Beyond employment and earnings, job training can increase aspirations among women who lack the confidence to see themselves as successful

Gender-based occupational segregation – where women are concentrated in low-paid or low-profit sectors – is a non-trivial source of the gender wage gap worldwide, accounting for as much as 50% of the gap in some countries (World Bank 2011). We see this phenomenon in both high- and low-income settings. A recent study that used nationally representative data from the US, for example, found that “gender differences in occupations and industries are quantitatively the most important measurable factors explaining the gender wage gap” (Blau and Kahn 2016). Among Sri Lankan entrepreneurs of small businesses, female ownership loses its power to explain gender performance differences once industry is taken into account (de Mel et al. 2009).

There is some evidence that women’s biases about their own potential can affect their performance and aspirations. In the Netherlands, for example, boys tend to choose more academically prestigious tracks than girls with similar levels of academic ability (Buser et al. 2014). A study by Spencer et al. (1999) found that when women are told that men have previously outperformed women, a gender gap in performance in a mathematics test is generated that is not present when women are first told that both sexes performed equally well on the test.

Changing aspirations

How can we change these aspirations? Sometimes what can help is simple exposure to a situation that demonstrates women can be equally as successful as men. After witnessing two election cycles of mandated female leadership in village government, adolescent girls in India were much more likely to express a desire for a job that requires an education, and the gender gap in school attendance disappeared (Beaman et al. 2012). After recruiters from the business process outsourcing industry showed up in rural Indian villages to recruit young women for work in the sector, young women (but not young men) in these villages began to invest more in related training (Jensen 2012), suggesting a shift in aspirations towards white-collar work among a group that previously may not have considered such a profession in their opportunity set.

This is what we think happened in a recent experiment in Nigeria, where my co-authors and I evaluated a job training programme that sought to prepare university graduates for work in the emerging sector of information and communication technology (ICT)-enabled services (Croke et al. 2017), which included business process outsourcing work and work in telecommunications in domestic firms. The training had no explicit gender component. Rather, it was designed to improve skills of both men and women in three competency areas: oral and written communication, basic computing, and cognitive skills. Graduates of the training were meant to take a certification test that had already been approved by firms in the nascent sector.

Even though globally this sector’s workforce appears to be predominantly female, government figures in Nigeria indicate that the sector was male dominated in 2010, when 67% of those employed in the information services sector were men (Holman et al. 2007, National Bureau of Statistics 2010). When we interviewed the target population and training centre directors in 2010 before the experiment, they were worried about women’s prospects in the sector, not just because of standard labour market discrimination but also because of women’s own confidence in seeking work in the formal sector. To explore this empirically, we added implicit-association tests to the application process. The implicit-association test is a computer-based sorting task developed by social psychologists that tries to measure automatic associations between a group (such as men or women) and concepts (such as employment, sectors, or professionalism). Because the differences in sorting times are often less than a second, these associations are considered implicit and automatic, or beyond conscious control (Greenwald et al. 1998, Nosek et al. 2002).

The experiment demonstrated that the training was successful in shifting employment to the ICT-enabled sector, with applicants offered training slots 1.7 percentage points (or 26%) more likely, on average, to work in the sector two years after the training. We found similar effects for male and female applicants, but a markedly different impact among women who exhibited a bias against women’s professionalism before the training. After the training, these women were three times more likely to find an ICT-enabled service job than unbiased women. Because the counterfactual level of employment in the sector was so low for women (2.2%), even unbiased women substantially increased their prospects of employment in the sector (an increase of 2.7 percentage points, or 119%) after they were offered training. Women who entered with a pro-female bias, however, did not benefit from training.

Why might this have happened? Our end line data do not permit an investigation of the mechanisms. We had to use a phone survey because two years after the training, applicants were scattered all over Nigeria. There is, however, some suggestive evidence from the certification test that followed the training that the training experience could have changed how women viewed their own abilities relative to those of men. Total scores were statistically indistinguishable across women and men, and women scored significantly higher on skills that might be more observable to peers (such as voice clarity, grammar, and accent), as opposed to skills that would have been tested via computers (such as keyboard skills, and internet and browsing skills).

The role of training

Recent reviews suggest that skills training programmes have low returns in low- and middle-income countries when it comes to employment generation or improvements in earnings (McKenzie 2017, Blattman and Ralston 2015). Judged by these metrics, the training programme in Nigeria is no exception, as it did not generate any impacts on overall employment or earnings (as measured two years after the training). These metrics alone, however, do not capture all of the potential benefits of training. In the case of Nigeria, training induced switching into a nascent sector and opened up the possibility set for a group that previously may not have imagined themselves as being successful in a work environment that requires professionalism. These two non-pecuniary benefits imply increased mobility across professions and a reduction in occupational segregation, which should ultimately decrease gender wage gaps in the labour market.

Photo credit: Kaizenify/Wikimedia UG Nigeria.


Beaman, L, E Duflo, R Pande and P Topalova (2012), “Female Leadership Raises Aspirations and Educational Attainment for Girls: A Policy Experiment in India”, Science 335(6068): 582-586.

Blattman, C and L Ralston (2015), “Generating employment in poor and fragile states: Evidence from labor market and entrepreneurship programs”, working paper.

Blau, F and L Kahn (2016), “The Gender Wage Gap: Extent, Trends, and Explanations”, NBER Working Paper No. 21913.

Buser, T, M Niederle and H Oosterbeek (2014), “Gender, Competitiveness, and Career Choices”, Quarterly Journal of Economics 129(3): 1409-1447

Croke, K, M Goldstein and A Holla (2017), “Can Job Training Decrease Women’s Self-Defeating Biases? Experimental Evidence from Nigeria”, World Bank Policy Research Working Paper No. 8141.

de Mel, S, D McKenzie and C Woodruff (2009), “Are Women More Credit Constrained? Experimental Evidence on Gender and Microenterprise Returns”, American Economic Journal: Applied Economics 1(3): 1-32

Greenwald, A G, D E McGhee and J L K Schwartz (1998), “Measuring Individual Differences in Implicit Cognition: The Implicit Association Test”, Journal of Personality and Social Psychology 74(6): 1464-1480.

Holman, D, R Batt and U Holtgrewe (2007), “Executive Summary”, in The Global Call Center Report: International Perspectives on Management and Employment, Ithaca, NY: Authors.

Jensen, R (2012), “Do Labor Market Opportunities Affect Young Women’s Work and Family Decisions? Experimental Evidence from India”, Quarterly Journal of Economics 127: 753-792

McKenzie, D (2017), “How Effective are Active Labor Market Policies in Developing Countries”, World Bank Policy Research Working Paper No. 8011.

National Bureau of Statistics (2010), National Manpower Stock and Employment Generation Survey, Nigeria.

Nosek, B A, M Banaji and A G Greenwald (2002), “Harvesting Implicit Group Attitudes and Beliefs from a Demonstration Web Site”, Group Dynamics: Theory, Research, and Practice 6(1): 101-115.

Spencer, S J, C M Steele, and D M Quinn (1999), “Stereotype Threat and Women’s Math Performance”, Journal of Experimental Social Psychology 35: 4-28.

World Bank (2011), World Development Report 2012: Gender Equality and Development, Washington, DC.

Job training Nigeria Gender