Charity Support, Advice, Impact Measurement, Philanthropy Impact – NPC

The human side of big data

Hard Edges takes as its subject people facing severe and multiple disadvantage. By combining previously separate data sets on homelessness, substance abuse and offending, the report offers a comprehensive picture of the lives they lead, calling on the government, local authorities and the voluntary sector to bring about deeper, systemic reforms.

I’m no expert on homelessness, substance abuse or offending, so I will leave commentary on most of the report’s findings to others (though several are striking, even to a generalist). Rather, my perspective is that of a researcher of the charity sector, working as I have for the last 11 years to better understand social issues to help inform the approaches of charities and their funders.

I was first struck by the complexity of social issues during research on the subject of child abuse back in 2006-7. Child abuse is a product of many risk factors—from deprivation to parental mental health, substance abuse and domestic violence. To even contemplate advancing ‘solutions’ to a problem such as this, you are rapidly drawn into a spider’s web of interwoven issues.

In other words, reality is messy and complicated.

Commissioned by the LankellyChase Foundation and carried out by Heriot-Watt University, Hard Edges makes such a significant contribution to our understanding of severe and multiple disadvantage because it acknowledges that to better meet the needs of service users we must first understand the extent of overlap between the homeless, offender and drug misusing populations.

This insight challenges the very fabric of the charity sector and public services, comprised as it is of organisations set up to deliver services in response to a single issue. Working in silos isn’t just a risk, it’s a reality, as many frustrated charity workers and public servants will attest.

To respond to this challenge, our approach to data and evidence needs to evolve. We need to start looking at the relationships between issues—the overlaps, correlations and causal links—to improve our understanding of where and how to act. And we need to do exactly what this research has done and bring together data sets on related issues to forge this analysis.

For this reason we call on government to pursue the open data and data sharing agendas vigorously as possible in our forthcoming manifesto.

The social sector is buzzing with excitement about the potential of big data. Whether or not this is justified, I think there’s a way in which integrated big data sets allow for the opposite: small data that helps us understand the personal, human perspective on these issues. Without this integration, we often end up with broad brush generalisations about what causes a social issue. With it, we start to see how, in each person’s life, there are different links between different situations, risk factors and events.

What struck me most in reading this report was the finding that only 1% of people who experience homelessness, substance abuse and offending together turn to their parents in a crisis—compared to 17% and 40% of those experiencing one or two out of three disadvantage domains respectively.  What a stark fact in light of our own experience and direct studies on the importance of parental support.

While there’s a climate of fear around data sharing—for many good reasons—this research reveals what’s possible if data is shared with the appropriate security and structures in place. We have to work towards a world in which this becomes easier if we are ever to reach a more nuanced understanding of how social issues come about, and how we might tackle them.