding institutions they
think will accept them and institutions select the best qualified
students that apply to them. Because school achievement is so
closely linked to social background, this makes it inevitable
that some institutions will be more socially skewed than others.
Chart 7 shows the relationship between A-level scores and family
background. The difference in performance between social groups
continues a pattern that began with GCSEs. There is a very strong
bias at this stage which makes it inevitable in a meritocratic
system that those institutions making the highest academic demands
will also be the most socially skewed.
22. The academic differences explain most
of the social differences between institutions. Even if university
admissions processes were completely fair, the social profiles
of the most academically demanding institutions are likely to
be unbalanced. By and large nobody seriously suggests that university
admissions processes are deliberately unfair, and indeed, what
research evidence there is confirms that they are not[21].
23. Most universities base their admissions
policies on A-level grades, informed by reports from schools and,
in some cases, interviews. But they need to be explicit about
the basis for their decisions. Generally they are not, but if
pressed most would say that they are admitting those students
that they believe have the potential to do best in their institution[22].
They have to make a judgement about which students have the potential
to do best, and generally A levels are a reasonable pointer. But
A levels do not tell the whole story. For example, it is now well
established that pupils from independent schools do less well
in university than pupils from state schools with similar A level
grades, as is shown in Chart 8.
24. This is not an attack on A levels. A
levels remain an extremely effective way of identifying academic
potential, and we would be foolish to undermine them in any way[23].
But it is clear too that universities need to use a certain amount
of discretion in making their judgements. In the past there has
been concern that when using discretion they have tended to favour
the better-off students. It is unfortunate that now, when some
are exercising discretion with evidence to back up the exercise
of that discretion, they are coming under fire from independent
schools which believe that universities are discriminating unfairly
against their pupils. Universities need to be careful though.
It would be intolerable if some students were excluded from universities
in favour of others because of bias, prejudice or unsupported
theories[24].
The knowledge that, on average, pupils from certain types of school
do worse at university than others is useful general information,
but it does not help in individual cases to decide whether to
admit one applicant in preference to another. Admissions tutors
badly need additional tools in order to be able to discriminate
between applicants.
25. The Sutton Trust has funded a pilot
of the American scholastic aptitude test (SAT), in British schools,
conducted by the National Foundation for Educational Research,
which showed promising results. It is certainly clear that something
further is needed, and the SAT is one tool which may be helpful
for this purpose.
26. This discussion illustrates that the
question of fair access, and of what would constitute a robust
admissions process, is extremely complex. We have to be very careful.
We need firstly to understand where the problems lie. Secondly
we need to be honest about the implications of all this. There
are things that universities ought to be doing to widen participation
and achieve fairer access, but if we misdiagnose the problem we
will prescribe solutions that may do more harm than good.
27. It would be possible to achieve one
notion of fair access by imposing, for example, quotas of different
social groups on institutions. Effectively, if we did this, we
would remove the right of institutions to select those applicants
whom they judge to be the very best. Selection would have to be
on some basis other than ability. This would make the whole application
process extremely haphazard from the student's point of view since,
by and large, students at present form a view of where they stand
in the intellectual pecking order and make their applications
accordingly. The basis for applications would need to be completely
rethought.
28. Fortunately, this seems highly unlikely[25].
What is more likely is that institutions will be given an idea
of the social profile that they "ought" to have, given
the average ability of their intake, the subject studied, and
so on—essentially the HEFCE benchmarks at present—and
then judged according to how close they are to these. However,
it should be noted that, according to HEFCE, these are not targets,
and even with this caveat, we need to be a little careful.
29. The benchmarks are quite good for their
purpose of comparing indicators, but they are not perfect. For
example, they rely on UCAS data, which record 30 points only as
the maximum a student can obtain at A level. But there are some
universities—two in particular—where the average entrant
may have four or five A levels, perhaps all at Grade A. That might
well mean that the benchmarks are misleading or inaccurate in
these cases. What Is more, they are based purely on previous exam
results and subjects studied; they take no account of other factors
or of potential that the raw scores may not reveal.
30. There is one detailed aspect of the
admissions processes in this country that makes higher education
admission more uncertain and less robust than it need be: most
applications are decided before the A-level scores of the applicants
are known. Most applications are decided on the basis of predicted
A-level scores, together with school reports and pupils' statements.
This provides a further, and apparently unnecessary, aspect of
the process where unfairness or bias could creep in. Moves to
a post-qualification admissions system would certainly ease the
job of institutions as they try to deal fairly between applicants.
Conclusion
31. The Government's concern with widening
participation and fair access is really very largely a concern
with young, largely full-time, traditional students. It barely
touches on the question of lifelong learning and the increasing
number of mature students who participate in education throughout
their working lives. Arguably, this is the more pressing educational
and economic concern, but the questions of widening participation,
and particularly fair access, are important for social reasons.
Those who undergo higher education when they are young achieve
the greatest benefits through their lives, and there is a strong
belief that the more prestigious the institution the greater the
benefit[26].
In social policy terms, therefore, widening the participation
of young people, and fair access, are extremely important.
32. It is apparent that higher education
itself has a limited role to play in widening participation. It
is important, but it is by no means the major player. Aspirations
and achievement have to be worked on at a much younger age for
this to translate into widening participation. Nor, despite what
remains to be done, should we lose sight of what has already been
achieved. There is very much greater participation in higher education
than there was, and the number of those affected by the differences
in participation by the different social groups is very much smaller
than in the past.
33. The Government is right to concern itself
with the question of fair access. In terms of the individual it
is not sufficient simply to participate in higher education. That
is important, but the institution they go to is also important,
and in social policy terms, frustrating though some find it, the
Government's concern with the "top" universities is
understandable.
34. There does appear to be a problem in
that, disproportionately, students from poor backgrounds enter
less prestigious institutions, which provide less benefit than
others further up the hierarchy. In part, this is again because
of differential performance at school by different social groups.
But, as discussed earlier, that does not appear to tell the whole
story.
35. Institutions need to ensure that they
get the applications. They also need to be more confident than
it is possible to be at present that they are identifying students
with the potential to succeed best at their institution. This
will not be achieved by imposing quotas on institutions. That
could be done, but it would imply a very different notion of fair
access. It would also imply far-reaching changes in the nature
and characteristics, not just of individual institutions, but
also of the whole HE system.
36. In pursuit of a more acceptable definition
of fair access, institutions should ensure as far as possible
that they admit those applicants most likely to do best academically.
To achieve this, the most urgent need, besides ensuring that more
young people from poorer backgrounds get to the starting point,
is to give admissions staff more effective means than they have
at present to identify potential. A levels play an extremely important
role, but alone they are not sufficient to do the job that is
required of them at the margins. Other diagnostic tools are needed.
March 2003
14 Statistic quoted in the report of the National Commission
Of Inquiry Into Higher Education in 1997. Back
15
For example, their drop out rates are similar-HEFCE Memorandum
to the House of Commons Education and Skills Select Committee. Back
16
This finding-from internal HEFCE research that will be published
shortly-is based not strictly on social class but on social classifications,
based on small areas. Back
17
It should be noted, however, that there is survey and focus group
evidence (eg "Attitudes to Debt" a report for Universities
UK and HEFCE by Professor Claire Callender and the Centre for
Higher Education Research and Information, 2000, ISBN 1 84036
094 1) that pupils from poor families are more debt-averse than
others, and therefore report themselves as less likely to participate
in HE if it involves debt. However, it remains to be seen if these
reported attitudes translate into behaviour. There is no evidence
so far that it has. It could also be that financial considerations
are affecting the institutions that poorer students attend-see
"Supply and Demand in Higher Education"-HEFCE 2001. Back
18
Louise Archer (2003) "The `value' of higher education",
Chapter Six in Archer, L., Hutchings, M. & Ross, A. (2003)
Higher Education and social class: issues of exclusion and inclusion
(London, Routledge Falmer). Back
19
A well-known and obvious example was the discovery in the 1990s
that the computer programme used by St George's Medical School
to manage their admissions coded ethnic minority applicants in
such a way as to make it less likely that they would be selected. Back
20
For example, students from poor backgrounds are known to travel
less far on average to attend HE. See HEFCE Report on Supply and
Demand in Higher Education (2001). Back
21
See, for example, Modood and Shiner "Help or Hindrance? Higher
Education and the Route to Ethnic Equality" (British Journal
of Sociology of Education No 2 2002). In modeling the probability
of receiving an offer, they found that Social Class was not a
significant factor. Ethnicity was, but only when an interaction
variable with institution type was introduced, which showed that
most ethnic minorities were significantly less likely to be made
an offer at pre-92 universities. There was, however, no reported
exploration of an interaction between Social Class and institution
type, so it is not clear whether such a specific effect also exists
for Social Class. Back
22
This is not the only possible basis for selecting students. In
"The Shape of the River" William G. Bowen and Derek
Bok offer three possible bases besides academic potential. 1.
That the applicants, even if they may not do better than others
while at university, have the potential to develop and become
leaders of their community or profession later in life. 2. That
the admission of one applicant in preference to another would
create a better mix of students on a course or in an institution
(this criterion is probably more applicable in some subjects than
others, and may well fall foul of the Human Rights Act in this
country). 3. That the admission of one candidate rather than another
is more likely to increase the university's endowment in due course. Back
23
We know, for example, that A Levels are a very good predictor
of the degree class that students will obtain. (Bekhradnia &
Thompson "Who Does Best at University?"-HEFCE 2002). Back
24
Examples of unsupported theories are that pupils from schools
without a track record of achievement at A level or of sending
pupils to university will do better than their peers with similar
A level grades; and that the same applies to pupils whose parents
have not themselves been through higher education. Back
25
In an interview with the Education Guardian on 4 March, Margaret
Hodge, Minister of State for Higher Education, said that the Government
had no plans to impose quotas on institutions. Back
26
Good research on this is not available, and there are suggestions
that the differences between universities-other than a small number
of the most prestigious-are relatively small and decline over
time. However, this is not a conclusion that can be reached with
confidence, and there is no doubt that the most prestigious universities
do indeed confer disproportionate advantage. Back
| 
