Data scraping to pull together budgets for Rwanda and Uganda

13 May 2014 ⋅ 

Written by Simon Peters and Gertrude Nandyona

The Development Initiatives Africa Hub analysts have been ‘data scraping’ to make the first spreadsheets of the Uganda and Rwanda budgets in soft copies. Gertrude Nandyona and Simon Peters, Development Initiatives Analysts, blog from the front lines of the Data Revolution to explain how they did it – and why it matters.

Why do we need soft copies of Uganda and Rwanda Budget data?

At the Development Initiatives Africa Hub, our analysts need budget data to inform the research we do on poverty and the financial resources available in the East Africa region. But accessing budget data in the East Africa region is no easy task. First, you need to know where that data can be found. If you’re thinking of just downloading a neat spreadsheet from the Government website then you’ll be disappointed. If you are lucky you can get it online but in PDF formats which make it difficult to use.

The inaccessibility of budget data was a huge barrier to the work that we wanted to tackle. Hence the beginning of what was to be a year-long ‘data scraping’ exercise, beginning mid-2013. We wanted budget data available in formats that would make our analysis much faster. So the data scraping journey began.

What is data scraping?

In simple terms, it is hunting and gathering data from various sources to create a dataset. For the budget data, we ‘scraped’ information from the Ministry of Finance’s budget estimates books (by extracting data and manually inputting it into spreadsheets) and used the information from the sector budget documents of both countries, including Medium Term Expenditure Frameworks, budget review documents and budget outlook papers to link the estimates to outputs. We have collected these together and now have soft copies of all relevant budget information for both countries in Excel spreadsheet format.

What data do we have?

The Uganda budget data covers five financial years: 2009/10 to 2013/14. The Rwanda data covers six financial years: 2007/08 to 2012/13. We expect to update these datasets on an annual basis.

Who will benefit from this data?

We want to make this information available to the many individuals and organisations who cannot access the budget books from governments. We also want to make this information more accessible to increase analysis and stimulate discussions on resource allocations in the region. We did it for the civil society organisations we work with, who really wanted the budget data, but did not know where to get it. We also did it so that governments recognise that there are people who (despite the bulk of the hardcopy budget books and PDF files) are daring enough to engage with it. This will allow us, and our civil society colleagues, to engage governments in discussions on resources and encourage greater accountability for these expenditures.

What have we done with the data so far, and what more will we do with it?

Small pieces of analysis have been carried out so far: using the Uganda budget to analyse government spending on urban youth, and also on disability. The data is being used to inform Development Initiatives’ 2014 Global Humanitarian Assistance report to understand what resources governments in the East African region are spending on humanitarian response. It will also be useful for much of the Hub’s upcoming research on East African poverty and the financial resources available for poverty reduction.

What’s next: sharing the data with others

During May 2014, we will be engaging the Ministries of Finance, both in Uganda and Rwanda, in a ‘data verification’ exercise, to ensure that the data is accurate and a true representation of the budget data. This is important to us as we would like to make this data available to the public online. If this is successful, we will share the budget data here on the Development Initiatives website and also on the Uganda Open Development platform, and plan launch events to engage the local community and encourage the use of the data – such as hackathons, and round table forums. Watch this space!

Why did we do all this work? Why is this important?

We hope we will make a difference by using this data ourselves, in our research, and engaging with governments about it. But of course the real difference will come if we are able to make the datasets available online, as people will engage with the data and turn it into useable information. As we all engage with the data, we will contribute to the Data Revolution. The ultimate objective of providing this data is to provide better information for decision makers, and encourage better resource allocation to meet the needs of citizens. We believe that access to information can translate in this way to better livelihoods for households and individuals in East Africa.

What can you do to be part of this work?

Tell everyone you know that Development Initiatives has soft copies of the Uganda and Rwanda government budget data, and please get in touch with us if you want further information. We hope to release the data online for use by others once it is verified – please stay tuned to this blog and to our Twitter feed @devinitorg for updates.