Progress towards gender parity in education is undeniable

2020 Gender Report

Download PDF

Credit: Plan International / Adolescent girls from the Solomon Islands

In 1995, the education world was a very different place from a gender perspective. The glaring discrimination to which millions of girls and women were subjected in many parts of the world had repercussions that are still with us today: directly, in the guise of the stubbornly high share of women in the total population of illiterate adults, and indirectly, with women counting the consequences of having had no say in matters related to fertility or participation in economic, social and political activities.

Female enrolment accounted for 55% of the total increase in primary and secondary between 1995 and 2018, growing by 180 million


A generation later, the daughters of those young women enjoy the benefits of major progress towards parity in a range of education indicators, especially in primary and secondary education. Still, new potential areas of inequality are emerging in access to digital learning opportunities. These need to be monitored closely and addressed in coming years. This section reviews the latest status of gender disaggregated data and some of their main trends in education. A short presentation also discusses whether and how some education-related indicators could be affected by Covid-19.



Globally, female enrolment accounted for 55% of the total increase in primary and secondary enrolment between 1995 and 2018, growing by 180 million, from 469 million to 649 million. Central and Southern Asia accounted for 47% of the total increase and sub-Saharan Africa for 38%. The global increase in women’s enrolment corresponds to a combined gross enrolment ratio in primary and secondary education from 73% to 89%. The world has thus moved towards gender parity (Figure 1). Globally, the gender parity index in primary and secondary education increased from about 90 girls enrolled for every 100 boys in 1995 to an equal number of both in 2018. At both levels, the global trends were led by Southern Asia, particularly India, where more girls than boys are now enrolled in primary and secondary education. Significant improvement has also been observed in other countries. For instance, in 24 countries where in 1995 there were fewer than 80 girls enrolled in primary education for every 100 boys, including Bhutan, Burkina Faso, Djibouti and Nepal, parity has been achieved.

Girls continue to be more likely than boys to have never enrolled in school at all. Globally, of the 59 million primary school-age children who were not enrolled in 2018, 12 million, or 20%, have never attended school and will probably never start if current trends continue. Girls made up three-quarters (9 million) of such children, and over 4 million of those girls were in sub-Saharan Africa (UIS, 2019).

Between 1995 and 2018, the percentage of countries that had achieved gender parity rose from 56% to 65% in primary, from 45% to 51% in lower secondary and from 13% to 24% in upper secondary education (Figure 2).

Still, considerable disparity at girls’ expense remains, mostly in low-income sub-Saharan African countries. In 2018, fewer than 90 girls were enrolled for every 100 boys in 7 countries in primary, 15 countries in lower secondary and 22 countries in upper secondary education. Fewer than 90 boys were enrolled for every 100 girls in 24% of the 152 countries with data for upper secondary education in 2018, a situation that has hardly changed since 1995.

The intersection of gender with other factors of disadvantage compounds disparity. In a group of low- and middle-income countries with gender disparity in primary school attendance, the parity index was even lower among the poorest and those living in rural areas, mostly at the expense of girls (Figure 3). In Cameroon, Guinea, Pakistan and Yemen, gender parity existed among the richest 20% of households and those living in urban areas, but high levels of gender disparity were apparent among the poorest 20% and those living in rural areas. In Pakistan, only 70 of the poorest girls attended primary school in 2018 for every 100 of the poorest boys. In a few countries, including Lesotho, Sierra Leone and the United Republic of Tanzania, disparity at boys’ expense at national level is further exacerbated among the poorest and those in rural areas.

Among 56 countries with data for 2000–18, primary completion rates improved faster, on average, for girls (by 17 percentage points) than for boys (by 15 percentage points). In Burundi, Cambodia and Sierra Leone, completion rates for girls rose by more than 40 percentage points. While overlapping vulnerabilities persist over time, completion rates for disadvantaged girls have been improving relatively in many countries. In Guatemala, the primary completion rates of the poorest girls increased from 21% to 54% between 2000 and 2015, faster than the rate of the poorest boys (from 35% to 58%), which meant the parity index increased from 0.60 to 0.92 (Figure 4).


In Pakistan, only 70 of the poorest girls attended primary school in 2018 for every 100 of the poorest boys

In one-third of the 86 countries with household survey data for 2013–18, girls were more likely to complete primary school than boys. But significant disparity at girls’ expense still exists in many countries. Fewer than 80 girls for every 100 boys completed primary in 4 countries (Afghanistan, Chad, Guinea-Bissau and Yemen), lower secondary in 15 countries and upper secondary in 22 countries.

At the upper secondary level, male completion rates were more than twice the level of female completion rates in Afghanistan, Benin, Chad and Togo. In Chad, 14.6% of boys and 5.5% of girls completed upper secondary school. In at least 20 countries with data, mostly in sub-Saharan Africa but also in Belize, Haiti, Pakistan and Papua New Guinea, hardly any poor, rural young women completed upper secondary school (Figure 5)



Globally, girls continue to outperform boys in reading, according to the 2018 Programme for International Student Assessment (PISA) of the Organisation for Economic Co-operation and Development (OECD). Among 15-year-old students, 80 boys achieved the minimum proficiency level in reading for every 100 girls across the 80 participating countries and territories. Girls’ advantage in reading was very large in, for example, North Macedonia and Thailand, where about 60 boys achieved the minimum proficiency level in reading for every 100 girls. In more than half the 38 countries and territories that took part in PISA in 2000 and 2018, girls extended their advantage in reading over boys in 2018. In Israel, the number of boys achieving the minimum proficiency level in reading fell from 91 to 71 for every 100 girls.

Girls and boys perform equally well in mathematics in more than half of countries. Girls do better in one-quarter of countries, especially in South-eastern Asian countries such as Indonesia, the Philippines and Thailand. Boys maintain a strong advantage in Latin American countries such as Colombia and Costa Rica (Figure 6). Boys (12.3%) are more likely than girls (9.5%) to be among the highest performing students (levels 5 and 6 in PISA) across OECD countries (OECD, 2019).

Gender disparity in learning outcomes intersects with other forms of disadvantage. In 24 countries, over 70% of poor boys scored below the minimum level of proficiency in reading in 2018, and in 24 countries, over 70% of poor boys scored below the minimum level (Schleicher, 2019). Detailed administrative data linked with school records have shown how boys from disadvantaged families in Denmark and in Florida, United States, had lower achievement scores and were less likely to complete secondary school than girls from similar backgrounds (Autor et al., 2019;Brenøe and Lundberg, 2018).

An important caveat is that the increasing gap in education attainment and reading achievement between girls and boys in some contexts does not necessarily seem to lead to disadvantage in adult skills for men. A study in the United States found that boys had more problems at school than girls, including lower learning aspirations and more frequent suspensions, when they were in households without a father, in poor quality schools and in less educated neighbourhoods. However, no evidence was found that this disadvantage had a greater impact on tertiary education, employment and income for men compared to women (Lei and Lundberg, 2020).


In 24 countries, over 70% of poor boys scored below the minimum level of proficiency in reading in 2018



Education opportunities for women have expanded even more strikingly at tertiary education level. Globally, women’s enrolment in tertiary education tripled, from 38 million to 116 million, between 1995 and 2018, accounting for 54% of the total increase in enrolment. The female gross enrolment ratio increased from 15% to 41% between 1995 and 2018, with the adjusted gender parity index rising from 0.95 to 1.14. Disparity at the expense of men was observed in all regions except Central and Southern Asia, where there is parity, and sub-Saharan Africa, where disparity at the expense of women persists, with 73 female students enrolled for every 100 males in 2018 (Figure 1). At the country level, gender disparity at men’s expense is recorded in 74% of the countries with data (Figure 2).


The 1995 Beijing Declaration and Platform for Action made improvement of women’s access to vocational training and science and technology a strategic objective


Northern Africa and Western Asia has seen rapid expansion of tertiary education participation, outpacing the global average in recent years. Yet country experiences vary. Tunisia had among the highest participation rates as recently as 2010 but has since stagnated at around 35%. Saudi Arabia’s enrolment rate more than doubled between 2009 and 2017, from 32% to 70%. In some countries in the region, such as Algeria, women have been the main beneficiaries of rapid increases in tertiary education enrolment. By contrast, Saudi Arabia, with some of the highest gender disparity levels (albeit at men’s expense), increased enrolment while achieving gender parity. Morocco, which in the early 1990s had one of the most gender-unequal tertiary enrolment ratios (30 women for every 100 men), reached parity in 2017. As recently as 2011, Morocco had the same low participation rate as Sudan (16%), but while the latter stagnated, Morocco more than doubled participation in seven years to 36% (Figure 7).

The 1995 Beijing Declaration and Platform for Action made improvement of women’s access to vocational training (Box 3) and science and technology a strategic objective. But while global expansion of tertiary education has favoured women, their subject choices are still heavily influenced by perceptions of their abilities to pursue certain types of studies. Girls and women face barriers that restrict their engagement in male-dominated fields (CEDAW, 2017). The perception of science, technology and mathematics as ‘male’ subjects turns gender differences into self-concepts. At all education levels, girls show lower values in self-efficacy – that is, perceived as opposed to actual abilities – in mathematics and science subjects, aside from life sciences.

Women are over-represented in education, health, arts, humanities and social sciences, and under-represented in some science, technology, engineering and mathematics (STEM) fields of study. It varies from less than 1% in Maldives to 41% in Oman. In OECD countries, just 20% of new entrants to short-cycle tertiary programmes and 30% of new entrants to bachelor’s programmes in STEM fields in 2017 were women (OECD, 2020).

Engineering, manufacturing, construction, and information and communication technology (ICT) programmes are highly male-dominated: globally, the percentage of females studying engineering, manufacturing and construction or ICT is below 25% in over two-thirds of countries. In OECD countries, on average, women account for less than 20% of entrants in tertiary computer science programmes and about 18% of engineering entrants (OECD, 2017). The share of female students in ICT programmes is around 10% to 12% in high-income countries, including Belgium, the Netherlands, Spain and Switzerland, compared with 58% in Myanmar and 51% in Tunisia.


Women account for less than 1% of the Silicon Valley applicant pool for technical jobs in artificial intelligence and data science


Results from the 2018 International Computer and Information Literacy Study of grade 8 students in 21 mostly high-income countries showed that although girls performed better than boys on measures of digital skills, they were less likely to want to study or find a job in an ICT-related field (Faulstich-Wieland, 2020).

Girls are less likely than boys to study in STEM fields

Gender segregation in fields of study affects career prospects and equality in work opportunities. In the 37 OECD countries participating in the 2018 round of PISA, 7% of girls but 15% of boys expected to work in science and engineering professions. The gender gap was particularly high in Colombia (10% of girls, 25% of boys) and Mexico (11% of girls, 30% of boys). The PISA results also showed that less than 1% of girls in OECD countries, but nearly 8% of boys, wanted to work in ICT-related occupations, with especially wide differences reported in Bulgaria, Estonia and Lithuania and in Poland, where 1% of girls but 19% of boys expected to work in ICT professions (OECD, 2019b).

These expectations were unrelated to performance. PISA results showed that in OECD countries, only 14% of girls who were top performers in science or mathematics expected to enter a professional field in science and engineering, compared with 26% of top-performing boys (Encinas-Martin, 2020). The gender gap among top performers persists even in countries that score highly on common gender equality measures, such as Norway (12% of girls, 33% of boys) and Sweden (20% of girls, 37% of boys).

Women account for less than 1% of the Silicon Valley applicant pool for technical jobs in artificial intelligence and data science. They are largely absent from the frontiers of technological innovation, where job growth is expected and pay typically is highest. Globally, only 6% of mobile application and software developers are female. At Google, women fill 21% of technical roles but account for only 10% of employees working on machine intelligence (UNESCO, 2019b).

Lack of gender diversity can have a serious multiplier effect as big data and algorithms become influential in day-to-day life (Rosenberg, 2017). Technology reflects its developers’ values, and there is an urgent need to avoid gender bias in deep learning systems. More diversity on artificial intelligence technology development teams could help prevent bias and empower women to become digital creators (UNESCO, 2019b; European Union, 2016).

Gender gaps in career expectations are related to deeply ingrained gender-stereotyped norms about which careers are suitable for men and women. They are passed on to children by families, teachers and wider societies (OECD, 2017). According to 2015 PISA data, parents are more likely to expect adolescent sons to work in STEM occupations, even when daughters perform as well as their male classmates in mathematics, science and reading (OECD, 2015, 2017).




Adult participation rates in formal and non-formal education and training in the previous 12 months were below 10% in more than half of the 99 countries with data available for 2011–18, according to the UNESCO Institute for Statistics database. Percentages ranged from less than 1% in 13 countries, mostly in Asia and sub-Saharan Africa, to over 60% in the Netherlands, New Zealand, Sweden and Switzerland. Women’s participation rates exceed those of men in Baltic states (e.g. by 14 percentage points in Estonia, 9 in Latvia) and Scandinavian countries (e.g. by 12 percentage points in Finland, 9 in Sweden) (Eurostat, 2019). One potential reason is gender segregation in education and employment. Too few males in these countries attend health, education and welfare courses: 9% in Estonia and 16% in Finland, well below the European Union average of 23% (European Institute for Gender Equality, 2019). In addition, these countries have higher than average labour market gender segregation by occupation and/or by sector (Burchell et al., 2014). Women are more likely to work as nurse and healthcare assistants and/or in the public sector, where opportunities for training are higher.

Analysing gender barriers to adult education and learning requires a clear methodological framework. The way barriers are categorized matters. One long-standing categorization (Cross, 1981) describes factors preventing participation as situational (e.g. life circumstances such as family responsibilities and lack of time), dispositional (e.g. determined by previous learning experiences and personal disposition towards learning) and institutional (e.g. structural conditions hampering access, such as cost, lack of support, rigid schedules and limited provision) (UIL, 2019). Dispositional barriers are less often investigated in surveys and thus are frequently underestimated (Rubenson, 2011). But when they are measured, they are shown in most countries to be the strongest factor hindering adult learning.

On average, across EU countries, almost 60% of adults surveyed said they did not participate in adult learning mainly because they saw no need for it. Cost and inconvenient schedules or locations were the most pressing institutional barriers. Among those willing to participate who did not, lack of time and family responsibilities were the most common situational barriers, according to data from the Programme for the International Assessment of Adult Competencies and the Adult Education Survey. While men were slightly more likely to mention scheduling as a barrier, women in all countries except Denmark were far more likely to mention family responsibilities. The tendency was higher in southern Europe, with up to two-thirds of female respondents in some countries unable to participate because of family commitments (Figure 8).

A comparative study based on 14 time-use surveys and 5 household surveys in 19 countries found that men allocated slightly more time to learning, leisure and social activities. Albania, Ghana, Pakistan and the Republic of Moldova reported the highest gender imbalances: Ghanaian women, for example, spent almost two hours less than men per day on these activities (Rubiano-Matulevich and Viollaz, 2019). Women were more likely to see cost as an obstacle and less likely to have scheduling conflicts, probably reflecting their lower labour force participation and higher part-time employment rates.
In sub-Saharan Africa, only 81 adult women were literate in 2018 for every 100 literate men



Despite the expansion of education across the globe, progress in literacy skills acquisition has been relatively slow, largely because much of the progress has been offset by rapid population growth in countries with low literacy rates. Globally, the number of illiterate adults fell by nearly 16%, from 916 million in 1995 to 773 million in 2018. The adult literacy rate increased from 76.5% to 86.3%. In Eastern and South-eastern Asia, the number of illiterate adults fell by two-thirds, from 228 million to 78 million. However, in sub-Saharan Africa the number of adult illiterates increased by 46%, from 139 million to 204 million, despite the literacy rate rising from 55% to 66%.

Gender disparity in adult literacy still prevails worldwide. Globally, the adult literacy gender parity index increased from 0.84 in 1995 to 0.92 in 2018. But while parity has been achieved in Latin America and the Caribbean and is about to be achieved in Eastern and South-eastern Asia, disparity persists in Central and Southern Asia and in sub-Saharan Africa, where 81 adult women were literate in 2018 for every 100 literate men.The share of females among illiterate youth has decreased globally since around 2005, although in Eastern and South-eastern Asia, a sharp decrease has been evident since 1990 (Figure 9a).

The share of women among illiterate adults, however, has remained constant for the past 20 years at around 63%. Eastern and South-eastern Asia has the highest adult female literacy rate and may be moving towards parity, but it also has the highest share of women among illiterate adults. Among women over age 65 in the region, 75% are illiterate (Figure 9b).

In 2018, fewer than 80 women were literate for every 100 men in 12 countries, mostly in sub-Saharan Africa. In Afghanistan, Benin, the Central African Republic, Chad and Mali, female literacy rates were around or below 30% and the gender parity index was below 0.60. In Chad, the women’s literacy rate in 2016 was as low as 14%, compared with 31% for men. The extremely low rate was confirmed by direct literacy assessments: 15% of women aged 15 to 49 had attended secondary school and so could be assumed to be able to read a simple sentence without difficulty. Of those who had not attended secondary school, 1.7% could read a simple sentence without difficulty and 5.5% could read part of the sentence. In Guinea, 4% of women aged 15 to 49 had attended tertiary education. Of those who had not, 11% read without difficulty and 9% could read part of a sentence. In both countries, one in six women could read a sentence without difficulty. In 12 of 20 regions in Chad, only 1 in 10 females could read, falling to 1 in 100 in the Lac and Wadi Fira regions (Figure 10)

The most disadvantaged women are further left behind in terms of literacy skills. In 59 countries, women aged 15 to 49 from the poorest households are four times more likely to be illiterate than those from the richest households (UN Women, 2020a). In Nepal, while the literacy rate of the richest women at 75% was more than twice that of the poorest women at 30%, it remained substantially lower than that of the richest men at 93% (UNESCO, 2019). Women with disabilities tend to be more disadvantaged. The widest gap was seen in Mozambique, where 49% of men with disabilities, but only 17% of women with disabilities, could read and write (UIS, 2018).


Even when women have access to the internet, they may be less able to use it for various reasons related to gender disparity


A number of countries have attempted to improve women’s access to adult literacy programmes as well as to expand the content of these programmes to make them more relevant. In Cambodia, a programme for women has expanded its scope from basic literacy to functional and information literacy, including financial literacy. In Eritrea, women and girls have been the primary focus of new literacy programmes offered through learning centres, resulting from collaboration with communities in remote areas. Morocco has prioritized women’s literacy programmes focused on developing socio-economic skills. Saudi Arabia has established equal access to high-quality education and has improved literacy rates among women. Systems have been shifting from a focus on eradicating illiteracy to increased concentration on continuing education (UNESCO, 2019).




Besides basic literacy and numeracy, ICT skills are increasingly seen as essential to function in and navigate a knowledge-based world. Access to and use of ICT tools have become indispensable in day-to-day life and, even more so, in emergency contexts such as the Covid-19 crisis (Box 4). Wide gender disparity exists in ICT skills distribution. Worldwide, 327 million fewer women than men have a smartphone. Even when women have access to the internet, they may be less able to use it for various reasons related to gender disparity. For example, when multiple members of a household need access to limited computing resources at home, women and girls may receive less access (UNESCO, 2020b).

The sixth round of the Multiple Indicator Cluster Surveys (MICS) includes a module with questions on the nine ICT skills being monitored as part of SDG global indicator 4.4.1. The data cover adults aged 15 to 49 and allow disaggregation of skills by individual characteristics. Clear gender patterns emerge across the 10 countries that have included the module in their survey. For instance, women in the seven poorest countries covered are less likely to have used a basic arithmetic formula in a spreadsheet, while parity exists in the three richer countries: Mongolia, Suriname and Tunisia. There are two additional features of interest in these three countries. First, young women are slightly more likely than men to have this skill. Second, there is a distinctive age profile, showing the rapid pace of ICT adoption by younger people (Figure 11).

The data also show wide socio-economic disparity in the distribution of basic ICT skills. In the seven poorer countries, the probability of women in the poorest 60% of the population having the spreadsheet skill is below 1%. In the three richer countries, 3% of women from the poorest quintile had this skill as compared with 27% in the richest in Tunisia, 35% in Suriname and 39% in Mongolia.