Misconception of foreign crime rate in japan

[Ewok85]

Is the OP playing down the criminality of visa violations?

No, they are saying that visa violations are a unique crime that can only be committed by non-Japaneses, and that you need to exclude them from the criminal statistics to create a reasonable comparison.

Apples and Oranges mate.

 

[diceke]

Yes, apples and oranges. The comparison of crime rates in the original article is not reasonable, to begin with.
A crime is a crime, nevertheless. Don't exclude that from the stats.

 

[diceke]

Has anyone actually tried calculating the stats on Bloomberg.com? I have no idea where some of these numbers come from, they don't quite match the stats on the National Police Agency website? Are they accurate?

Hmmm? Anyone?:?

Bloomberg - Are you a robot?

 

[Glenski]

Has anyone actually tried calculating the stats on Bloomberg.com? I have no idea where some of these numbers come from, they don't quite match the stats on the National Police Agency website? Are they accurate?
Hmmm? Anyone?:?

Bloomberg - Are you a robot?

Statistics in this country are manipulated enormously to suit the individual or government agency. I find it hard to believe much of it anywhere here.
That quote doesn't give enough material to show how bogus the numbers are, but let's give it a go with some assumptions.
In 2005, Japanese committed 1,776 crimes/100,000 people.
In 2005, non-Japanese committed 2,380 crimes/100,000 people.
If we use the 2003 figure that 46% of non-Japanese crimes were visa related and apply that to the 2005 figures, that means in 2005 non-Japanese committed 64% x 2,380 = 1,523 crimes/100,000 people. HEY! Fewer than the Japanese!
And, if we look at the last part of that quote, including or exluding visa offenses, foreigners' crime rate was lower than Japanese.
So, what's the beef about non-Japanese now?
People also like to throw out the rising number of crimes by foreigners, but they fail to realize that the number of foreigners is also growing. They should not count number of crimes, but a percent or rate of them.

 

[diceke]

In 2005, Japanese committed 1,776 crimes/100,000 people.
In 2005, non-Japanese committed 2,380 crimes/100,000 people.
If we use the 2003 figure that 46% of non-Japanese crimes were visa related and apply that to the 2005 figures, that means in 2005 non-Japanese committed 64% x 2,380 = 1,523 crimes/100,000 people. HEY! Fewer than the Japanese!

What I'm questioning is the accuracy of these specific numbers, like "2,380 crimes/100,000 people", "46% visa related crimes". Are they accurate? If so, what are the sources?
First, the article mentions the number of reported crimes per 100,000 people among 2 million "foreign residents", but then he's citing the number of crimes that lead to the arrests of "visiting aliens", which is a totally different breed of "aliens". The article is confusing the two, and it's misleading us.

visiting aliens in 2003
27258 penal law offenses, 13357 special law offenses (including 10550 immigration act offenses), 40615 total offenses
10550ﾂ≫ぎ40615=0.25975

"other" aliens in 2003
10277 penal law offenses, 1462 special law offenses (62 immigration act offenses), 11739 total offenses
62ﾂ≫ぎ11739=0.0005

aliens offenses total in 2003
37535 penal law offenses, 14819 special law offenses(10612 immigration act offenses), 52354 total offenses
10612ﾂ≫ぎ52357=0.20269

Therefore, the violations of the immigration act account for 20% of total offenses for all aliens, 26% for visiting aliens, 0.05% for "other" aliens. "46% visa related crimes" is overinflated. (Notice that not all immigration act offenses are related to visas, and it's also possible for Japanese nationals to violate this Act.)

 

[Ewok85]

Yes, apples and oranges. The comparison of crime rates in the original article is not reasonable, to begin with.
A crime is a crime, nevertheless. Don't exclude that from the stats.

A crime is not statistically relevant when the crime cannot be committed equally by both groups that are being compared.

Remember to read what you are quoting too - the statistics have crimes committed and people arrested - a single act can have multiple convictions and fall under several categories making it sound much worse.

 

[diceke]

Remember to read what you are quoting too - the statistics have crimes committed and people arrested - a single act can have multiple convictions and fall under several categories making it sound much worse.

Right. The number of people remanded for immigration offenses is 9,274, which indeed makes up 46% of all arrested/remanded visiting aliens.

Also, there are statistical differences between (1) the number of reported crimes, (2) the number of crimes that actually lead to arrest, and (3)the number of arrested people. So it seems like the Bloomberg article is guilty of mixing up all three, because you can't accurately infer (1) the number of reported crimes of 2 million "foreign residents" from either (2) or (3)of "visiting aliens".


The analysis in the OP is seriously flawed to begin with.

 

[Ewok85]

Its a common mistake with statistics, or just a plain manipulation. When I get time I'll get you the official crime stats (in Japanese though) and give a nice writeup

 

[diceke]

Its a common mistake with statistics, or just a plain manipulation. When I get time I'll get you the official crime stats (in Japanese though) and give a nice writeup

I tried to find out the crime "rates", but the "rates" are mostly rough estimates.

 

[Ewok85]

The national police website has accurate reported, arrest and convicted rates. Like I said, when I have time >_<