Setting the scene
The World Economic Forum’s Gender Gap Index (GGI) (e.g. WEF, 2016) measures the gap between men and women with respect to: economic participation and opportunity, educational attainment, political empowerment and health and survival. It is often called the “emancipation index” and is a number between 0 and 1 with 1 representing gender equity. In 2016 the top three countries were Iceland, Finland and Norway. Australia was in 46th place. What is particularly concerning is that Australia has barely improved over the last ten years due to a particularly sharp decline in the last two. In the chart below I’ve compared the performance over the last 10 years of Australia (in the gold) and a European “average” (elaborated below).
All stake holders in Australian society should find this chart depressing, those with daughters will find it alarming.
As a mathematician involved in mathematics education I have a particular interest in how Australia measures up with the rest of the world when it comes to the overall mathematical achievement of Australian school children and, just as importantly, the extent of the mathematical gender gap. The Organisation for Economic Co-operation and Development (OECD) produces a triennial survey (e.g. OECD, 2016), the Programme for International Student Assessment (PISA), which covers 15-16 year olds worldwide, assessing them in a range of areas including mathematics. In the next chart I present the last 5 years of PISA results for overall maths scores for Australia as compared to the average of the same group of European countries featured in the previous chart. That we have two time series lines per Europe and Australia reflects the fact that I’ve split out girls and boys PISA scores. It will come as no surprise that in both the European and Australian time series the lower line represents the scores of girls.
It’s difficult to find an optimistic reading of this chart. From either a European or Australian perspective the persistence of the gender gap in mathematical performance is worrying. From an Australian point of view the consistency of the overall decline just adds to the sense that we’re headed in the wrong direction. As I’ll discuss later, recent research suggests that these two charts are telling a consistent if disturbing story, that a societal gender gap, appropriately defined, may be related both to poor performance of boys and girls in mathematics as well as to a gender gap in school mathematics. This research provides clues as to how we as a society can improve the outlook for girls in mathematics and their prospects within the wider society. It is also important as it directly refutes some widely held and deeply entrenched positions related to gender difference in mathematics.
The historical and policy context
In 1971 less than a third of students in tertiary education were female. In 1977, an OECD report(see e.g. McInnis, 1996) described Australia as having the “…highest level of occupational segregation by sex of all the countries it studied.” The desperate need to improve the quality of girl’s education was certainly apparent but it wasn’t until 1987 that the Commonwealth Schools Commission published its “National Policy for the Education of Girls in Australian Schools” (CSC, 1987). This policy, which was embraced by both politicians and educators, was founded on admirable values and principles. Noteworthy is the third of these: “Equality of opportunity and outcomes in education for girls and boys may require differential provision, at least for a period of time.” A clear statement of support for the idea that single-sex schooling might be an appropriate tool to tackle the woeful imbalance of the time and one that I’ll return to in light of recent research. The 1987 National policy also stipulated that five years into the Policy’s implementation a comprehensive review would be undertaken to assess progress towards its objectives. That review resulted in the “National Action Plan for the Education of Girls 1993-1997” (AEC, 1993), the purpose of which was to sharpen the focus of the plan laid down in the original Policy by highlighting certain priorities for the Policy’s further implementation. For example, it stressed the importance of “Examining the construction of gender”. This was an explicit acknowledgement of the pernicious nature of gender stereotyping across society and a call for action in terms of policy making, curriculum design and teaching practice to encourage student awareness of this systemic and therefore somewhat hidden negative influence.
Let me now return to the depressing statistics with which I set the scene and reflect on how some recent research allows us to interpret this data. The persistent gap in mathematical achievement in schools between girls and boys has led to an enormous body of research which has thrown up at least three distinct explanations.
- Greater male variability hypothesis
- Single gender classroom hypothesis
- Societal gender stratification or inequity hypothesis
The first of these is a somewhat euphemistic version of the notion that there may simply be a biological difference which accounts for girls mathematical underachievement. Though one might have hoped that such a view had long since been discredited, as recently as 2008 Machin and Pekkarinen (Machin & Pekkarinen, 2008) concluded “that gender differences in the variance of test scores are an international phenomenon”.
The single-gender classroom hypothesis, the idea that girls do better in all-girl classes, is just as old as the biological difference view and one, as we’ve seen, given some support in Australia’s 1987 National Policy. An example often cited here is that in many Muslim countries (with typically gender-segregated classes) girls are in fact more likely to outperform boys, a phenomenon Fryer and Levitt (Fryer and Levitt, 2010) considered to be “consistent with the hypothesis that mixed-gender classrooms are a necessary component for gender inequality to translate into poor female math performance”.
A recent and comprehensive meta-analysis, that of Kane and Mertz (Kane and Mertz, 2012) not only explains the limitations in the above cited analyses but establishes a coherent framework from which to interpret and understand the current state of the gender gap in mathematical achievement. With regard to the greater male variability hypothesis they point out that if true it should persist across cultures but demonstrate this to be false. As for the single gender classroom hypothesis they explain that in the specific case of Muslim countries what is actually happening is that the girls are in fact performing poorly with the boys doing even worse and that this is in part due to little emphasis being placed on maths in the typically religious schools in combination with the weaker girls being encouraged to leave school early. More generally, analysing data from countries in which 17% or more students attend single-gender schools (including the Muslim countries mentioned) they found no consistent trends. Instead, they found that a particular component of the GGI, economic participation and opportunity, correlated strongly with overall maths scores. In other words, countries in which women and men have comparable earning potential and similar promotional opportunities within the workforce, generally have higher mathematics scores and smaller gender gaps.
WEF (2016), The Global Gender Gap Report 2016, World Economic Forum, Geneva, Switzerland, 2016. (https://www.weforum.org/reports/the-global-gender-gap-report-2016)
OECD (2016), PISA 2015 Results (Volume I): Excellence and Equity in Education, OECD Publishing, Paris. (http://www.oecd.org/education/pisa-2015-results-volume-i-9789264266490-en.htm)
McInnis, S. (1996), Girls, Schools…..and Boys, Promoting Gender Equity Through Schools: Twenty Years of Gender Equity Policy Development, Parliamentary Research Paper 24 1995-96, Parliament of Australia.
CSC (1987), National Policy for the Education of Girls in Australian Schools, Commonwealth Schools Commission, May 1987.
AEC (1993), National Action Plan for the Education of Girls 1993-1997, Curriculum Corporation for the Australian Education Council, 1993.
Machin, S. and Pekkarinen, T., Global sex differences in test score variability, Science 322 (2008), 1331-1332.
Fryer, R. and Levitt, S., An empirical analysis of the gender gap in mathematics, Am. Econ. J.:Applied Econ. 2 (2010), 210-240.
Kane, J. and Mertz, J., Debunking Myths about Gender and Mathematics Performance, Notices of the AMS 59:1 (2012) 10-21.