Friday, September 6, 2013

Quick hit: rape and violence in Sweden is not increasing

For people who have heard that Sweden is breaking apart and rapes and violence are at RECORD HIGHS due to the immigrants: it's bullshit. This is the way rape and violence have developed in Sweden over the past couple of decades (note the time-scale, they vary a bit):
The upper line is for victims of "any violence" (16+ years of age), the lower for "assault" (16-79 years of age) (percent).

These lines are for "sexual crimes", upper line women, lower line men, and the total in the middle (percent).

Now some people may say "hey, that doesn't fit with this article I read!". Well, that's because you read a shit article. Any article that uses "reported crimes" for Swedish criminality is utter garbage. Why? Let's take a look at deadly violence - with deadly violence we can relatively easily show the real rate, since dead people are examined for cause of death. Here's a graph about deadly violence in Sweden:
Number of cases of deaths due to deadly violence in Sweden. Thick line according to reported crimes, lower two lines are cause of death-statistics from hospitals and the Swedish Crime Prevention Council.
Notice how murders have tripled according to reported crimes since 1990 while hospital statistics don't bear it out? That's because the police changed their system of reporting in 1990 and stopped removing duplicates, among other things. In general, crime statistics suck because a lot of it is based on trust of  the police and willingness to report, rather than the actual amount of crime. However, reporting about crime in a panicked fashion makes the media happy, so these are the graphs they use:
Amount of reported sexual crimes ("other sexual assault", rape, sexual abuse, indecent exposure).

Amount of reported assaults (lower lines against women and against children).
The graphs I showed in the beginning are according to crime victim surveys. These are by no means perfect, but the results are much less sensitive to the system used for police reports and willingness to report. They are a far better indicator than crime reports, which should never be used for that purpose.
Of course, if you want to argue that violence in Sweden should have decreased more since the 1980s when considering the lead theory, that's more of an interesting argument, but essentially, very little has changed over the past couple of decades in Sweden, crime-wise.
All data from here.
One extra thing, this is a graph over the number of people treated (per 500 000 people) in hospitals due to violence, with knife and gun violence in separate lines:
 

No comments: