13

Before researching this, I have assumed countries in which people feel safer tend to be actually safer. But take a look at this infographic which was recently circulating the Croatian media:

Apparently, Croatia feels like the safest country in Europe, and the UK feels like one of the least safe. However, if you look at the actual crime statistics for those two countries, the murder rate in Croatia is actually higher than in the UK. And gun crime is by two orders of magnitude more common in Croatia than in the UK.

How is that possible?

Improve this question
14
  • 41
    Because Numbeo does not use scientific methods to arrive at its conclusions?
    – ccprog
    Jul 4 at 20:31
  • 3
    @IllusiveBrian Comparing crime across countries is notoriously difficult. How crimes are defined, the culture of the country (influences charges and convictions), and how its reported all varies from country to country. Even a simple concept like murder isn't so simple. The devil is in the details and that can very much complicate comparisons. For example, some countries will include attempted murder in reported murders. This may help: sciencedirect.com/science/article/abs/pii/….
    – David S
    Jul 5 at 22:08
  • 9
    Because perceptions are relative. Most adults in Croatia can remember a time it was decidedly unsafe to walk anywhere in daylight. Of course just being exposed to low levels of crime feels much safer.
    – Stian
    Jul 6 at 12:54
  • 8
    But on a more relevant note; for the love of all that is sensible, do not take such random statistics seriously. Not even a little bit.
    – user46811
    Jul 7 at 6:30
  • 3
    In my city in France, in the western suburbs of Paris, I can walk in the middle of the night and there is zero chance to have any problems. If I was to walk in the middle of a cité (a "complicated neighborhood" with huge blocks of flats) and yell things that can make people angry, I may have problems (o not). If I was to disturb the drug trafficking there I would certainly have problems. I also spent a few years as a young teenager taking the Paris metro "red line" (a dangerous metro line according to the press) in the evenings and never ever had or saw anything remotely dangerous.
    – WoJ
    Jul 7 at 15:53

9 Answers 9

Reset to default
47

Murder rate (or general crime rate) and "safety walking alone at night" are not necessarily derived from the same sorts of crimes.

Consider for example a country where domestic violence is higher than in other countries. Still rare as an absolute value, but higher than in the other country we compare it with. This means that if you live in a troublesome household, you are at risk. However, if you are not, you are not at risk.

This has nothing to do with becoming a victim of a gang on the streets at night.

A country where solving disputes in the family can more often involve violence, but there are no gangs roaming the streets looking for random victims to assault, might have a higher average crime rate, but you would still be safer walking alone at night.

Note: This answer is not about a specific country. It is a general explanation with the point being there are many different types of crimes, and that different cultures and different countries can have different distributions among these different types of crimes. Therefore one sort of crime being higher does not mean a different sort of crime (for example being attacked by strangers while walking alone at night) cannot possibly be lower.

Improve this answer
3
  • Is there any more specific explanation regarding to the country set of the question, or should be it applied as is?
    – alamar
    Jul 5 at 5:59
  • 11
    @alamar my answer was a general explanation not necessarily specific to the countries in question. My point was simply that different cultures and different countries can have different types of crimes, so some sort of crime being higher does not mean another sort of crime (being attacked by strangers while walking alone at night) cannot be lower.
    – vsz
    Jul 5 at 6:39
  • In some highly oppressive cultures, your father can beat you to death, and it won't be reported as a crime at all. Even in western countries, domestic abuse was a crime for only maybe the last 60ish years.
    – Nelson
    Jul 6 at 7:48
26

Because Perception and Reality are often different.

There is a lot of selection bias, right from reporting by media to gossip to what political parties say to what each person wants to believe.

Everyone's perception is a result of all these biases masked on real statistics. These biases play out differently in different countries.

Improve this answer
9
  • 11
    In addition to this, the kind of crimes, or gossip/social media coverage of crimes, that scare people as opposed to the type of crimes that show up in statistics could explain part of the gap. In the Netherlands there is a lot of theft that goes unreported because police is not interested, but people still discuss it online. Same with random street violence. Another one is appearances: in Croatia the streets at least in the evening belong to laid back families with kids, not rowdy teenagers.
    – Ivana
    Jul 4 at 22:25
  • 5
    I don't know the exact contents of the Numbeo survey, but they track whether people feel safe or not. This is pretty much an apples and oranges comparison to how safe they actually are (statistically). Simply walking through an area which looks rundown and underpoliced may make someone feel unsafe, even if they are in no actual danger of being a victim of a crime.
    – GiantSpaceHamster
    Jul 5 at 14:30
  • 2
    @o.m. theft is not robbery (mugging). My perceived safety in an area with high theft but low personal robbery is not the same.
    – CGCampbell
    Jul 6 at 12:23
  • 5
    This was my immediate thought. At least in the U.S., we spend so much time worrying about child abductions by strangers, using statistics about missing children in general (800K/year, but most are runaways, not abducted at all) and child kidnapping in general (100K-250K/year, but almost all of them are custody-dispute driven parental abductions). The actual number of children kidnapped by strangers is 100-150/year, and while even one is horrible, the risk is vastly lower than people think it is, and in fact it's been getting lower for decades (where perception of danger rose for decades).
    – ShadowRanger
    Jul 6 at 13:36
  • 1
    I'm actually surprised by how low the perceived safety in the Netherlands is according to this survey.
    – user46811
    Jul 7 at 6:17
20

Most people never witness a violent crime, but read about it. So perceived safety is strongly related to how much crime is reported. How often it is reported: Same crime can be reported when it happens, when the victim leaves hospital, when an arrest is made, when the person is released without charge, when they arrest the real criminal, when they go to court, get convicted, go to jail, apply for parole, eventually get released. That’s one crime reported 15 times. No wonder people will feel unsafe.

In the uk, using a gun for a crime is a bad idea. Police will try to get you very hard. Few criminals do that, they use knives instead. So if you look at gun crime, Uk rates are very low.

Who becomes a victim? In some places criminals attack criminals. I feel quite safe even while two gangs are killing each other, because I’m not part of that. Or if there is lots of crime against some minority, the majority may feel safe. Or if you have places with lots of crime, you just don’t go there.

So I feel safe/unsafe either because for some reason my perception of risks doesn’t match reality, or because my risk is different from the risk of other people.

Improve this answer
17
  • 1
    Also not just reporting but general (non-news) media representation: With all the London knife crime memes and drill lyrics, if you're in that sort of media bubble you can very easily get very very skewed impressions/perceptions of danger.
    – Hobbamok
    Jul 5 at 18:43
  • 3
    Personally, in France I feel unsafe because of the police, walking around with machine guns in heavy military gear, giving an impression of a warzone. I wonder if that's a factor as well. I have not seen that anywhere else.
    – gerrit
    Jul 6 at 6:31
  • 2
    The idea that most people never witness a violent crime seems extremely detached from reality to me, everyone knows places you should avoid in their cities because of violence.
    – Eduardo Wada
    Jul 6 at 10:28
  • 4
    Part of the problem with this whole question is explicitly pointed out here, right at the beginning. "Most people never witness a violent crime" ... As I said in another comment, my perceived "safety" would be high in a city with high violent crimes of a familial or 'hidden' nature, yet in the same city, if the mugging rates (without violence, aka simple robbery) were high, I'd never venture out. Safety is a very personal and not well defined thing. In another Stack we disallow questions about safety without well defined parameters, just because mine is not yours. It is opinion.
    – CGCampbell
    Jul 6 at 12:26
  • 2
    No. You specifically said "everyone knows places you should avoid in their cities because of violence." That does not mean those people have personally witnessed or experienced the violence, merely that they possess a belief the violence exists (regardless of that belief's accuracy or legitimacy). When people act to reduce their exposure, their exposure is, absent some very odd situation, reduced. You are trying to argue for the nonsense that because people avoid violence, they would be more likely to witness it. The exact opposite is true, regardless of how much violence actually occurs.
    – Nij
    Jul 8 at 10:58
12

Perception is not often well-informed, and will probably reflect the direction of change rather than the actual longer-term average. The UK is suffering slightly rising violent crime levels starting from an extremely low base.

A long while ago when I visited New York, people were telling me things like "relax, its pretty safe these days". Well, compared to a decade earlier, it definitely was. Compared to London, it definitely wasn't!

On the same trip I visited a friend in a small rural community on the Oregon coast, and they told me to "just let myself in when I get there -- we might be out but we don't lock the door" As a Londoner I found that incredible. But I now live in a village in the UK where nothing bad happens if I forget to lock the door ... but I still lock it because I reckon its a good habit to stay in.

Croatia went through a civil war in living memory. Many people doubtless think back to that, and the high levels of civil unrest before it (and after? I don't know). So today's reality is "much better".

Violent crimes will be front-page news (murders nationally so in the UK) where they are rare and precisely because they are rare. But because of this, people feel more frightened. Where crime levels are higher, these crimes "aren't news" and don't get reported to anything like the same extent. So people go on personal experience and their friends' experiences, which may actually be reassuring. Almost all of Europe is safe by USA standards (and the USA can point at South Africa or Guatemala as worse).

Also, is it violent crime against ordinary members of the public? Or fighting between criminal gangs? The latter makes front page news in the UK but may not ever impinge on most people's lives. Elsewhere it may not be news.

Improve this answer
3
  • 2
    As far as I know, Americans feel rather unsafe, in spite of violent crime sharply decreasing over time.
    – FlatAssembler
    Jul 6 at 14:06
  • I agree with you its all just a matter of security theater, and what is focused on by the news.
    – Questor
    Jul 6 at 18:58
  • 2
    @FlatAssembler it very much depends on where you live, even down to the neighborhood. In my neighborhood, it is very safe, but I know there are places in the same city where I would be very unsafe. The existence of such places doesn't make me feel unsafe though, since the violence doesn't reach anywhere I live.
    – Esther
    Jul 6 at 21:11
7

Croatia is a pretty well-known tourist-friendly country, thus it's quite possible that the primary tourist attractions in there, along with popular tourist routes, are guarded better than towns with no tourism value but where people suffer from general poverty, unemployment and such. Thus, the overall crime rate in the country is high due to those regions, but perceived safety displays low crime rate due to two factors: tourist bias, as most tourists never reach highly criminalized regions, and home population bias due to those localized in such regions have rarely raised their voice to pipe up for their "country" being unsafe, instead any voicing from such regions would be "this region is unsafe" instead, these votes might even not contribute to perceived safety measurement.

TL;DR crime rate is not uniform across countries, and some parts are indeed as safe as perceived.

Improve this answer
1
  • 1
    I wonder if it might be worth mentioning - to say it another way - that noone is by themselves a statistical average. A place might have a high average of crime or anything else, but your experience is a singular one, not an average. If whatever the statistic is requires doing a certain thing, then your chance of the whatever becomes 0 if you don't do the required thing and far above average when you do.
    – Mike M
    Jul 6 at 11:38
4

Because the source of the data might not be reporting accurately on reality

@ccprog has an excellent point about Numbeo being the source of these statistics, which has a number of major problems.

In particular, the "Safety" and "Crime" indexes have a number of issues highlighted in the wikipedia article on Numbeo, ranging from being simple to maliciously alter, to using poor statistical methods: https://en.wikipedia.org/wiki/Numbeo

The main problem, really, is that Numbeo does no sampling, relying instead on crowd sourced statistics. So the chart should read "How safe do people who answer Numbeo questionnaires feel walking home at night?" - there's not really any stats that can allow us to cleanly extrapolate this to the rest of the country.

I suspect this is only slightly more accurate than going onto, say, each countries subreddit, asking people there how safe it feels, and aggregating the data in some way.

Even worse, there's no indications of number of responses to this survey. Maybe Numbeo is unheard of in France, and so only a couple of paranoid people have responded. Maybe hundreds of thousands respond in Croatia, dragging the data back to a more optimistic number. We are simply unable to say from this dataset if more people feel unsafe in croatia than france while walking home at night.

Improve this answer
3
  • The countries with a high perceived security tend to be smaller ones when you compare otherwise similar demographics (e.g. Germany and Austria, or the Scandinavian countries, or the Iberian peninsula). For perceptions there is a difference between "here" and "abroad," and larger countries have higher totals which get disseminated over the whole country.
  • In several EU countries, there are political parties with an interest in inflating perceived crime numbers, notably the "walking alone at night" kind with immigrants as perpetrators. The much higher danger to people is in the own home, from a partner, but that is what people don't want to think about.
Improve this answer
0
2

I'd identify two key reasons.

Fear and statistics

Even assuming the numbers you posted are true and accurate, "how safe people feel to walk alone on the streets at night" isn't the same as "how likely are you to be murdered".

As a general rule, you are more likely to be murdered by someone you know. Here's the FBI's Crime in the US 2019 Expanded Homicide Data, showing that out of the roughly 50% cases where the offender is known, 20% (which is about 9.9% of those 50%) were strangers. Previous years show similar numbers, and there weren't more up-to-date numbers.

The point though isn't to be extremely precise. Numbers might vary from country to country, by gender, and by a number of other factors, but in general, it's true. You are more likely to be murdered by someone you know. A lot more likely.

But is it what you fear? Probably not. Without going deep into psychology, humans have a tendency to be more afraid of the unknown, and "walking alone at night" is prime overactive imagination, "I'm gonna get murdered by a serial killer" dubiously-rational fear territory.

Media coverage

Not all crimes get coverage proportional to their statistical occurence or likelihood.

This is true for both news reports, there is after all an economic imperative to generate views which tends to favor more spectacular reports.

This is also true for fiction, there are for instance a lot more police procedurals and other crime dramas featuring homicide detectives than FinCEN analysts.

Imagery of violent crime perpetrated by random strangers is probably going to have a more powerful influence on your fears than a pie chart from the FBI.

Improve this answer
1

Crimes that "make you feel unsafe walking at night" are mainly robberies or assaults. They don't involve guns or murder.

I acknowledge vsz and several others said this too. Perception vs Reality is also a good point. I feel this answer has 75% of the truth in a much smaller statement, which is a virtue in itself.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .