Western standards behind the gender equality paradox

Previous research has asserted that women and men in gender-equal countries differ more in their preferences than women and men in less equal countries, for example by making more traditional educational choices. This relationship is known as the gender-equality paradox. However, new research from Uppsala University now shows that it is not possible to draw these conclusions from the data studied. The paper, published in the journal PNAS, shows that the question is based on Western perspectives and conditions and cannot be applied to other countries.

“I became interested in this paradox when I noticed that all the more gender-equal countries were Western,” says Mathias Berggren. “That made me want to investigate whether the methodology really held up and how they had designed their studies. There are major data problems, for example with personality measures developed in Western countries.”

The so-called gender-equality paradox has been a recurrent but controversial thesis in social research over the past decade. It assumes that when both women and men have more freedom to do what they want, i.e. when gender equality is high, they are more prone to adopt traditional gender roles. This could explain why, for example, female students in Sweden are less likely to apply for engineering, technology or maths programmes than women in less gender-equal countries such as Algeria.

Men and women are innately different, and more gender equality only makes it easier for them to show their true colours, the thesis goes. The paradox has received widespread attention both in research and in social media, and has led some to argue, for example, that gender equality efforts may be dysfunctional or meaningless. In recent years, the thesis has been questioned, and now researchers at Uppsala University have decided to thoroughly scrutinise the methodology to discover potential flaws.

The research team hypothesised that the link between more gender equality and larger gender differences is actually due to Western measures not working as well in other cultures. In other words, when – usually Western – researchers have found large gender differences for some characteristic in, say, the United States, it is often possible to find the same phenomenon in countries with similar cultures, for example in Western Europe.

“However, it does not work as well in countries that differ more from the West. For example, personality can manifest itself differently in different cultures and in some cases reliability drops significantly outside the Western world. This makes it very difficult to find the same large gender differences – because the search is simply carried out in the wrong way. For example, statements such as ‘I tend to vote for liberal politicians’ may systematically differ between people with different personalities in the West, but do not necessarily do so in countries with limited voting rights,” Berggren explains.

In the article published in PNAS, the researchers re-analysed data from multiple studies that demonstrated the gender-equality paradox.

When they took account of cultural clusters of countries as well as reliability and other statistical indications of measurement quality, the correlations between increased gender equality and larger gender differences disappeared. The researchers also tested a number of other variables besides gender equality, such as indications of economic development, but there were no general associations with gender differences in personality and the like after statistically controlling for cultural clusters and measurement quality.

The researchers chose to look at a wide range of other variables, such as cognitive abilities, school performance and aggression. Again, after controlling for reliability and cultural clusters, they did not observe any major differences associated with higher gender equality. In all, millions of participants from over 130 different countries were analysed.

In their study, the researchers go on to show that fictional explanations also exhibit the same association with gender differences as the evolutionary thesis, as long as the variables are similarly linked to the Western world.

Overall, the article suggests that there may be no connection at all between greater gender equality and larger gender differences.

“There is currently no strong evidence that increased gender equality leads to women and men more clearly revealing any underlying distinctions. We show that this type of study provides very poor evidence for such conclusions. If researchers want to investigate the effects of gender equality reforms, they need methods that better show cause and effect, for example using extensive data over time and/or natural experiments,” Berggren concludes.

Fact box: The gender-equality paradox

Since 2001, scientific studies have been presented showing larger gender differences (initially in personality) in more gender-equal countries. The researchers have compared countries with high and low levels of welfare and high and low levels of gender equality (according to UN indicators, for example). The finding that has received most attention is that in countries with a low gender equality index, more women study subjects leading to traditionally male occupations (STEM programmes: Science, Technology, Engineering, Mathematics), than in more gender-equal countries such as Sweden. This finding is known as the gender-equality paradox and has been used in many contexts. Among other things, it has been adduced as evidence that active gender equality policies do not pay off and that it is not worth countries working to promote equal opportunities between women and men.