Richard Florida and Charlotta Mellander of the Atlantic took a crack at examining the statistical relationship between gun related deaths and several commonly associated causes.
Firearm-related deaths were positively associated with states that voted for McCain (.66) and negatively associated with states that voted for Obama (-.66).
[.] fatal gun violence is less likely to occur in richer states with more post-industrial knowledge economies, higher levels of college graduates, and tighter gun laws. Factors like drug use, stress levels, and mental illness are much less significant than might be assumed.
The authors correctly point out the difference between correlation and causation, and I have problems with some of the indicators used, but this analysis is one step closer to reality than most of the other crap articles floating around our news media lately on this topic. As usual with politically fiery articles like this, the comments are an entertaining read. Thanks to Tom Dawkins for the link!
Comment Form
What is Chart Porn?
An addictive collection of beautiful charts, graphs, maps, and interactive data visualization toys -- on topics from around the world.
Categories
- Bailout (118)
- Chartporn Related (3)
- Commentary (21)
- Culture (669)
- Emerging Markets (66)
- Employment (245)
- Environment/weather (133)
- Finance (298)
- Food (92)
- Global Economy (373)
- Graphic Design (bad) (26)
- Graphic Design (general) (183)
- Graphic Tools (23)
- History (158)
- Housing (162)
- Humor (204)
- Innovative (183)
- Interactive (545)
- Internet/tech (97)
- Maps (578)
- News Media (34)
- Politics (329)
- Reference (97)
- Science (331)
- Source: Economist (101)
- Source: FT (92)
- Source: NYT (147)
- Source: Ritholtz (76)
- Source: USA Today (27)
- Source: Washington Post (90)
- Source: WSJ (135)
- Sports (58)
- Stock Market (74)
- Uncategorized (2)
- Updated regularly (76)
- US Economy (553)
- Video (22)
- Aram Korevaar: This chart is now being used as a projection in which countries such as China see themselves as in a [...]
- David: Welcome back Chart Porn! [...]
- J S: Thanks for the great story. Miss reading this blog. Hope to see you more active again. [...]
- jake: I lived in a DC row house for 6 years, and I'm writing this comment from my tiny 1 bedroom apartment [...]
- ronny pettersen: Hilarious and unfortunately accurate... ;-) [...]
4 Responses to What Causes Gun Deaths?
Matt
January 17th, 2011 at 14:09
Ah, so voting for Obama will lead to less gun violence. Too bad the counties Obama won have twice as many murders as the ones McCain won:
“The murder rate for counties carried by Obama was 6.56 per 100,000 inhabitants, less than half the rate claimed in the message. The rate for counties carried by McCain was 3.60 per 100,000, much higher than claimed in the message.”
http://www.factcheck.org/2009/01/unreported-stats/
Perhaps they’re using knives? Nice chart, though!
JohnF
January 19th, 2011 at 11:17
What is also interesting is that the correlation coefficient of voting for McCain and of voting for Obama is identical apart from the sign. This indicates that when doing their regression analysis, they probably used “1” for McCain and “-1” for Obama: duh. If they had regressed according to the % of votes for McCain and % of votes for Obama, they’d not have ended up with that identical coefficient. Seriously, it’s extremely unlikely.
Given what Matt says (6.56 murders per Obama county, 3.60 murders per McCain county), the results are even less believable.
On the up side, at least they were *trying* to do the right thing. A step in the right direction!
Shan
January 20th, 2011 at 15:40
Matt, I think you’re misreading the statistics:
“Note that these figures include accidental shootings, suicides, even acts of self-defense, as well as crimes.”
It’s not just murders, it’s any kind of gun-related death. It’s really getting to the core of gun violence and not just murder rates.
Matt
January 25th, 2011 at 15:56
Shan, really? You really think the sole purpose of the article was to find random correlations between “gun violence” and various demographics? Or was it to make a political point. Well, the political point I’m making is that murders occur more frequently in counties that voted for Obama. Or it’s a “random” correlation. Tick your pick. The fact that the “study” left out the murder correlation makes it biased against McCain voters, and in favor of the average Atlantic reader who probably voted for Obama.
Side note: I really don’t know why self defense and suicide is lumped into “gun violence”. The first, it’s either you or a cop defending you. Seems like a good thing if defense actually works (e.g. someone is coming at you with a knife, you can shoot him with a gun). The second, people who commit suicide have plenty of options. We should really only be concerned about gun violence against innocents. Of course, we also care about suicide, but guns aren’t the cause.