montestewart wrote:I was curious about the data from Sweden, so I went to the source, and as far as I can tell, the chart above was created by an individual, using data from U.S. and Swedish national agencies (websites links provided beneath the graph). The U.S. agency (DOJ) website had a tool for compiling data on violent crime, and that looks to correspond with the blue line in the U.S.-Swedish comparison. I was unable to find any comprehensive table or tool on the Swedish agency website to produce a comparable historical graph for violent crime.
I agree that the definition of "violent crime" is likely to vary from country to country, so there's no use in trying to directly compare the aggregate numbers. However, that doesn't mean that the fact that the trend lines in violent crime show a massive deviation starting in 1990 isn't relevant. As long as there is consistency in the way they are gathering the numbers over time, we can infer that Sweden's numbers are continuing to climb relative to American numbers.
montestewart wrote:In other words, I have no idea how the person arrived at the figures, and glancing at various other websites, charts, and graphs regarding Swedish crime, I think the numbers are wrong, because most of the categories that I found seemed to show crime rates holding steady or even declining in some cases (like homicide). In searching for accurate statistics, I found numerous sites and references to a few things:
1) Many references to the problems with comparing crime statistics of one country to that of another country. In this case, I couldn't find a definition for "violent crime" on the DOJ website, and saw no comparable category on the Swedish webiste. Without the definition, I didn't know which individual categories to use in producing a total of "violent crime" in Sweden. A number of websites noted such things as the rate for reported kidnapping is higher in Canada than it is in Mexico, or that a U.N. sponsored survey found as many as 25% of males admitted to having participated in r***s (I'm not supposed to spell this out, right? I think I was warned once.) in some regions of Asia, though those same regions had very low rates of reported r*** in official statistics. Many variables (definitions of crimes, competence of police, confidence in judicial system, official fudging of numbers for political reasons, etc.) can influence crime statistics and make country comparisons difficult.
In light of these criticisms, I dug deeper for other data. I got this from
A Comparative Criminology Tour of the World by Dr. Robert Winslow of San Diego State University.
INCIDENCE OF CRIME
The crime rate in Sweden is high compared to other industrialized countries. An analysis was done using INTERPOL data for Sweden. For purpose of comparison, data were drawn for the seven offenses used to compute the United States FBI's index of crime. Index offenses include murder, forcible rape, robbery, aggravated assault, burglary, larceny, and motor vehicle theft. The combined total of these offenses constitutes the Index used for trend calculation purposes. Sweden will be compared with Japan (country with a low crime rate) and USA (country with a high crime rate). According to the INTERPOL data, for murder, the rate in 2001 was 10.01 per 100,000 population for Sweden, 1.10 for Japan, and 5.61 for USA. For rape, the rate in 2001 was 23.39 for Sweden, compared with 1.78 for Japan and 31.77 for USA. For robbery, the rate in 2001 was 95.83 for Sweden, 4.08 for Japan, and 148.50 for USA. For aggravated assault, the rate in 2001 was 667.42 for Sweden, 23.78 for Japan, and 318.55 for USA. For burglary, the rate in 2001 was 1323.90 for Sweden, 233.60 for Japan, and 740.80 for USA. The rate of larceny for 2001 was 6988.81 for Sweden, 1401.26 for Japan, and 2484.64 for USA. The rate for motor vehicle theft in 2001 was 495.21 for Sweden, compared with 44.28 for Japan and 430.64 for USA. The rate for all index offenses combined was 9604.57 for Sweden, compared with 1709.88 for Japan and 4160.51 for USA. (Note that Japan data are for year 2000)
The data reported to INTERPOL make it appear that Sweden is perhaps the most crime ridden country in the world; however, these findings should be tempered by comparison with data reported to the United Nations. In the UN reports, murders are referred to as "intentional homicides." Aggravated assaults are referred to as "major assaults," and larcenies are referred to as "thefts." According to the United Nations Sixth Annual Survey on Crime, crime recorded in police statistics shows the crime rate for the combined total of all Index crimes in Sweden to be 6981.48, per 100,000 inhabitants in 1997. This compares with 1345.94 for Japan (country with a low crime rate) and 4930.06 for USA (country with high crime rate). For intentional homicides, the rate in 1997 was 1.77 for Sweden, 0.54 for Japan, and 6.80 for USA. For major assaults, the rate in 1997 was 37.93 for Sweden, compared with 20.91 for Japan, and 382.31 for USA. For rapes, the rate in 1997 was 14.71 for Sweden, 1.31 for Japan, and 35.93 for USA. For robberies, the rate in 1997 was 75.04 for Sweden, 2.23 for Japan, and 186.27 for USA. For automobile theft, the rate in 1997 was 890.75 for Sweden, 213.49 for Japan, and 505.99 for USA. The rate of burglaries for 1997 was 1664.41 for Sweden, 175.81 for Japan, and 919.35 for USA. The rate for thefts in 1997 was 4296.87 for Sweden, compared with 931.65 for Japan and 2893.41 for USA. It should be observed that the above data reveal that comparatively speaking, Sweden has a low crime rate in regard to murder and major assault, a medium crime rate in regard to rape and robbery, and an exceedingly high rate in regard to property crimes (burglary, larceny, and auto theft). The discrepancies between the data reported to the United Nations for 1997 and those reported to INTERPOL for 2001 are partly explained by Sweden’s peculiar method of reporting murder and assault. Murders reported to INTERPOL included both attempted and completed acts of murder, while "major assaults" included both simple and aggravated assaults. These statistical anomalies for Sweden actually obscure the actual low to moderate rate of the most serious crimes (murder, aggravated assault, rape, and robbery), and this point is important. Sweden actually prioritizes its treatment of crime in the criminal justice system giving priority to the serious crimes, and often diverting property crimes to out of court settlement by the police or prosecutors, often in the form of "day-fines."
http://www-rohan.sdsu.edu/faculty/rwinslow/europe/sweden.htmlIt would appear that in 1997 and 2001, Sweden had fairly high crime rates with respect to robbery and theft, but low violent crime rates. Rape was higher proportionally than murder and assault, but still relatively low.
However, when looking at the trends, violent crime seemed to be rising:
Between 1995 and 2001, according to INTERPOL data, the rate of murder increased from 9.00 to 10.01 per 100,000 population, an increase of 11.2%. The rate for rape increased from 19.00 to 23.39, an increase of 23.1%. The rate of robbery increased from 65.00 to 95.83, an increase of 47.4%. The rate for aggravated assault increased from 616.00 to 667.42, an increase of 8.3%. The rate for burglary decreased from 1615.00 to 1323.90, a decrease of 18.0%. The rate of larceny increased from 5861.39 to 6988.81, an increase of 19.2%. (Note: larceny data are from 1996 – not reported for 1995) The rate of motor vehicle theft decreased from 659.00 to 495.21, a decrease of 24.9%. The rate of total index offenses increased from 2983.00 to 2615.76, an increase of 8.6%.
This was during the time period when crime of all kinds was decreasing in the U.S. This data is fairly old, but it does correspond to the trends that I posted in my graphs earlier. Crime is increasing in Sweden while it is decreasing in the U.S. during the late 90's. If those trends continued in the 2000's, then my graphs look accurate with respect to trends.
montestewart wrote:2) related to the above observation, I saw numerous references (including on Wikipedia and on State Department website) to changes in the way Sweden records crime that have created the appearance of a rise in crime, if the statistics are not properly assessed. For example, if three people jointly commit a homicide, apparently in Sweden that can now show up in statistics as three homicides. If a man r***s his wife every night, that is apparently now recorded as multiple, individual crimes In Sweden (whereas in the U.S., it apparently would be recorded as a single crime). This change in statistical reporting seems to have greatly contributed to the appearance of a rapid increase in reported r*** in recent years.
That wikipedia page is interesting. It reads as if it was written by someone with the sole goal of refuting allegations that Sweden's crime rate is rising. Yet, if you look carefully at the graph, it tells a different story:

The graph is visually deceptive because of the logarithmic scale. But if you look closely, you can see that assaults (yellow) are up from about 500 to about 900. Rapes (green) are up from about 90 to about 200. And robberies (pink) are up from about 70 to 90. Burglaries (blue) are down dramatically. Murders increased from 2.5 to 3.5, give or take.
A new system of crime reporting was implemented in 1990 which may indeed have caused a sudden increase in reported crime rates at the time, but there's no explanation as to how a one time change in 1990 would continue to cause a steady increase in reported violent crime for the next 20 years.