Before You Change Perceptions, Check the Reality

Ask Londoners about the biggest issue facing the city and they typically say knife crime. Ask instead about their local area, and knife crime ranks much lower than issues like antisocial behaviour, drugs and burglary. Working in policing has shown me how easily perceptions can shift – one subtle change to the frame can change the response entirely. It has also taught me that perceptions matter more for the Met than any other organisation I’ve worked with.

Your view of a supermarket is shaped by your frequent shopping experiences. With the police, there are far fewer opportunities to form an opinion through direct interaction. Unless you’re the unfortunate victim of a crime – or a suspect – you have little reason to interact with the police at all. Perceptions of the police are therefore often constructed at a distance: from what people hear in WhatsApp chats, see on social media and pick up from the world around them. So what does policing teach us about how perceptions are formed?

The Closer Something Gets, the More the Answer Changes

Let’s return to the conundrum at the start. Why are people more concerned about knife crime in London than in their own neighbourhood? One explanation is psychological proximity. Most of us don’t experience “London” directly. It is too big and abstract, so we construct an idea of it from the stories that stick in our minds.

Knife crime is unusually salient because it is both frightening and severe. It’s no wonder individual incidents become part of the London-wide picture. And whether it’s knife crime or general crime, this mental picture is often much darker than the underlying data. According to recent research from KCL, the public reckons that 30% of Londoners experience crime each year when the real figure is 12%. Nearly two-thirds think overall crime has risen since 2019, when it has actually fallen slightly.

Our neighbourhood is different to this idea of London. We walk through it. We see the graffiti by the station. We witness the antisocial behaviour on the high street. Our local concerns are therefore grounded in day-to-day experience. “What is wrong with London?” and “What is wrong where I live?” might sound similar, but psychologically they are very different.

Where You Live Matters Too

We also tend to explain attitudes through demographics. I often hear that “young people hate the Met while old people love it.” This is only half right. My analysis has found that, even after controlling for demographic differences, where someone lives has a meaningful relationship with their attitudes towards the police.

A 30-year-old living in Richmond isn’t necessarily going to have the same views as a 30-year-old living in Lambeth – even if they are the same age, have a similar income and share other demographic characteristics. I don’t fully know why geography is so important. It’s probably a combination of crime levels, population density and historical relationships with policing. But that is precisely what makes it interesting. Attitudes are not produced only inside our heads. They are partly produced by the environments we inhabit.

I’ve collected plenty of examples of this for Uncommon Thread, my resource that finds patterns across seemingly unrelated sectors. One of my favourites comes from Germany: more than three decades after reunification, electoral maps still show a striking divide between the former East and West, with the AfD performing particularly strongly across much of the former East.

Sometimes Perception Is Reality

There is another danger when talking about perceptions: assuming they must be wrong. Take police visibility. If people say they see police officers patrolling less regularly than they used to, it would be easy to reach for a behavioural explanation. Perhaps negative stories are more memorable. Or perhaps people simply don’t notice when officers are there.

But there is a more straightforward possibility: there really are fewer officers. Met officer numbers fell from 34,523 in April 2023 to 31,181 in March 2026, according to London Assembly analysis, while the force continues to operate under significant financial pressure.

Behavioural science is great at understanding why perceptions diverge from reality. It is less helpful when used to explain away every uncomfortable finding. Sometimes, people think something simply because it’s true. Or, as the line often attributed to Freud goes: “a cigar is just a cigar.”

When the Number Is Right but the Conclusion Is Wrong

The opposite problem happens too. A number can be accurate while the story we attach to it isn’t. I see this with measures of police visibility. In some boroughs, around four in ten residents say they regularly see officers locally; in others, it can be closer to one in ten.

Those numbers are real. But it would be wrong to conclude that the first borough has four times as much policing. What we are measuring here is reported visibility – not actual patrol time – which can overestimate or underestimate what is happening on the ground. Officers concentrated around busy town centres will be seen by large numbers of people, while officers working in residential areas will be less visible.

Similarly, the borough of Bromley is 12 times larger than Kensington and Chelsea – so visibility is never a simple like-for-like comparison. First check the measurement is right, then check the story you’re telling about it.

Funnily enough, this learning is just as applicable to the world of sport. Sir Alex Ferguson sold defender Jaap Stam to Lazio in 2001. The explanation later attached to the decision was that some of Stam’s defensive statistics, including his number of tackles, appeared to be declining.

But fewer tackles do not necessarily mean a defender is getting worse. A player with better positioning needs to make fewer tackles in the first place. Ferguson later described selling Stam as a mistake.

Diagnosis First. Persuasion Second.

All of this has made me much more cautious about the phrase “changing perceptions”. Before trying to change them, we need to understand where they came from.

• Is there genuinely a gap between reality and perception?
• Is someone’s immediate environment shaping what they believe?
• Are we measuring the perception badly?
• Or are people simply describing reality accurately?

Those aren’t academic distinctions. They require completely different responses. If people are misperceiving something, better communication might help. If their environment is shaping their experience, a comms campaign will do very little.

If we have interpreted the measure incorrectly, we need better analysis. And if the perception is accurate, no amount of reframing is going to fix the underlying problem. Behavioural science is enormously useful for understanding the gap between reality and perception. But sometimes its most useful contribution is helping us work out whether there is a gap at all.