Research Brief: July 2026

Welcome to the July research briefing from the Society of Evidence Based Policing. This presentation introduces six of the month’s twenty-two studies. We will explain what each study suggests, where caution is needed, and how practitioners might use it without overstating the evidence. We begin to research from Wales, United Kingdom. Victims who felt officers listened and took them seriously reported markedly higher satisfaction with the police response. There is an important qualification. This identifies a strong association, not proof that one behaviour alone caused satisfaction. Experience, case outcome and vulnerability may overlap. For practice, the useful question is this. Treat communication as an observable part of service quality. Measure whether people understood what happened, felt heard and knew what would happen next. Next, we turn to research from England and Wales. Participants saw value in carefully targeted tools, but described governance, testing and institutional capability developing more slowly than adoption. There is an important qualification. This is a detailed qualitative map of a fast-moving system, not an impact evaluation or a measure of how frequently each risk occurs. For practice, the useful question is this. Start with the operational problem. Require realistic evaluation, publish limitations and define the human reviewer’s actual task. Next, we turn to research from Sweden. Recorded assaults fell during the winter break, especially among young people, while residential burglary rose and visitor destinations saw increases. There is an important qualification. Staggered holiday dates and pre-trend checks strengthen the comparison, but it remains an analysis of recorded crime rather than a randomised trial. For practice, the useful question is this. Use school calendars as an operational variable: prevention and patrol patterns can shift with the changing mix of youth, residential and visitor demand. Next, we turn to research from Santiago, Chile. The programme did not produce a statistically clear reduction in overall crime frequency or harm. Officers completed around 30% of scheduled visits. There is an important qualification. Random allocation gives a strong comparison, but low delivery and activity in control locations make the programme’s main test harder to interpret. For practice, the useful question is this. Patrol dose is part of the intervention. Track visits, supervise delivery and report both assignment and what actually reached each location. Next, we turn to research from Metropolitan Police, United Kingdom. Participants agreed that visibility, engagement and problem-solving were central, but did not share a consistent definition or operating doctrine. There is an important qualification. The study offers a useful picture of one large force at one point in time. It does not show that one model produces better outcomes. For practice, the useful question is this. Translate the neighbourhood promise into an operational minimum: ownership, protected time, problem selection, community influence and measures of delivery. Next, we turn to research from United Kingdom. Personal resilience and internal locus of control were associated with better wellbeing, but exposure and organisational conditions still shaped risk. There is an important qualification. Self-report and a single time point cannot establish cause and effect or show whether resilience training changes later outcomes. For practice, the useful question is this. Do not turn workforce protection into an individual test. Combine selection and preparation with supervision, workload control, peer support and follow-up.