Women make up close to half of the labour force in most developed countries, yet occupational segregation remains high (Blau et al. 2013). There are several possible explanations for the lack of workplace integration, including different preferences of men versus women (Goldin 2014). But it is also possible that workplace segregation is caused by gender stereotypes which cause belief distortions about whether a woman can fit into a traditionally male-dominated environment (Bordalo et al. 2016). Related to this explanation, gender role attitudes have been shown to strongly correlate with female employment rates and the gender wage gap across countries (Fortin 2005).
Understanding the link between gender attitudes and occupational segregation is a critical input into policies which have the goal of integrating men and women in the workplace. Contact theory posits that biases and beliefs of a dominant group can be changed via intensive exposure to members of a minority group (Allport 1954). The argument is that mixing groups can break down stereotypes and promote understanding. There have been several studies which have randomly assigned college roommates of an opposite race or ethnicity to study the predictions of contact theory (e.g., Boisjoly et al. 2016, Carrell, Hoekstra and West 2015, Finseraas and Kotsadam 2017). However, until recently, there had not been similar field experiments which assigned men and women to work and live together to see whether gender attitudes and real-world outcomes could change.
A field experiment during army boot camp
In recent work (Dahl et al. 2020), we explore whether integration can change gender attitudes and related outcomes. We conducted a field experiment which randomly assigned women to teams which were traditionally all male. Our setting is the Norwegian military, which has only 15% women in its ranks (a fraction which is similar in the US). The Norwegian Army’s North Brigade has been at the forefront of integration and since 2010 all of its battalions have had mixed-gender squads. With the help of the Norwegian Defense Research Establishment, in 2014 we randomised female Army recruits to some squads but not others during boot camp.
Boot camp, also known as basic training, lasts for eight weeks. It is an intense period of training which involves both field exercises and classroom time. The objective is to prepare soldiers physically and mentally for military service. Recruits are grouped into squads, which are typically composed of six soldiers. These squads train together as a team and live in the same room. Since recruits are not allowed to leave base during boot camp, they spend most of their daytime and night-time hours together.
In our setting, treated men live and work closely with women in their squad (usually two women and four men), while untreated male recruits will only have other men in their squad (usually six men).
We conducted surveys of gender attitudes before boot camp, at the end of boot camp, and six months after boot camp. We also tracked the military occupations of soldiers after boot camp and the education and occupational choices made after the end of service.
Short-run changes in attitudes and occupations
We find that intensive contact with female recruits during boot camp causes men to have more egalitarian attitudes in the short run. At the end of boot camp, treated men are 14 percentage points more likely to think mixed-gender teams perform as well or better than same-gender teams. This is a 26% increase relative to the control mean of 52%. Treated men are also 8 percentage points more likely to believe men and women should share household work equally, relative to the control mean of 66%. And finally, treated men are 14 percentage points more likely to not completely disavow feminine traits, relative to the control mean of 63%. These changes are equal to between one-third and one-half of the gap in mean attitudes across genders.
These are sizable changes in attitudes, but do they translate into differences in real outcomes? To examine this question, we look at the gender makeup of military occupations after boot camp ends. We find that men in mixed-gender teams during boot camp go on to serve in military occupations with more women in them – there is a 22% higher fraction of women in the occupations of treated men relative to the controls. Since assignments are based on soldier preferences, ability, and staffing needs, we interpret this result as at least in part reflecting the altered preferences of recruits.
A natural question is whether the short-run changes in gender attitudes and occupational choices persist. After boot camp ends, individuals are assigned to new squads whose members serve in a variety of military occupations. An important consequence of the reshuffling is that both the extent of exposure – in terms of number of women – as well as the intensity of exposure – in terms of teambuilding and spending time together – are typically lower after bootcamp. We find that treated men’s gender attitudes converge to those of the controls, eliminating the short-run differences observed at the end of boot camp. Congruent with this result, we find no difference in future education or occupation choices in the long run – treated men do not choose fields of study or occupations with a larger fraction of women in them.
We caution that these results do not mean that integration cannot change attitudes and choices in the longer run. Instead, the sharp contrast between the short- and long-run results suggests that to maintain changed attitudes and behaviours, exposure likely needs to be maintained for a longer period. Our intervention, while intense during boot camp, was relatively short compared to the overall military experience. And our setting is an extreme example of a male-dominated environment, where attitudes may be the most difficult to change permanently without continuing exposure.
Will integrating women hurt the military?
Our research provides general insights into how integration can matter for attitudes and occupational choices. However, it also has direct implications for the integration of women into the military. We also studied how treatment affects male soldiers’ performance and satisfaction with service. We find no evidence that integration hurt squad members on these two margins, either during boot camp or subsequent service. Treatment also did not affect real outcomes – there was no effect on promotion probabilities or performance evaluations conducted at the end of service. Policymakers in several countries have argued that having women serve alongside men in the military would hurt performance and cause dissatisfaction among male recruits, but this evidence suggests otherwise.
Authors’ note: This column is based on research by the authors (Dahl et al. 2020), which received generous financial support from the Norwegian Research Council.
Allport, G W (1954), The Nature of Prejudice, Addison-Wesley.
Blau, F D, P Brummund and A Y-H Liu (2013), "Trends in Occupational Segregation by Gender 1970-2009: Adjusting for the Impact of Changes in the Occupational Coding System", Demography 50(2): 471-492.
Boisjoly, J, G J Duncan, M Kremer, D M Levy and J Eccles (2006), "Empathy or antipathy? The impact of diversity", American Economic Review 96(5): 1890-1905.
Bordalo, P, K Coffman, N Gennaioli and A Shleifer (2016), "Stereotypes", The Quarterly Journal of Economics 131(4): 1753-1794.
Carrell, S E, M Hoekstra and J E West (2015), "The Impact of Intergroup Contact on Racial Attitudes and Revealed Preferences", NBER Working Paper No. 20940.
Dahl, G B, A Kotsadam and D-O Rooth (2020), "Does Integration Change Gender Attitudes? The Effect of Randomly Assigning Women to Traditionally Male Teams”, NBER Working Paper No. 24351 (originally issued 2018).
Finseraas, H and A Kotsadam (2017)(, "Does personal contact with ethnic minorities affect anti-immigrant sentiments? Evidence from a field experiment", European Journal of Political Research.
Fortin, N M (2005), "Gender role attitudes and the labour-market outcomes of women across OECD countries", Oxford Review of Economic Policy 21(35): 416-438.
Goldin, C (2014), "A grand gender convergence: Its last chapter", The American Economic Review 104(4): 1091-1119.