In the run up to the launch of the 2022 Aid Transparency Index, our own Alex Farley-Kiwanuka explains the monumental effort undertaken by her team to combine four global datasets in order to track Women’s Economic Empowerment funding across three countries. Alex also shares why she feels the Aid Transparency Index is so important for maintaining and improving international funding data and making it useable for researchers like herself.

It is now well established that investing in women’s economic empowerment has a catalytic effect in reducing poverty for all people and achieving gender equality. Women’s economic empowerment (WEE) is key to the realisation of women’s rights and their full participation in society and the world of work, helping both to reduce poverty and increase inclusive growth. Speaking bluntly, this is about ensuring that women have access to decent work, that they are paid fairly and equally for that work, that they have control over their time, lives and bodies, and that ultimately they have increased agency, voice and meaningful participation in economic decision-making at their household and institutional level.

What we’ve tracked and why

One of the barriers to effective investments is the limited understanding of who is funding what and with what results. In October 2020 we initiated  an effort to overcome this barrier, launching “Women’s Economic Empowerment: building evidence for better investments”. This project has since tracked funding to WEE as well as women’s empowerment collectives (WECs) and women’s financial inclusion (WFI), which are accelerators for WEE. We have also analysed Covid-19 funding , assessed the gender intentionality of international donors’ funding approaches, and examined funding for unpaid care work. We have studied Kenya, Nigeria and Bangladesh.

Early on we needed to develop definitions and frameworks for the four focus elements: women’s economic empowerment, women’s financial inclusion, women’s empowerment collectives and gender integration. In total 45 key stakeholders and subject matter experts helped us develop our overall methodology, engaging in an iterative process, to fine–tune and contextualise our project’s approach to each country.

How we built the dataset

Using our methodology as a foundation we then had to build a dataset for our analysis of international donor funding. This was a serious undertaking. At one point our entire team was involved, then we had to draw in experts from other parts of our organisation, then our laptops started crashing under the weight of the data. The resulting data collection methodology is 33 pages long!

Our dataset combined information from four sources.

• The Country Development Finance Data tool (which gave us access to International Aid Transparency Initiative (IATI) data)
• The OECD Creditor Reporting System
• The Consultative Group to Assist the Poor (CGAP) Funder Survey
• The Foundation Directory Online (Candid)

With one large dataset in hand, we then had to identify and classify all appropriate projects from among the many thousands of development programmes. This involved a combination of using existing markers and sector codes, coupled with word searches on titles and descriptions (noting that word searches needed to be undertaken in multiple languages to ensure completeness). We then needed to clean the data and take out duplicates – please don’t ask us how long that took! As a second step we then analysed the dataset against COVID-related terms in order to determine the extent to which COVID related funding was (or wasn’t) supporting existing WEE efforts.

What struck me at this point was how much we needed to rely on some of the more basic elements within the data. I think we were hoping that we’d be able to do much of our analysis using the data elements such as sector codes, markers and titles, but ultimately, we ended up relying heavily on the information provided in the description field. The Aid Transparency Index rewards publishers for descriptions which include a description of what the project intends to do and how it intends to do it, and if possible, information about who the project intends to benefit and where the project will be implemented. While many international funders provide comprehensive descriptions for each of their activities simply don’t, making it nearly impossible to find the activities you’re looking for if you are using datasets of the scale which we were.

The importance of international funding data

The analysis which our team has been undertaking simply wouldn’t have been possible ten years ago. While the OECD CRS data has been relatively comprehensive for many years, the data published in the IATI Standard, with all its documents and richness, has only become truly useable in the last few years. And it wasn’t as if it happened overnight; rather, the data set has reached a maturity in terms of its completeness, timeliness and quality that means that we, and many others, can and are now using it. Driving this improvement has been the Aid Transparency Index – run by the research team on the other side of the office from where I sit. The Aid Transparency Index is effectively a giant data quality checking effort, and it provides detailed feedback to publishers to help them improve. The Index doesn’t cover every publisher, but it does cover those providing 84% of the data currently held in the dataset. And one area where the Index team is continually looking to improve is how the Index can respond to the needs of data users. For example, other researchers have recently had the same experience as we have whereby project descriptions are insufficient to do basic word searches across large data sets. The Index team are currently looking at how they can adjust the weightings associated with the description indicator in the Index to reflect the importance of this field and further improve data quality for all users.

Sharing our analysis

In July, just after the launch of the Aid Transparency Index, we’ll be launching our reports for Nigeria, Kenya and Bangladesh. These will illustrate the breadth of what’s possible given the state of aid data today, and they’ll also demonstrate the level of granularity which is achievable.

We’ll be sharing our findings with stakeholders in Kenya, Nigeria and Bangladesh to support their evidence-based advocacy for WEE policy and funding. These findings offer insights for funders that encourage more effective and coordinated funding for WEE. Our reports will include recommendations on how funders can improve their reporting and publication of data related to WEE, so that we can monitor progress and impact more sustainably. As governments and funders increasingly focus on recognising, reducing, redistributing, representing and rewarding care and domestic work, we hope that our findings will provide insights on unpaid care work and inform policy and investments. We look forward to exploring ways to expand this work and monitor progress on funding commitments.

Other reading:

Following the money: using data to track development spending in Nigeria