Section · 5 articles
Reading police numbers honestly — taking them one at a time, showing what they do and do not support, and revisiting a first interpretation when the evidence demands it.
British policing has promised to electrify a fleet that is four-fifths diesel — but the pursuit, the armed response car and the 3am rural shift are governed by physics a procurement target cannot waive. A guide for the people signing the forms, with two interactive tools: match the car to the role, fund the charging first, and do not let a demonstration write fleet policy.
A fastener manufacturer's PR ranking named Worthing the worst town in the UK for shoplifting. The real driver was a police reporting scheme working as intended — and the force didn't say so for fourteen months. A case study in reading crime data, and in explaining it before someone else does.
Surveys tell us whether public confidence in the police rose or fell. This dataset shows the language that ran alongside it — a quarter-century of UK news headlines about the police, reduced to words, tone and themes. The two curves turn out not to be the same shape.
You have seen the headline: domestic abuse rises 38% when England lose. It comes from a real study — but the real story is more careful, more troubling, and more useful than the number that travels. A More or Less Policing look at what the evidence actually supports.
A new strand, in the spirit of Radio 4's More or Less: police statistics are everywhere, and most of them are easy to misread. Here's how to read them — starting with the trap inside every stop-and-search figure.
