Think women aren’t good at maths? Depends on where you’re a woman.
Do you know the difference between Celsius and Fahrenheit? Can you interpret information from line graphs in news articles? Calculate how many wind turbines would be needed to produce a certain amount of energy (given the relevant information)?
These may seem like basic tasks, but if you are a woman living in the UK, Germany or Norway, the chances are you would struggle with them more than a comparable man. If you live in Poland, however, you might even outperform a male counterpart.
Why this variation in skills, and why does it appear in some countries and not others?
For some, these findings, from the 2011 international survey of adult skills, run by the OECD, will confirm their existing beliefs. In spite of women being more academically successful than men, the perception that ‘women can’t do maths’ is widely held. A recent experiment  showed that both genders believe this to be true: both male and female subjects were more likely to select men to perform a mathematical task that, objectively, both genders fulfil equally well. In her successful book ‘The Female Brain’, Louann Brinzedine argued that women are ‘hard wired’ for communication and emotional connection, while men’s brains are oriented towards achievement, solitary work and analytical pursuits.
Another camp of social scientists argue that such narratives misrepresent the facts. Janet Shibley Hyde and colleagues insist that, at least in the United States, men and women’s cognitive abilities are characterised by similarity rather than difference. Reviewing findings across many studies of gender differences on standardised mathematics tests, these authors found that ‘even for difficult items requiring substantial depth of knowledge, gender differences were still quite small’.
The fact that gender differences show up on an international survey of numeracy skills is a puzzling addition to an already contentious picture. Of course, not all maths tests are created equal. The difference may in some way reflect the way the survey conceptualises skills. Distinct from mathematical ability, applied numeracy skills are described as:
‘the ability to use, apply, interpret, and communicate mathematical information and ideas’.
Crucially, individuals who are ‘numerate’ should be able to apply these abilities to situations in everyday life. Perhaps these ‘everyday’ maths skills are more biased by gender than the measures used in other studies?
Numeracy: the ‘new literacy’
I argue that we should take these gender differences seriously. More and more, jobs now require numeracy skills, both to perform basic tasks and to support ICT skills. Outside work, numeracy skills are increasingly required to make sense of the world around us. They help us to grasp concepts such as interest rates and inflation, which help us to deal with money. Moreover, according to the British Academy,
‘the ability to understand and interpret data is an essential feature of life in the 21st century: vital for the economy, for our society and for us as individuals. The ubiquity of statistics makes it vital that citizens, scientists and policy makers are fluent with numbers’.
The importance of numeracy has been recognised recently in the UK with the establishment of an All-Party Parliamentary Group for Maths and Numeracy, the National Numeracy charity, and initiatives such as Citizen Maths.
Particularly curious is the large variation across countries in the size of the gender difference. Figure 1, below, shows that, among adults aged between 16 and 65, the male advantage in applied numeracy skills is particularly large in Germany, the Netherlands and Norway, while it is virtually non-existent in Poland and Slovakia. The graph shows raw differences in average skill scores; although gaps reduce somewhat when controlling for age, family and immigration background and education, they remain.
Figure 1: Mean numeracy skills by gender, International Survey of Adult Skills, 2012
Any genetic component is unlikely to vary internationally , suggesting a substantial role for cultural, institutional or economic factors that vary across countries.
My PhD study
Given that the survey tests adults who have many experiences behind them, isolating the causes of gender differences and cross-country variation is far from simple. We are socialised into gendered preferences, motivations and skills from our earliest years . We go on to make gendered choices in our educational lives, our careers and our leisure activities. All of these life domains contribute to the skills we end up with in adulthood. To some, a choice-based explanation is unproblematic; determining one’s own destiny is a core value in many contemporary societies. However, this side-steps the question of where preferences come from. Skill differences in adulthood may well reflect individuals’ choices; however, the choices themselves are likely to be influenced by a complex mixture of cultural, educational, economic and institutional factors; which vary in their salience across countries.
In my PhD study, I focus on education and labour market explanations. A key task for my research is disentangling why gender differences in numeracy skills are relatively large in countries typically considered ‘gender egalitarian’. For example, Scandinavian countries consistently top the rankings of the World Economic Forum’s Global Gender Gap Report, and are held up as bastions of gender equality. Yet Norway, Sweden and Denmark show among the largest gender differences in adults’ applied numeracy skills. Poland, Slovakia and Spain are not known for being particularly progressive on gender equality, yet they show among the smallest differences.
School and skills
One possibility is that gender differences arise from what girls and boys are exposed to while they are at school. Despite a similar basic structure, education systems across the world differ in the extent to which subjects are optional or compulsory. For example, in the UK, mathematics was not compulsory in upper secondary education until recently; whereas in other countries this has long been the case. Where numerate subjects are not compulsory, they may be less valued, and this could have created more scope for gender to affect subject and career choices. There is also wide variation in the types of mathematics learning boys and girls are exposed to across countries, as well as between schools and classes within countries.
Work and skills
Another possibility is that differences in skills are related to the types of jobs that women and men pursue once they leave education. In the majority of countries in the study, occupational segregation is still widespread in spite of female’s superior performance in education, and is partly to blame for the continuing gender pay gap. Gender occupational segregation is particularly rife in Scandinavian countries, although this has been improving in recent years . Countries with strong gender segregation in jobs promote gender norms about what careers are appropriate and accessible for men and women. This is likely to drive the early choices that contribute to skills in adulthood. In contrast, in some countries gender segregation of jobs is less pronounced, which may set more egalitarian norms for skill development. Moreover, given the link between more demanding, highly skilled jobs and skill development in adulthood, concentration into lower paid, more routine jobs could affect the extent to which women are able to gain skills at work. In some countries’ labour markets, women may perceive weaker incentives to develop mathematical skills than their male counterparts, preferring more typically ‘feminine’ ones, such as communication and literacy skills.
In my view, skills gaps are among the hurdles we need to overcome in order to attain full economic equality between men and women. Using international comparisons, my research aims to locate gender differences in applied numeracy skills within a broader, institutional context. This is important both to correct the assumption that differences are ‘fundamental’ or ‘natural’, and to design effectively-targeted policies to equalise skills. I use a variety of quantitative techniques in my research which isolate factors associated with gender differences at both the individual and country levels. This should broaden the discussion beyond the common focus on encouraging girls to make gender ‘atypical’ choices in education, which neglects both males and the broader social context in which skill differences develop. Moreover, while there is a large amount of research on gender and education, skills inequalities among adults are less often addressed. Yet they affect adults’ lives in profound ways . I hope to show some of the ways in which skill differences among adults are not fixed by early experiences and biology, but malleable according to social context.
 Reuben, E., Sapienza, P. and Zingales, L. (2014). ‘How stereotypes impair women’s careers in science.’ Proceedings of the National Academy of Sciences, 111 (12), 4403-4408.
 Hyde, Janet S., et al. (2008) Gender similarities characterize math performance. Science 321 (5888) pp. 494-495 (p.495)
 Penner, A.M. (2008) Gender differences in extreme mathematical achievement: An international perspective on biological, social, and societal factors. American Journal of Sociology 114 (supplement) S138–S170.
 Maccoby, E. E., and D’Andrade, R. G. (1966) The development of sex differences. Stanford University Press.
 Bettio F and Verashchagina A (2009) Gender Segregation in the Labour Market: Root Causes, Implications and Policy Responses in the EU. Brussels: European Commission.
 Carpentieri, J. C., Lister, J., Frumkin, L., & Carpentieri, J. (2010). Adult numeracy: a review of research. London: NRDC.