Research Methodology

Spotlight is the research series of the Community Impact Policy Institute, the thinktank and research arm of Fedcap, and its approach is one we call Research in Real Time. Rather than studying complex social and economic challenges only through aggregate data or after the fact analysis, Spotlight follows them into communities to understand how they are experienced, how communities respond, and which solutions show the greatest potential to make a difference. Lessons and promising practices are captured as they emerge, not simply once a programme or policy has concluded, and what is learned connects directly to the people shaping policy. Spotlight on Breaking Britain's Barriers applies this approach to economic inactivity, community by community. The research unfolds across three sequential chapters, examining early indicators in childhood, through youth disengagement at the transition out of compulsory education, to adult economic inactivity. To investigate this pathway, we use a mixed methods design combining aggregate data analysis with on the ground research in localities.1

The quantitative strand draws on government statistics, published datasets and Fedcap UK's outcome data. The qualitative strand will consist of in depth, purposive interviews with customers of the interventions, service users, frontline delivery staff, local partners and elected representatives. Interviews are semi structured and probing, designed to draw out honest, person centred responses, and all are recorded, transcribed, coded and thematically analysed.2 They are not testimonials, and the study does not treat them as such. Their function is to explain the nuances numbers cannot. Each trajectory is heard from more than one vantage point, including family members, teachers, delivery staff and local representatives, so that accounts are corroborated or contested rather than taken alone, and we deliberately retain what is honest and unresolved, the gaps and lingering difficulties, rather than presenting only success.

Two limitations qualify the interpretation and generalisability of these findings. Firstly, our sample is drawn from Fedcap participants, to whose data we have direct access. This access lets us reach and hear candidly from people rarely represented in research. It also carries a selection effect that should be kept in mind when reading the findings: the sample is purposive rather than representative, and because participants are, by definition, people the system reached, those it lost entirely are the least likely to be present. The chapters that examine where the system loses people should be read with that absence in view. Secondly, we are not claiming population level causal inference. Programme evidence is presented as case study rather than impact evaluation, and the study makes no counterfactual claims about what would have happened without the intervention. Our design does, however, allow us to identify the interventions at play and to assess what kinds of policies have the best chance of making a difference. Research of this kind is judged by its credibility, transferability and trustworthiness rather than by statistical generalisability.4 Thus, well-chosen local cases can carry insight well beyond their setting.5