Development Initiatives (DI) works with partners and allies to strengthen the quality and use of data and evidence that can inform decision-making to tackle poverty and inequality in all its forms. We have conducted several independent projects focusing on gender data and this briefing presents key challenges we have encountered in that process alongside opportunities we have identified to improve the gender data landscape. Strengthening this landscape is an intentional focus of DI’s work going forwards, supporting the use of robust evidence to understand and reduce gender inequalities.

---

► Read about recent global trends in inequality

► Find out more about DI's current work on gender

► Sign up to our newsletter

---

All over the world people are treated differently, discriminated against and excluded because of their gender. This is evident in the workplace, the home, in politics and in society at large. However, this need not be the case . The aim to achieve gender equality and women’s empowerment runs throughout the UN’s Agenda 2030; it is a universal value alongside the commitment to leave no one behind. These aims are integral to achieving all the Sustainable Development Goals (SDGs) through pursuing a transformative approach to development. More specifically, the fifth SDG also sets out to ‘achieve gender equality and empower all women and girls’ with concrete targets and indicators for every country.

Data and evidence that provide insight into differences by gender and the associated intersecting drivers of exclusion can be used to ensure that policies and interventions are appropriately targeted to reduce poverty and inequality in a society.

Existing gender data sources

Many surveys, tools and indices seek to measure gender inequality, highlighting where and how exclusion and discrimination is gendered. At the international level, the World Bank’s gender data portal has data for more than 900 indicators. The SDG monitoring framework identifies 14 indicators for SDG 5, as well as indicators for 13 other goals that have explicit gender equity targets, making 51 gender indicators in total, which are officially monitored by the UN . There are several publicly accessible composite indices that also raise awareness of gender-based differences in people’s outcomes, as well as providing evidence on potential causal drivers of gender inequality that should be addressed. Independent tools such as the SDG Gender Index use data from official and non-official sources to look beyond SDG 5 to include the gendered dimensions of all SDGs, covering 98% of the world’s girls and women. The UN Development Programme (UNDP) publish the Gender Inequality Index (GII), which measures reproductive health, empowerment and the labour market, and the Gender Development Index (GDI), which measures the disparities in development outcomes between men and women. In 2023, the GDI will be replaced with two new indices. The Global Gender Gap report , published by the World Economic Forum, measures gender disparities across four dimensions: Economic Participation and Opportunity, Educational Attainment, Health and Survival, and Political Empowerment. Each of these provide a range of underlying data points and have their own uses and limitations .

Gender data also includes the collection of information on issues where gender differences are explicitly recognised. These issues include GBV , female gentital mutilation (FGM) and male gang violence . Important methodological progress has been made with respect to collecting inclusive, ethical and safe data on these sensitive topics, particularly when it comes to protecting the interests of the most vulnerable people.

Missing gender data is a barrier to effectively tackling gender inequality

Across all dimensions of poverty and inequality, people’s experiences and priorities may differ depending on their gender. This necessitates taking a gendered approach to data that intentionally listens to people’s different perspectives and ensures that nuanced information is captured on the issues that matter to people of all gender identities. But we are far from having the visibility we need on what gender inequalities look like locally, nationally and globally across the diverse aspects of our lives. For example, less than half the data needed to measure progress against SDG 5 is available. Women, children and marginalised groups face barriers to inclusion in even the most fundamental data recognising people’s identities, as collected through Civil Registration and Vital Statistics (CRVS). This missing data can negatively impact countless outcomes for rights, livelihoods and wellbeing, including access to basic services.

Without timely and robust gender data that is effectively used and leveraged , it is challenging to make well-informed decisions about how to most effectively reduce inequalities and improve outcomes for people left behind. The rest of this briefing explores four different gender data gaps, outlining associated opportunities where DI can support partners and allies to fill those gaps. While these gaps are presented as separate issues, there are clear overlaps and dependencies between them. The final section highlights the need to understand and address these data gaps in the context of wider imperatives for inclusion and the protection of rights in local, national and global data ecosystems.

Challenge: The most prominent data on poverty and inequality is measured at the household level, and so cannot identify differences in outcomes between people of different genders within the same household.

Every year, DI publishes an economic poverty factsheet that draws on internationally comparable World Bank data on global incomes and prices to estimate levels and trends in people living below the extreme poverty line ($2.15 per person per day). It provides headline findings on who is furthest behind from an economic perspective. The underlying data comes from large-scale nationally representative surveys, where questions on income and consumption are answered on behalf of everyone in that household. Separate questionnaires for women and men focus mainly on health and reproductive issues. Similarly, household-level data from the international Demographic and Health Survey (DHS) and Multiple Indicator Cluster Survey (MICS) are used to calculate the OPHI/UN Multidimensional Poverty Index (MPI). Assuming that everyone in the same household has the same outcomes does not acknowledge the reality that there may be within-household inequalities , including when it comes to the control, availability and use of financial resources. Strong evidence indicates that people within some households have different experiences of multidimensional poverty based on their gender. For example, in some countries in Africa, the Middle East and South Asia, boys are more likely to go to school than girls, and are more likely to eat first and eat more in times of food scarcity . But household-level surveys do not capture the vital data that reveals such gender-related inequalities within households.

Opportunity: Strengthen a diverse national data ecosystem that goes beyond household measures.

A range of alternative data can be used to complement traditional household survey data. Census data, administrative data, targeted surveys, registries, and non-traditional data such as mobile phone records and community-generated data, are typically collected at the individual level and can be effectively disaggregated. These can be powerful complements to household surveys. For example:

• The Gender Data Navigator has been created by the International Household Survey Network (IHSN) to identify useful surveys that allow for a gendered analysis on a country-by-country basis.
• The World Bank’s Strengthening Gender Statistics Project is supporting 12 different national statistical offices (NSOs) to assess the availability and quality of sex-disaggregated statistics and then work to close these data gaps through the data collection, analysis and dissemination phases of the data life-cycle. This effort is much smaller in scale than their household-level data collection projects.
• UN Women’s Women Count programme has been working with NSOs to: ‘Analyse Gender Data from a LNOB Perspective’; produce disaggregated statistics for a prioritised list of gender indicators; and develop methodologies and produce disaggregated statistics for selected National Priority Gender Equality Indicators (NPGEI).
• Mobile phone data, specifically call data records, have been used in Pakistan to identify gender inequalities in social networks.
• Tax data can reveal gender disparities in taxpayers at the very top of the income distribution. In nine high-income countries studied, at least three out of four of the top 1% of income earners were men; in Norway, it was almost nine in ten.
• Surveys deliberately designed to capture information at the individual level include the data collected to inform Equality Insights , which are specifically designed to enable a gendered analysis of poverty.
• Within the household survey community, efforts are also being made to address this challenge. The LSMS+ project is focusing on collecting data that can be disaggregated for asset ownership and living standards with guidance from the UNSD on how to produce these statistics.

Challenge: Analysis of outcomes by gender alone can miss the extent to which gender discrimination or exclusion potentially compound or exacerbate other inequalities.

Gender parity has been achieved in Nigeria when it comes to primary school enrollment , whereby just as many girls as boys are enrolled into school. However, this statistic simplifies the diverse experiences of different boys and girls across the country. In particular, significant gender gaps in education exist in rural and marginalised communities , particularly in northern Nigeria. Global analysis of poverty that seeks to disaggregate data by both sex and age reveals that gender gaps are widest for people aged between 25 and 34, particularly in households with dependent children. This is a critical insight that can inform policies with respect to addressing the gender norms and support associated with childcare. Similarly, in the political sphere, women now make up at least half of all parliamentarians in several countries, while others have female heads of state. However, this does not mean that all women across all decision-making structures have equal opportunities to rise to positions of power and influence. Rather, an individual’s socioeconomic status interacts with their gender to affect their outcomes and opportunities. As a consequence, all inequalities should be counted, understood and challenged in their own right, as well as being part of an intersectional analysis of who is furthest behind due to these compounding disadvantages. However, this analysis is limited when the underlying data is not fully inclusive, or where there is a risk of identifying vulnerable individuals with multiple characteristics. For example, in DHS and MICS surveys, only women of reproductive age are counted, meaning that women over the age of 50 are excluded , while data with small sample sizes has limited power for intersectional analysis and can risk inadvertently disclosing information about an individual, even when it has theoretically been anonymised.

Opportunity: Growing recognition of ‘intersectional feminism’ and the need to understand concurrent and overlapping forms of discrimination through more nuanced data analysis.

• The need for disaggregation is not limited to the gender variable and an intersectional analysis requires understanding how multiple identities compound to affect a person’s outcomes, requiring data to be captured for other critical variables too. Indeed, as stated by the UN Statistical Commission, “Sustainable Development Goal indicators should be disaggregated, where relevant, by income, sex, age, race, ethnicity, migratory status, disability and geographic location, or other characteristics, in accordance with the Fundamental Principles of Official Statistics .”
• The Global Partnership for Sustainable Development Data (GPSDD) has produced a package of resources to support organisations seeking to take an intersectional approach to data.
• Colombia has demonstrated its commitment to an intersectional approach in official statistics. This is elaborated in their guide on the inclusion of the differentiated and intersectional approach in the statistical production of the Colombian National Statistical System.
• Targeted data collection from specific groups and individuals with intersectional identities can provide important insights into people’s complex identities and characteristics. Qualitative insights , in particular, and community-generated data can effectively tell human stories, sharing the voices and experiences of marginalised individuals.

Challenge: Non-binary and transgender individuals are often invisible in data that uses binary categorisations of gender.

It has been estimated that between 0.1–2% of people identify as transgender or do not conform to cisgender classifications. These groups tend to face exclusion and discrimination as a direct result of their gender status in all aspects of their lives, including education , work and as victims of hate crimes . However, evidence of the size and experiences of these groups is limited because so many data-collection instruments use exclusively binary man/woman gender classifications, which are sometimes used interchangeably with male/female sex categorisations. For official administrative documents such as birth certificates and passports, there are only 16 countries in the world where non-binary classifications are included. This is despite the fact that the UN recognises self-determined gender as a cornerstone of a person's identity and, in particular, provides recommendations for how to establish legal recognition that takes into account gender-diverse persons. Even when a non-binary category is offered in an instrument, people may feel reluctant to openly share their gender status where they feel there would be a risk or stigma associated with that identity, including from the enumerator collecting the data. Because this group is a minority, with necessarily smaller sample sizes, the risks associated with an individual being personally identified in any dataset are also greater, while at the same time the ability to say something statistically meaningful about non-binary people is limited by the smaller sample sizes. Measuring and understanding the structural inequalities faced by these individuals, based on their identity, is therefore particularly important to target intervention appropriately to leave no one behind, but it can be highly risky to those individuals. It is important that frameworks for data ethics and governance are fit for purpose for data that reflects the identities and experiences of non-binary and transgender people.

Opportunity: Precedents have recently been set for the measurement and recognition of people who identify as non-binary in official data-collection tools and processes.

• In countries such as the UK , Canada and Argentina , census questionnaires now include options that allow for the identification of transgender individuals, making a clear and explicit distinction between sex at birth and self-identified gender. Guidance and examples from a range of countries demonstrating how to do this are available in a UNECE report.
• In Nepal , the approach to collecting data on non-binary people evolved to respond to stigma and fear that may have limited people’s willingness to honestly self-report their identity. The language used in data-collection tools was revised and enumerators received training in collaboration with the LGBTIQ+ community.
• In 2022, Colombia implemented the Voluntary Register for the Visibility of Sexual and Gender Diversity to create a database to gather and record information on the lives of the LGBTIQ+ community. This data source has been designed to inform public policies and programmes that target the LGBTIQ+ population, as well as to create statistics that contribute to the recognition of sexual and gender diversity in official statistical data.

Challenge: Outcomes and issues that are more relevant to women and girls are less prominent in headline indicators of prosperity and progress.

It has long been acknowledged that GDP is a limited measure of societal progress , yet it continues to dominate headlines and politicians continue to orientate policies towards achieving economic growth as measured by an increase in GDP. One of the limitations of this approach is the masking of inequalities within a population and, from a gender perspective, the time spent by men and women is valued differently. In particular, men tend to dominate in higher-paid formal employment. Their income and output are included in GDP estimates, and policies to foster growth seek to enhance the productivity of this time spent in work. Conversely, women tend to spend more hours working in the informal or unpaid care economy. This is the case in high-income countries such as the UK where women spend almost twice the amount of time on unpaid work compared with men, work that is estimated to be worth 56% of GDP , but it is particularly pronounced in low-income countries where women must compensate for the lack of care services, or access to time-saving technologies. In many countries, the lifestyle changes associated with the Covid-19 pandemic exacerbated these gender inequalities. Time spent on care and domestic work is not valued as productive and instrumental to growth yet it is critical to support the foundations of society and families; it limits the amount of time women have to spend on other work, education, rest and leisure; and ultimately it contributes to gendered poverty . The need to recognise, reduce and redistribute unpaid care and domestic work is explicitly identified as target 5.4 under SDG 5 to achieve gender equality. However, data for the indicator associated with this target is missing for 151 countries.

Opportunity: Implement emerging standards for the collection of time-use data to systematically measure, understand and address gender inequalities in day-to-day life.

• There are now standards for the international classification of activities for time-use statistics, produced by the UNSD and the EU which, when applied consistently, can allow for international comparisons, particularly in response to policy concerns about gender equality and unpaid forms of work. A more flexible instrument has also been designed for Latin America and the Caribbean that can respond to the classification requirements and socioeconomic characteristics of each country, while still facilitating harmonisation of statistics across the region.
• A toolkit has also been developed by Oxfam for the explicit purpose of measuring and understanding unpaid care and domestic work, which includes gathering information on the main factors that affect people’s responsibilities for this work, such as access to care services, infrastructure and social norms.
• A review of time-use evidence from Asia finds that infrastructure investments have the potential to reduce the burden of unpaid care work, but rarely do these investments seek to do this explicitly in their design or delivery: an opportunity that could be exploited.
• Organisations such as the OECD and ILO have drawn on time-use data to estimate the scale of unpaid care work, and have applied labour market wage rates to convert this work into monetary figures: 9% of global GDP ($11 trillion) or 15% for OECD countries . While this calculation fails to capture opportunity costs for women, or value the investment this works makes in terms of human and social capital, the translation into monetary equivalents can help to raise the profile of unpaid care work and integrate it into policies that target growth and progress.
• Looking beyond conventional economic models, more pluralistic approaches to progress, including feminist economics and wellbeing economics are gaining momentum. These provide an opportunity to challenge the norms of what is valued and prioritised in our societies. The Indian presidency of the 2023 G20 summit puts a special emphasis on promoting sustainable lifestyles, values, wellbeing, and accelerating the SDGs. This creates the space for leaders to steer progress towards more gender-equitable goals and use gender data to do this in an effective and evidence-informed way.

The opportunities presented in this briefing provide some encouraging technical solutions to effectively fill specific gender data gaps. However, it is also important to look behind these targeted solutions and address the political, social and financial context that can determine whether solutions will be implemented and the extent to which they can be a force for change in the interests of the people furthest behind.

Launched in 2018, the Inclusive Data Charter has quickly built high-level political momentum for inclusive data, garnered commitments from 30 diverse champions and signatories, including DI, and supported these signatories in accelerating and deepening progress in their attention to data.

Financially, addressing data gaps can be expensive. DI has stressed the need for better, more coordinated and nationally driven financing for data ecosystems in general to strengthen the foundations on which gender data is built. Data2X has calculated the specific financing gap for gender data – $500 million a year – until 2030. Welcome commitments, such as from the Gates Foundation , which has dedicated US$80 million to tackling gender data gaps, can go only some way to filling this financing shortfall. In 2020, Paris 21 found funding for gender data had fallen by 55% compared with the 2017–2019 average. Longer-term, less volatile and more substantive commitments, particularly for low- and middle-income countries , is necessary to build robust and responsive gender data ecosystems, while financing for existing data, particularly international household-level surveys that mask gender inequalities, must be repurposed to collect gender-sensitive data.

DI’s ‘Good with data’ podcast also discussed the role that inclusive data governance, management and operations within the data sector itself can play in challenging inequalities to, in turn, improve the outcomes in terms of producing and using better gender data. Our work on assessing data throughout its life cycle, from collection and storage to analysis and use, recognises opportunities to address power imbalances and issues of rights, privacy and ownership over data at all stages, from who participates in the decision-making around data policies, governance and financing, to the people involved in decisions around data accessibility and use. Applying a gender perspective includes:

• Examining gender inequalities in the staffing and leadership of data agencies such as NSOs.
• Looking at how processes for data collection can be more sensitive to cultural and social stereotypes that may introduce bias into the data.
• Trying to advance the use of data, particularly by women’s rights organisations .

Many national statistical systems have still not fully adopted gender mainstreaming as a standard operating procedure, an approach which could systematically expand gender data availability and improve official gender data products.

There is still a significant path to travel , but – together with our partners – DI is working to close the gender data gap. Collectively, we are identifying opportunities to strengthen gender data ecosystems to improve the evidence base that informs efforts to leave no one behind. Please contact us for more information and to discuss the kind of work we can support.

Thank you to Amanda Austin (Equal Measures 2030), Lauren Harrison (Paris 21), and Joanne Crawford (International Women’s Development Agency) for reviewing this briefing and providing comments and suggestions. The content and positions within the brief are those of Development Initiatives alone.