ONS DFID Data Science Hub

The Office for National Statistics (ONS) and the Department for International Development (DFID) created the Data Science Hub in 2019. It is based in East Kilbride near Glasgow at one of DFID’s headquarter offices.

We use data science to apply the tools, methods and practices of the digital age to create new understanding and improve decision-making. Our focus is to use data science to make more effective development progress in low and middle-income countries as they work towards the Sustainable Development Goals.

We can draw on the deep technical expertise of our Data Science Campus colleagues alongside the world class development skills, knowledge and experience of DFID. It includes data scientists, trainers and developers, business support and user engagement staff. The team includes expertise in machine learning, natural language processing and spatial analysis, as well as colleagues with epidemiological backgrounds.

As part of the wider ONS Data Science Campus team, we work with data science colleagues in multilateral and bilateral development agencies and academic and research institutions in the UK and internationally.

Our work falls into three main areas:

  • Building and working with analytical tools to support development progress and decision-making.
  • Providing training related to data science techniques.
  • Developing mentoring relationships with emerging data science work in our development partners. These include colleagues in DFID, national statistics offices in developing countries and other parts of their governments, and those working in non-government organisations.

Building analytical tools

These can be either developed by the Hub team or extensions and modifications of tools that have been built by others.

For example we have been exploring the potential to provide information on cattle numbers in South Sudan. This is a key aspect of their economy but difficult to measure on the ground. We use satellite imagery to automatically identify the cattle camps – and examine if estimates of cattle numbers  can be drawn from indicators using high resolution images.

In relation to extending existing tools, we are using the work done by the Data Science Campus on shipping indicators in the UK to look at shipping activity in developing countries, and its relationship to economic activity. This is particularly relevant in the context of examining the secondary impacts of COVID-19.

Providing training related to data science

We provide a range of different training courses including in coding languages (for example R and Python) and data science techniques (such as machine learning and natural language processing). In addition we also provide training to advocate for the greater use of data science in decision-making in developing countries, as well as demonstrating examples of how this can be done.

Mentoring colleagues in DFID and in developing countries

We work with partners who want to incorporate data science activities into their work in developing countries. Analysts and emerging data scientists in our partner organisations are encouraged to identify and develop data science activities that can help their work. Our team work to guide, advise and problem-solve with them. We work with many organisations but we have a particular focus on supporting the national statistics offices where the ONS International Development Team have embedded advisors (Ghana, Kenya and Rwanda) as well as working with the United Nations Economic Commission for Africa.

How to work with and contact us

We are always happy to discuss potential work and support with new partners to see whether and how this fits with our current priorities and resources. If you would like to do this, please get in touch:

Tom Wilkinson, Site Lead, DFID Data Science Hub

Tim Harris, User Engagement Lead, DFID Data Science Hub