UPDATE: WUWT commenter P Wilson points out one single map that refutes this entire theory, see below the “read more” at the end of the post. – Anthony
To add to the Numberwatch big list of things supposedly caused by global warming, there’s now an oddball “irrefutable” (their words) story circulating around the net since Friday from Craig Anderson, a psychologist from Iowa State university known for video game violence studies, shown at left.
A Google News search reveals a number of news outlets picking this story up. The source for all these stories seems to be this one article in Newswise:
Researchers Present Study on How Global Climate Change Affects Violence
In that article, they cite it as:
Released: 3/19/2010 1:00 PM EDT
Source: Iowa State University
Problem is when you go to Iowa State to look for the source of the press release, it can’t be found. For example look at the Iowa State News site at: http://www.news.iastate.edu/ it is not listed on the page, nor if you look at the release page http://www.news.iastate.edu/releases/ page. Or do a search using their search engine.
On that search I found a vignette done apparently on Feb 26th, but no official Iowa State news release. Here’s the meat of the vignette, which looks like it was written for an internal newsletter:
He found that increases in average annual temperature or global warming, has an increasing effect on murders and assaults in this country, even after controlling for a variety of other factors.
“For every one degree increase in average temperature, we can expect an increase of 4.58 additional murders and assault cases per every 100,000 people,” Anderson said.
“There are obviously other factors involved,” he continued. “I would never claim that temperature alone would be the main factor that causes violent crime to be higher. However, there is now considerable evidence from a variety of sources that suggesting that high temperature is one cause that contributes to violent behavior, including violent criminal activity.”
Note to Anderson: correlation is not causation
Iowa State’s College of Liberal Arts and Sciences news page also does not list the story about Anderson’s claims on global warming driving increased violence. I did find a mention that Anderson has a paper in the journal Current Directions in Psychological Science but the latest 2010 edition is apparently not online.
It appears Anderson may have done his own press release, because I certainly haven’t been able to find any evidence that Iowa State official made any sort of news release of Friday March 19th, as cited by the “ground zero” Newswise story.
It is odd that Iowa State doesn’t have any official release. Maybe something will turn up Monday at the Iowa State News site.
In the meantime, his last offical news release on video games and violence gets a thorough drubbing from Techdirt:
===================================
from the except-for-the-details dept
excerpt:
So it seems a bit ridiculous for anyone — especially a professor who has been solidly on one side of the debate for many years, to stand up and claim that he has conclusively shown that violent video games make kids more “aggressive” (found via Slashdot). First, note the choice of words: not violent, but aggressive. Iowa State psychology professor Craig Anderson, who has already staked his reputation on saying that violent video games have a negative impact on kids, isn’t about to back down. He claims that he went through 130 studies and concluded that the support is unequivocal:
“We can now say with utmost confidence that regardless of research method — that is experimental, correlational, or longitudinal — and regardless of the cultures tested in this study [East and West], you get the same effects,” said Anderson, who is also director of Iowa State’s Center for the Study of Violence. “And the effects are that exposure to violent video games increases the likelihood of aggressive behavior in both short-term and long-term contexts. Such exposure also increases aggressive thinking and aggressive affect, and decreases prosocial behavior.”
Of course, reality is a bit more fuzzy. The same journal that is publishing Anderson’s new paper is also publishing a commentary from other researchers who disagree and suggest that Anderson has a pretty bad selection bias problem. But the biggest problem — as we noted above, is that all of these “violent video games are bad” studies seem to show incredibly weak effects that don’t appear to be significant in any meaningful way. As the commentary shows:
Psychology, too often, has lost its ability to put the weak (if any) effects found for VVGs on aggression into a proper perspective. In doing so, it does more to misinform than inform public debates on this issue.
Meanwhile, just last year, two Harvard Medical School professors also went through a whole bunch of different studies on violent video games and came to the exact opposite conclusions as Anderson did. It found little actual evidence to support Anderson’s claims, and found significant problems with research suggesting there was a serious link between violent video games and actual violence. Among that report’s findings:
- In the last 10 years, video games studies have been overwhelmingly popular compared to studies on other media.
- Less than half of studies (41%) used well validated aggression measures.
- Poorly standardized and unreliable measures of aggression tended to produce the highest effects, possibly because their unstandardized format allows researchers to pick and choose from a range of possible outcomes.
- The closer aggression measures got to actual violent behavior, the weaker the effects seen.
- Experimental studies produced much higher effects than correlational or longitudinal studies. As experimental studies were most likely to use aggression measures of poor quality, this may be the reason why.
- There was no evidence that video games produce higher effects than other media, despite their interactive nature.
- Overall, effects were negligible, and we conclude that media violence generally has little demonstrable effect on aggressive behavior.
Oh, and I almost forgot to mention. Dr. Anderson has apparently embraced a whole new type of science called “Human Thermodynamics”. Here’s an encyclopedia cover at the EoHT Wiki of which he is a member:
There’s even has an equation to quantify the human thermodynamic effect, nicely presented in a non-violent manner. From the EoHT Wiki main page:
Tattoo (or inking) of the Clausius inequality; photo by Marco Fantoni (March, 2008); an example of art thermodynamics. In the photo, showing a hand holding both a new and burnt match, “the hand represents the capacity of the human mind to analyze and understand natural phenomena [such as] the power and imperative of irreversibility.” [3]
He found that increases in average annual temperature or global warming, has an increasing effect on murders and assaults in this country, even after controlling for a variety of other factors.
“For every one degree increase in average temperature, we can expect an increase of 4.58 additional murders and assault cases per every 100,000 people,” Anderson said.
“There are obviously other factors involved,” he continued. “I would never claim that temperature alone would be the main factor that causes violent crime to be higher. However, there is now considerable evidence from a variety of sources that suggesting that high temperature is one cause that contributes to violent behavior, including violent criminal activity.”
UPDATE: WUWT commenter P Wilson shares this map circa 2009 and asks:
What does it show? Rather than Austrialia havin inexorable crime rate, the highest crime rates seem to be in relatively cool countries.
WUWT?
Indeed, according to the map, the top ten countries for crime are:
1. Iceland
2. Sweden
3. New Zealand
4. Grenada
5. Norway
6. England and Wales
7. Denmark
8. Finland
9. Scotland
10. Canada
With the exception of Grenada, all are cooler climate countries. So much for Dr. Anderson’s theory of heat in the form of AGW = crime.
Maybe that’s why Iowa State never published a press release, they were just too embarrassed to do so.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Steve Goddard (16:59:06) :
You will find that this was addressed quite well by Stephen Levitt (best known for his publications and blog called “Freakonomics”). Levitt demonstrated that crime statistics correlated very well with the introduction of abortion legislation in the USA*. I believe there was some debate on his use of statistics in that particular paper, but upon later review the discrepancies did not affect the conclusion.
In a nutshell crime statistics dropped off in each of the states of the USA according to when the Roe vs. Wade legislation was brought into effect (it is a generationally lagged effect). This is by no means an argument for or against the legislation, but simply pointing out the objective fact that this lagged relationship occurs. So folks, please don’t go O/T and start a flame war on the issue 🙂
It also begs the question what is going to happen to Australian crime rates in about a decade as a result of the $5,000 baby bonus the previous government introduced – in effect a reverse of the roe vs. Wade legislation, as it increased birth rates. One has to sit back and objectively muse which people would alter their procreative choices based upon a $5,000 government hand out, and then ponder the long-term effects of such a policy. But I digress O/T…
* http://papers.ssrn.com/sol3/papers.cfm?abstract_id=174508
Robert E. Phelan (20:19:56) :
If crime rates are, in fact, rates such as murders per 100,000 population, then they would not necessarily be related to the size of the population, no? Unless one were to argue the fact that increased population densities leads to higher crime rates … hmmm, UHI relates to population density… no, just not going there LOL 😀
Scratch that about Buffalo being low. I sorted badly. Houston still isn’t good though.
St Louis is surprisingly large. I wouldn’t have picked that one.
Hotter climates foment aggression? How many times have armies perished in large numbers in the bitter cold? Napoleon, Hitler’s Russian voyage? How many times has there been an equivalent in hot regions? Lawrence of Arabia?
The map doesn’t prove anything because it’s talking about crimes *reported*, and not all of those crimes are violent. In many high crime countries, crimes occur but just aren’t reported because the population doesn’t feel like the authorities will do anything about it. If you look at the places where people report the most crime, it has a lot to do with how people view the police in that country.
Another thing it has to do with is the fact that in many of the Scandinavian countries as well as in the UK, people insure a lot more of their belongings (laptops, cameras, bicycles, etc), much more so than we do in the USA. In order to claim that insurance, you need to report the crime. I’m willing to wager that bicycle theft (usually not a violent crime) probably pushes some of those Scandinavian countries much higher than they normally would be!
Anyway, regarding the connection to global warming…. not so sure about that one. But crimes reported isn’t such a good statistic, maybe you want to look at violent crime rates, because that’s what the guy’s theory is about.
Bulldust (21:51:09) :
Yeah, I’ve loaned my copy out, but from memory Levitt titles one of his chapters like “Why drug dealers live with their mothers”.
I wonder if this sort of thing now gets reported as an assault?
Headline: “Yakima kindergartner expelled for making a gun with hands”
http://www.kndu.com/global/story.asp?s=11979866
Mooloo (21:50:06) :
I seem to recall Michael Moore (now shoot me because this is a dubious reference, as I am the first to admit) saying that Canadian gun ownership per capita is higher than in the USA, but the murder rate is far lower. These would be interesting stats to verify.
Also interestingly, if morbidly, Australia has the dubious distinction of beating the USA in the number killed by a single gunman in a killing spree, this despite our relatively low gun ownership and murder rates. Go figure…
http://en.wikipedia.org/wiki/Port_Arthur_massacre_(Australia)
Bulldust (21:54:56) :
Unless one were to argue the fact that increased population densities leads to higher crime rates …
Yeah, Bulldust, in fact they are. Urban crime rates are always higher than rural crime rates.
Geoff Sherrington (21:59:58) :
I sympathise – my copy of Freakonomics is out on loan and I get the feeling it isn’t coming back. I must stop loaning out my text books. Maybe if I owned a gun the books would come back…
The question is…will Al Gore…blame the crazy antics of a few University of Arkansas students in in erecting [arww bad choice of words] a snow-p****s in a spring snowstorm in the good ole’ south…on Global Warming?
Probably.
http://www.uark.edu/home/11136.php
You have to pan around with the webcam a bit….but it is there. Focus on the center of the pan….and you will find it. Heh you can’t miss it.
Hilarious! Quick…”before I melt away.” [To borrow words from Frosty the Snowman].
Chris
Norfolk, VA, USA
Bulldust (21:54:56) :
Sorry, I don’t mean to be either abrupt or cryptic even when in fact I am. You are right about rates, they are generally measured in incidents per thousand or hundred thousand. If you check the FBI’s Unifrom Crime Reports which are available here:
http://www.fbi.gov/ucr/ucr.htm
you’ll see that crime rates differ: the cities have higher rates than rural and suburban areas, blacks and hispanics have higher rates of both crime and victimization than whites, men are both the perpetrators and victims of violent crime more often than women (oddly enough, men murder men far more often than women murder women…. in fact, 90% of the victims of female murderers are men). The vast majority of street crime (crimes counted by the UCR) are committed by young men and the vast majority of these crimes appear to be committed by a fairly small cadre. Population Density does seem to have an effect, but the link to UHI would be spurious at best…
Ando, you do not have the remotest idea what you are writing about. Europe under-reports crime by an incredible factor. In Britain you are likely to be arrested if you report a crime by a Muslim or ‘black’ against you. France no longer keeps statistics on minority versus minority crime. There are many places in Europe that no longer keep statistics on real property crime. In fact trespass has been written off the British books. Norway now has a rape rate many times higher than that of America. So they deal with it by not cataloging it. Sweden has a crime rate in excess of South Africa or Brazil in at least three districts. Please.
Robert E. Phelan (22:20:33) :
Bulldust (21:54:56) :
Unless one were to argue the fact that increased population densities leads to higher crime rates …
Yeah, Bulldust, in fact they are. Urban crime rates are always higher than rural crime rates.
~~~~~~~~~~~~~~~~
So this must be as a result of the ebil UHI! (Urban Hate Island?)
So thats what the Hockey stick is for,i get it now.
You can get a good swing out of a hockey stick.
so there’s a correlation between temperature rise and violent behavour?
hmmmm,
Global warming is caused by violence!
Pass the Nobel please
/Mango
“davidmhoffer (18:59:03) :
Now wait a second.
First he says that video games cause more violence.
Then he says that global warming causes more violence.
Did he subtract the video game contribution from his data? Has it not increased exponentially in the last 50 years just like global warming?”
I think he uses one dataset for two papers; using two different hypothesises. Probably he’ll come up with more.
@Mooloo (21:50:06) :
As I said: …this can all be proven in any direction you like with statistical manipulation.
I’m sorry, even as a long time reader of this blog who enjoys reading the articles, i can’t take seriously the map that P Wilson has put up as evidence refuting this theory of higher violence in warmer countries etc.
you only have to start looking at other pieces of crime data that are violent (rather than all crime data), say murder rates instead of total reported crime rates to see that the map is total rubbish.
(http://en.wikipedia.org/wiki/List_of_countries_by_intentional_homicide_rate)
While i agree that the theory that global warming causes violence is probably nonsense, using that map, and trying to justify that norway/iceland etc is a more dangerous place than south Africa or iraq, makes this whole site look very amateurish. With your map you are comparing apples with oranges.
ot,
Saatchi & saatchi founded by Maurice and Charles saatchi is a world class advertising agency, the saatchi brothers now own M&C Saatchi another world class agency.
Question to Charles Saatchi,
How do you feel about Global Warming?,
Answer, Global warming is obvious nonsense but it makes nice people feel good about themselves to do their bit for the planet. It`s vanity of the grotesque kind to believe that mankind and our carbon footprint has more impact of the future of earth than nature, which bends our planet to it`s will, as it sees fit.
This would be just a harmless fad, exept it is now costing the west trillions of dollers to go greener with the blessing of the caring souls and the Kyoto protocol, while much of the world sill lives in grinding poverty or simply staves to death.
Dail Mail 21/3/2010.
” 4.58 additional murders and assault cases ” – This is how the media normally phrase such things but what they mean is 4.48 assault cases & 0.1 murders”. Conflating the 2 & putting murders first is simply scaremongering which is what the media & much of government are there for.
In America every day thare are 12 million cases of human cannibalism & eating burgers.
Many years ago a type of rotary engine was known as a Lenin Trotsky engine,
or two cranks per revolution, maybe an update to a Gore Anderson would be
timely anyone agree??
Neil Craig (04:33:45) :
” 4.58 additional murders and assault cases ” – This is how the media normally phrase such things but what they mean is 4.48 assault cases & 0.1 murders”. Conflating the 2 & putting murders first is simply scaremongering which is what the media & much of government are there for.
….
Aggregating perormance measures is very interesting. I have often considered why casualties in battle were always ‘killed and wounded’ when there is a huge distinction. In the British MSM now we rarely get any wounded stats. I believe it’s because the ratio od wounded to dead is much higher than previous wars [body armour, med-evac, toepopper mines etc, small calibre rounds etc]. 1 dead doesn’t sound as bad say as 21 ‘dead and wounded’. Whereas a similar action in WW2 may have had 6 dead so adding in the 15 wounded dilutes the impact of the stats.
cheers David
“There are obviously other factors involved”
Yes. Like unemployment and poverty, to name a couple. If we try to stop global warming, based on the the ridiculous notion that it is caused by the burning of fossil fuels, there will be a hell of a lot more of both. So, even if Professor Anderson is correct, it is doubtful that any attempt to stop global warming will lead to a net reduction in violent crime.
Regards,
Trevor
Phil B (04:08:47) :
You’re quite right. The map is nonsense and refutes nothing. The fact that Iceland MAY have the world’s highest crime rate says nothing about crimes of violence. One source here
http://www.nationmaster.com/graph/cri_mur_percap-crime-murders-per-capita
ranks Iceland at number 42 for murders, #12 for assaults and #10 for rapes. The United States, which did not make the top ten on Wilson’s map, ranks 24th, 10th and 9th respectively. Wilson’s map compares apples to potatoes.
There is in fact a correlation between crime and temperature and crime and seasonality that has been the subject of analysis and debate for at least a hundred years. The following link provides a quick summary of the issue as viewed by sociologists and criminologists:
http://www.odum.unc.edu/odum/content/pdf/Bollen%20Hipp%20Bauer%20Curran%20Bollen%202004%20SocForces.pdf
Dr. Anderson goes off the rails when he links the issue to Global Warming…. but then, perhaps it just shows that any field can get better funding when it can be linked to global warming.