Girls Rock, Boys Roll: An Analysis of the Age 14–16 Gender Gap in English Schools
| Author | Carol Propper,Deborah Wilson,Simon Burgess,Brendon McConnell |
| DOI | http://doi.org/10.1111/j.0036-9292.2004.00303.x |
| Published date | 01 May 2004 |
| Date | 01 May 2004 |
GIRLS ROCK, BOYS ROLL:
AN ANALYSIS OF THE AGE 14–16
GENDER GAP IN ENGLISH SCHOOLS
Simon Burgess
n
, Brendon McConnell
nn
, Carol Propper
n
and Deborah Wilson
nn
Abstract
We investigate possible explanations for the educational gender gap at age 16. We
employ a national dataset of matched exam results of the cohort of pupils who took
Key Stage 3 tests in 1999 and GCSEs in 2001. Our key result is the sheer
consistency of the gender gap, across both the attainment and the ability
distribution, with regard to both raw outcomes and value added. It is primarily
driven by performance differentials in English. The generality of the gender gap
suggests its source is not within-school practice, which means that policy directed
at improving such practice may be misplaced.
I Intro ductio n
The differential educational achievements of boys and girls in the UK have
generated debate since the 1970s (OfSTED, 2003). Early work focused on raising
the participation of girls, particularly in the traditional boys’ subjects of maths,
science and technology. The recent focus of this debate, however, has been the
perceived underachievement of boys relative to their female peers – the so-called
‘gender gap’ – as measured in the UK by results attained at the four Key Stages
of the National Curriculum. The gender gap at age 16 (when pupils take GCSEs)
is seen as cause for particular concern, illustrated by the degree of media
coverage given to the annual publication of results. While gender-related
differences in performance are also of interest internationally (OECD, 2003;
Elwood and Gipps, 1999; Shaw, 2002), we focus here on the gender differentials
in English secondary schools in measured outcomes at age 16.
The purpose of this paper is to investigate more fully gender differentiated
patterns of performance and to go beyond the ‘average’ statistics that grab the
headlines. We employ a national dataset of the matched exam results of the
entire cohort of approximately half a million pupils who sat the compulsory
exams at age 14 (known as Key Stage 3 tests) in 1999 and took school leaving
exams at age 16 (these are GCSEs or GNVQs) in 2001. We consider the gender
n
University of Bristol, CASE, LSE and CEPR
nn
University of Bristol
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
209
gap, at subject level and in aggregate, with respect to three measures of
attainment: the percentage of boys and girls gaining at least 5 GCSEs (or
equivalent) at grade C or above (%5A
n
-C); their total GCSE points and a
measure of value added between 14 and 16.
1
We investigate patterns of
differential performance both across the attainment (performance at age 16 in
GCSEs or GNVQs) distribution and across the ability distribution, as proxied
by a pupil’s average performance at Key Stage 3. We compare differences across
different types of schools, in terms of good or poor performance, gender mix,
admissions policy and percentage of pupils eligible for free school meals. We
also consider whether Local Education Authorities (LEAs) have an impact on
the gender gap in schools within their boundaries.
Our key result is that the gender gap is effectively constant across all the ways
we cut the data. It is consistent across both the attainment and the ability
distributions, whether we use raw GCSE test scores or value added as the
outcome measure. It is not affected by school quality, nor by a wide range of
observable school characteristics. We show that the gender gap is primarily
driven by performance differentials in English. We find that it is negatively
related to an increase in eligibility for free school meals (a marker of the poverty
of pupils attending a school) and to the proportion of boys in the within-school
cohort.
The rest of the paper is structured as follows. Section II briefly reviews the
related literature and Section III details our dataset. Section IV presents our
results and the final section concludes.
II Previous Literature
First, we review the evidence on the existence of and changes in the gender gap.
We then consider both the potential explanations for and the possible strategies
that have been put forward to reduce it.
In England in 2001, girls outperformed boys on average at each Key Stage
(DfES, 2002a). In all Key Stage 1 tests, taken at age 7, a higher percentage of
girls achieved the expected level than boys, with a much larger gap between their
respective performances in reading and writing than in mathematics. By the age
of 11 (Key Stage 2), girls outperformed boys in English and science, with boys
doing slightly better in maths. At Key Stage 3 (age 14) there was no difference in
performance in science, but the gender gap was maintained in the other subjects.
In 2001 at Key Stage 4 (GCSEs/GNVQs), 44.8% of boys achieved 5 or more
passes at grades A
n
-C compared to 55.4% of girls (DfES, 2002b). An analysis of
1995 data found a similar picture across all Key Stages (Arnot et al., 1998).
Stobart et al. (1992) found a gender difference in maths GCSE in favour of boys,
but noted that this was steadily decreasing. Certainly by 1999, the gender
differential in England in GCSE maths had been reversed (Atkinson and
Wilson, 2003). Similarly, in Scotland, girls’ relative performance in Higher
Grade maths was better than boys for the first time in 1995 (Powney, 1996).
1
This first measure is widely used, by the government and the media, to rank schools
according to pupils’ performance.
S. BURGESS, B. McCONNELL, C. PROPPER AND D. WILSON210
rScottish Economic Society 2004
To continue reading
Request your trial