Today’s inflation figures showed prices rising faster than at any time in the past three years. But that was not the most remarkable aspect of the data. Until recently, CPI – the monthly update on inflation – has paid little attention to housing costs, featuring instead a basket of goods and services. Now, the Office for National Statistics has put almost all of its focus on a newer and controversial measure called CPIH which adds owner-occupiers’ housing costs to the basket. The idea is to reflect the costs associated with living in one’s own home, something that is estimated to take up a tenth of household expenditure. But so far it has failed to win the official badge of excellence, classification as a “national statistic” by the nation’s fact-checker, the UK Statistics Authority. The ONS hopes that the change helps to bring official measures closer to individual experience. But rushing to include it before the way it is calculated has the kitemark of reliability risks fuelling, rather than healing, the scepticism that was exposed and partly legitimised in the EU referendum campaign.

The trouble with statistics is sometimes the statistics themselves – last year, MPs on the Treasury select committee were scathing about the “lack of intellectual curiosity” and failure to respond to consumer need of the ONS – but sometimes it’s also a failure to understand what statistics can, and can’t, explain. Their introduction in the “Whitehall knows best” years after the war transformed public policymaking. But in an age when people see themselves as individuals first, part of a community second, and when geographical and individual inequalities have grown sharply, official statistics have sometimes felt inadequate as a way of describing what life is really like. It was hard, for example, to believe in the idea of immigration as a national economic bonus when it was experienced as oversubscribed GP lists and overflowing classrooms.

One response has been to provide much more localised, and sometimes individualised, analysis of the big picture. Economic growth nationally looks quite different when looked at regionally. The UK growth rate between 2014 and 2015 was 2.1%. But in the East Midlands, it was only 1%, and while it was higher in the West Midlands (1.6%), people in Solihull were still contributing less to GDP than their neighbours in Nottingham. Inflation itself affects different income groups differently. The economist Paul Donovan talks about inflation inequality: it is expensive to be poor, cheaper to be rich. His calculations reveal a striking contrast between the rises in costs faced by the richest 20% and poorest 20% in a number of advanced economies in recent decades. In the UK, between 1997 and the beginning of 2017, prices for the richest fifth rose 40% but for the poorest fifth they increased by 51%.

The ONS boasts that confidence in its statistics remains high: four out of five people trust them. It’s politicians whose use of the information is questioned. The Department for Education has just had to acknowledge that the student satisfaction survey was not nearly robust enough to be used to assess university teachers’ excellence. The £350m NHS Brexit bonus was officially discredited. Politicians must behave better. Treating data with the proper respect is an important part of persuading people to adjust their prejudices to accommodate the facts.