Our Programmes



Sign up to our newsletter.

Follow ODI

Progress on gender inequality: how data can help

Written by Nicola Jones


Globally, more women have access to paid work now than in 2000. Recently highlighted by Phumzile Mlambo-Ngcuka, Executive Director of UN Women, it’s one of the many reasons to celebrate gender-related Millennium Development Goals (MDG) progress in the lead up to International Women’s Day (8 March).

However, there is still work to be done.

Yes, more girls are in school, fewer women and girls are dying in pregnancy and childbirth, and more women are leaders in their communities and countries. But despite growing recognition of the close links between better gender equality and better development outcomes, progress on gender-based disadvantage has been slow and uneven.

MDGs 3 (on gender equality) and 5 (maternal mortality) will not be reached by 2015 in any region, with the exceptions of parts of East Asia and Latin America and the Caribbean.

Misleading national averages

Just as importantly, the MDG focus on national averages has masked stark differences between social groups. In her excellent report on the challenge of intersecting inequalities, Naila Kabeer argues that if the MDGs were to promote a pathway to social justice, far more attention should have been paid to the divergent patterns of poverty within and across countries.

For example, data on extreme poverty in Latin America show that poverty rates for Afro- and indigenous populations are many times higher than for the majority population:  twice as high in Brazil, three times higher in Mexico and a whopping eight times higher in Paraguay. In India 2006 Government data found that while the national average poverty rate was 29%, it was 74% for Adivasi and scheduled tribes in Orissa state. 

When gender and age are factored in, the level of disadvantage and exclusion is often compounded. In 2010, UNESCO found that although education levels among indigenous populations in Latin America were consistently below those of majority populations, levels for poor indigenous women were often far worse, at less than half the indigenous average. In the case of maternal mortality, it is adolescent girls, especially those under 15, who suffer disproportionately.

As UNFPA’s 2013 Motherhood in Childhood report highlighted, girls aged 15 or younger who get pregnant are twice as likely to die as older teens and women. Similarly, WHO has underscored that being young  is a known risk factor in terms of violence, including sexual violence, at the hands of an intimate partner. While global statistics are woefully lacking on sexual and gender-based violence, recent data for Liberia paint a stark picture: court records from 2013 show that 94.5% of all rape cases affected children under the age of 18 years.

‘Data revolution’

There are growing calls for a ‘data revolution’ from development analysts and practitioners alike to shine a light on such discrimination. If this revolution is to foster social justice for the most marginalised new research by ODI on adolescent girls' wellbeing among Viet Nam’s highly marginalised Hmong ethnic minority suggests that three things need to happen.   

First, incentives to collect and disaggregate data need careful thought at every level, from the national to the community, given that evidence of disadvantage is politically sensitive. Where honest reporting risks embarrassing tensions with legal norms (e.g. reporting on rates of child marriage in contexts where it is illegal), officials’ performance needs to be linked to progress against a baseline rather than absolute rates if evidence-informed policy and programming is to become a reality.

Second, levels of data disaggregation need to be razor-sharp to allow the adequate tailoring of programmes. In Viet Nam, for example, poverty and social development indicators have been disaggregated by Kinh (ethnic majority) and ‘ethnic minority’ for some time. But recent research has found that ethnic minority groups may, in fact, have very varied development trajectories, with some faring quite well (the Khmer, for example) and others lagging far behind (the Hmong).

Finally,we must not overlook the value of qualitative and participatory evidence. In the case of adolescent girls, for example, we need to learn about their own experiences of poverty and well-being by listening to their voices.

Our recent qualitative and participatory research with 13-19 year-old Hmong girls in Viet Nam found that they saw income poverty and access to education as important, but they also wanted more chances to speak in confidence to their peers and to knowledgeable adults about sexual and reproductive health, intra-familial violence and their future career options. 

In short, and to borrow the recent title of a World Bank report, the picture is this: ‘Well Begun but Not Yet Done’. On International Women’s Day we can be heartened by the MDG-supported progress made towards reducing gender inequalities – but also strive for post-2015 world that values true and lasting equality for all women and girls.