Merthyr Food Poverty Data Discovery

We have recently completed the Discovery phase of the Merthyr Food Poverty Data project, which was formed to understand if citizen data held across multiple council systems could in some way be combined to help officers identify, at the earliest opportunity, citizens at risk of falling into poverty.

Background

Merthyr Tydfil council has a total population of around 60 thousand. Median income levels are the second lowest in Wales with around a quarter of pupils eligible for free school meals.

The council’s work on tackling poverty spans all policy areas including Education, Community Regeneration and Social Services. Its approach is to focus on early intervention and prevention.

Taking learnings from its pandemic response, the council recognised that doing things the same way as in the past was not going to be enough to deal with the cost-of-living crisis facing its citizens. A new perspective needed to be found.

What was needed was a reliable way to identify and track citizens at risk of falling into poverty well before their situation became acute. Gaining a better understanding of existing data was deemed the way to solve this aim.

The council leadership recognised the need for specialist help so it enlisted the support of the Welsh Local Government Digital team, and a discovery project was formed.

What we did

Over a three-month period, we collaborated with multiple functions within the council, and a handful of its third sector organisations including local food pantries, housing providers, and national charities.

From the beginning we had support from Welsh Government Ministers, its Communities and Tackling Poverty Department who provided financial support and helped us network with poverty organisations outside of Merthyr, and its Data Science Unit who provided helpful advice on, amongst other things, best practice data handling.

The project encompassed three main pillars of activity: user research, data research, and data handling.

User research

Our aim was to conduct research to provide a citizen perspective of living in food poverty in Merthyr.​ To achieve this, semi-structured interviews were carried out with a random sample of citizens. Analysis of these interviews helped Chris, our User Researcher, build several citizen personas to enable the data-focused team to identify which data is likely to be most useful.

Data research

Fundamentally, this is a data project, and any potential solution needed to include new or improved ways of using data. So, we focused our attention on how council officers engaged with information daily and what, if any, additional information would help them be more effective and targeted in their interventions.

We interviewed over 20 officers and support people from across Merthyr council, the third sector, and other public sector bodies, to get an understanding of what information is used to deliver council services to citizens, or what is used to oversee the performance of contracted-out food poverty initiatives.

We also needed to explore a way to bring data into a single place, so in parallel to our conversations with the officers, and with the support of Merthyr ICT, we looked at potential suitable platforms based on leveraging existing private sector relationships.

Seven Merthyr council and three external partner IT systems were analysed to gauge potential usefulness with data from each categorised as demographic, health status, financial payments, and social services.​

The SAIL Databank team at Swansea University also advised us on how its collection of 60 plus research databases could add value.

Findings showed that within the council, most officers accessed/updated information via one, sometimes two, applications related to their service. Knowledge of other sources was limited. Where access to other applications was required, it was usually done on an infrequent, as needed, basis.

Data handling

It was important to ensure that any suggested solution would adhere to good practice around data handling, to secure and protect citizen data.

Following guidance received from several sources we concluded that any data sources originating from Merthyr council’s own internal applications should be covered by existing internal data sharing and data protection agreements. Where information is gathered or shared from/with third sector partners then existing data protection agreements would need to be reviewed and updated accordingly.

Additionally, policies relating to data protection and confidentiality were strictly followed throughout Discovery. At no point was citizen data requested or accessed.

What we found

Our research demonstrated that information relating to individuals and households in food poverty is scattered across many systems with access often kept within specific teams. Data is often duplicated with some systems having more accurate up to date information than others. When combined this means extracting knowledge and actionable insights is difficult.

It became clear that no one application provided a wide enough, all-inclusive view of a citizen and the combined life challenges they face, so investigating ways to join up information from multiple applications would likely be beneficial.

As would investigating potential ways to combine data sources and process a set of intervention indicators, which when combined could provide an early pointer of a citizen slipping into food poverty.

Recommendations

The team agreed that the stated discovery goal had been achieved, and recommended investing in an alpha phase to further investigate and test potential solutions to meet the requirements identified in discovery.

The Merthyr leadership team reviewed our discovery findings and recommendations and have agreed to continue to support the project through alpha. We’ll share more details about the alpha phase of the project in a future blog post so watch this space for updates. You can also sign up to our newsletter to hear more about the work we’re doing as a team.