Degree performance of Economics students in UK universities: absolute and relative performance in prior qualifications
| DOI | http://doi.org/10.1111/j.0036-9292.2004.00305.x |
| Date | 01 May 2004 |
| Published date | 01 May 2004 |
| Author | Jeremy Smith,Robin Naylor |
DEGREE PERFORMANCE OF
ECONOMICS STUDENTS IN UK
UNIVERSITIES: ABSOLUTE AND
RELATIVE PERFORMANCE IN
PRIOR QUALIFICATIONS
Robin Naylor and Jeremy Smith
n
Abstract
We exploit individual-level data on full cohorts of UK university leavers who were
enrolled for an Economics degree during the period 1984–85 to 1992–93 or the
academic year 1997–98. We analyse the determinants of degree performance and
find that performance depends not only on prior qualifications as measured by
A-level scores, for example, but also on the individual student’s ranking (based
on their A-level scores) within their cohort at their university. The students’
performance is also enhanced by having previously studied mathematics, although
there is no benefit from the prior study of Economics. We find that attendance at an
Independent school has a negative effect on degree performance, and that there are
comparatively few social class background effects. Females are more likely than
males to obtain a good degree.
IIntroduct ion
In recent years, there has been a remarkable growth in the demand for
information concerning student academic performance in UK universities.
Statistics on student performance are of interest to four main groups: potential
university applicants, prospective employers of graduates, the universities
themselves, and public-policy makers in government. This growth in the
demand for information reflects, in large part, the move towards a quasi-market
for higher education in the UK. The introduction of student loans and home
student tuition fees in 1998 together represent one aspect of this process and
have led students to become more discerning over the characteristics of their
selected universities. This demand for information is likely to increase in the
context of the current trend of public policy towards top-up fees. There is likely
to be a particular increase in the demand for information over those factors
which are likely to influence a graduate’s future career prospects, such as degree
classification outcome: see, for example, Smith, McKnight and Naylor (2000)
and Naylor, Smith and McKnight (2002).
n
University of Warwick
Scottish Journal of Political Economy, Vol. 51, No. 2, May 2004
rScottish Economic Society 2004, Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK
and 350 Main Street, Malden, MA 02148, USA
250
In response to growing demand for information, there has been a growth in
the supply both of official and of unofficial statistics on higher education
institutions. Examples of the former include (i) the publication of official
statistics ranking university departments by teaching quality and research
excellence and (ii) the increased availability of university statistics through
HESA, the Higher Education Statistics Agency. ‘Unofficial’ information has
been increasingly disseminated through newspaper articles collating information
from various sources and constructing their own ‘league tables’ of universities.
A common ingredient in these league tables is the proportion of students
obtaining ‘good’ honours class degrees – typically defined as First or Upper
Second Class Honours degrees. Of course, such information on its own, without
any analysis of the determinants of the probability of achieving any particular
level of degree performance or of why this might vary across universities or
subject areas, is of relatively little value. Official league table exercises explicitly
recognise this fact. The Government has developed a set of performance
indicators which include measures of academic achievement (measured as
progression rates) by university. These performance measures are qualified by
‘context’ variables, designed to take account of the possibility that academic
performance in higher education will vary across students by factors such as
previous academic (and socio-demographic) background.
The analysis of student degree performance has a long tradition in the UK.
Many previous studies have focused on differences in degree performance by age
and gender, but typically have not controlled for a wide range of personal or
other characteristics. A number of studies have focused on the correlation
between A-level scores and degree performance. Sear (1983), found a correlation
coefficient of about 0.3 between A-level grades and degree classification, using
grouped USR data. Peers and Johnston (1994) found a correlation of 0.28 across
all universities (including both pre- and post-1992 universities), with a greater
correlation for the older universities. Chapman (1996) finds large differences in
the correlation coefficient by subject area (for example, he calculates 0.47 for
Biology and 0.23 for Politics). He also finds that the relationship is very stable
over time for most of the individual subjects.
On the effects of family and demographic characteristics, Blundell, Dearden,
Goodman and Reed (1997) find that family background characteristics such as
parental occupation and education are significant determinants of higher educa-
tional attainment. Dearden, Ferri and Meghir (2002) report similar findings.
A smaller number of studies have used the individual-level data provided
through the USR to look at degree performance. Smith and Naylor (2001)
analyse the degree performance of all UK university students leaving a UK
university in 1993 and find a significant effect associated with having attended an
Independent school. McNabb, Sarmistha and Sloane (2002) using the 1993
leaving cohort focus on gender differences between men and women and find
that, on average, women perform better than men. Although, interestingly, they
find that the predicted probability of a First Class degree is greater for men than
for women. They find that these differences in performance cannot be explained
by differences in personal attributes. Bratti (2002), again using the 1993 leaving
DEGREE PERFORMANCE AND PRIOR QUALIFICATIONS 251
rScottish Economic Society 2004
To continue reading
Request your trial