Gender differences in GCSE - Cambridge Assessment

1 Gender differences in gcse Tom Bramley, Carmen Vidal Rodeiro & Sylvia Vitello Cambridge Assessment Research Report 20th October 2015 2 Author contact details: ARD Research Division Cambridge Assessment 1 Regent Street Cambridge CB2 1GG Cambridge Assessment is the brand name of the University of Cambridge Local Examinations Syndicate, a department of the University of Cambridge . Cambridge Assessment is a not-for-profit organisation. How to cite this publication: Bramley, T.

2 , Vidal Rodeiro, , & Vitello, S. (2015). Gender differences in gcse . Cambridge Assessment Research Report. Cambridge , UK: Cambridge Assessment . 3 Introduction This report presents some analysis of gender1 differences in examination outcomes, focussing mainly on GCSEs taken in England. This is a topic of perennial interest, because it connects with a number of issues, including: Gender equality in the workplace (opportunity and remuneration); The proportion of women in STEM related jobs; Gender choices and outcomes in subjects studied at HE level; Gender stereotypes; Whether there are Gender differences in cognitive ability, and the role of genes and the environment in creating or maintaining them.

3 Whether certain types of school structure ( single-sex schools), curriculum, teaching style or Assessment style are better suited to one or other Gender ; Whether there are Gender gaps in exam performance and whether such gaps are closing or widening. We start by briefly summarising recent research on Gender differences in cognitive ability and achievement in academic examinations. The bulk of the report presents some analysis of Gender differences in subject choice and examination outcomes, mainly at gcse . Background A large proportion of research on Gender differences in cognitive ability and educational attainment has focused on mathematics, verbal performance and the cognitive abilities related to them.

4 Mathematics and related abilities Historically, the widely held view is that males outperform females in tests of mathematical ability (Halpern, 1986; Hill, Corbett, & St Rose, 2010; Hyde, Fennema, & Lamon, 1990). Early reviews of empirical research in this field concluded this was a robust finding (Halpern, 1986, p. 57) or, at least, it was one of several fairly well-established Gender differences (Maccoby & Jacklin, 1974, p. 352). Although subsequent formal analyses of these data indicated that Gender differences in mathematical ability were often small in size (Hyde et al.)

5 , 1990), recent research continues to show some differences but they vary according to certain factors, including level of mathematical ability, type of mathematical ability and examination format. There is little evidence of a male advantage in high school mathematics tests in either the US or the UK. In the US, trivial differences between boys and girls mathematics results have been found in all school years between Grade 2 (7 8 year olds) and Grade 11 (16 17 year olds) (Hyde, Lindberg, Linn, Ellis, & Williams, 2008). At the end of high school, it is girls who have been achieving, on average, higher Grade Point Average points (in mathematics and science combined) than boys since at least the 1990s (Hill et al.

6 , 2010). Similarly, the UK Department for Education and Skills (2006) report on Gender and education emphasises a female advantage amongst students gaining an A*-C grade in GSCE mathematics in England, although this pattern has changed from the pre-1991 pattern when males outperformed girls (DfES, 2006a). Despite a female advantage in school results, males in the show a persistent trend to perform better in high-stakes examinations such as in the mathematics sections of the SAT and ACT examinations which are taken for entry to universities in the (Hill et al.

7 , 2010). Recent international standardised tests examining 10-11 year-old (TIMSS), 13-14 year-old (TIMSS) and 15 year-old 1 In fact these are sex differences but we follow the usual practice of describing them as Gender differences . When describing data from GCSEs and A levels we use boys and girls but in some other contexts we use males / females as appropriate. 4 (PISA) performance show inconsistent Gender differences across countries, although most countries either show a male advantage or no Gender difference.

8 There is more consistent evidence that, if there is a male advantage for mathematics, it occurs amongst the high achievers rather than throughout the continuum of ability levels. For several decades, males have been overrepresented amongst high mathematics performers in high-stakes mathematics in the , although the male to female ratio has decreased over time and has stabilised to a much lower rate since the 1990s (Miller & Halpern, 2014; Wai, Cacchio, Putallaz, & Makel, 2010). Male overrepresentation has also been found amongst 15 year olds who were high mathematics achievers in the PISA 2012 international test in most countries and economies that took part (OECD, 2014).

9 In contrast, this Gender gap was not found for science performance. The PISA results also showed that the proportion of girls was marginally larger amongst the lowest mathematics achievers. Several reasons have been proposed to explain why males outperform females in certain mathematics tests but not others and why this advantage is found especially amongst high achievers. Spelke (2005) argues that Gender differences could result from a sample bias; for example if fewer boys take a particular test they may represent a more selective ( , higher achieving) sample than the girls who take the test.

10 Gender differences may also arise at test level because tests consist of a variety of items assessing a complex mix of capacities and strategies that vary differentially favour boys compared to girls depending on their composition (Spelke, 2005, p. 954). This hypothesis is supported by empirical evidence showing that boys and girls outperform each other on different mathematical operations and related cognitive abilities. Researchers have highlighted various trends in the literature, including that males tend to perform better on certain spatial ability tasks, mental rotation, reasoning and problem-solving tasks, geometry, trigonometry and statistics while females tend to perform better on algebra and test items that require understanding numerical patterns or representing quantities and quantifiable attributes of objects (Ceci, Williams, & Barnett, 2009; Liu & Wilson, 2009; Miller & Halpern, 2014).