GP Interviews Michigan Prof on Media Violence Study Confusion

December 7, 2007 -
In late November GamePolitics and other media outlets reported on a study published in the Journal of Adolescent Health which indicated a strong linkage between violent media and aggressive behavior in children.

The study was authored by L. Rowell Huesmann (left) and Brad Bushman, both professors at the University of Michigan.

A comparison of public health threats included in the report generated a fair amount of media attention and was widely misinterpreted (including here, initially). Here is an example, taken from the Reuters report:
Violence depicted on television, in films and video games raises the risk of aggressive behavior in adults and young viewers and poses a serious threat to public health, according to a new study.

After reviewing more than 50 years of research on the impact of violence in the media, L. Rowell Huesmann, of the University of Michigan, and his colleague Brad Bushman concluded that only smoking posed a greater danger.

Only smoking is a greater risk to public health than video games?

The more we thought about that, the less sense it made. So we went right to the source. Study author L. Rowell Huesmann is an Amos N. Tversky Collegiate Professor of Communication Studies & Psychology as well as the Director of the Research Center for Group Dynamics Institute for Social Research at the University of Michigan.

GP: Your study has been widely reported in mainstream and video game enthusiast sites as saying that violent media is the number two public health threat behind smoking. However, what I read your study to say is that the correlation between media violence and aggression is second only to that between smoking and lung cancer among [select] public health threats [those listed by Huesmann]. Am I correct about this?

LRH: Yes, you are correct. And furthermore, there may be other public health threats that are also greater that I don't know about. The real point is that the effect size is almost as big as smoking on lung cancer and bigger than a number of other threats that many people consider serious. This means the effect size is big enough that we should all really be concerned about the effects of violent video games on children.

GP: Also - among the threats listed I don't see [major] things like obesity & diabetes, guns & homicide, drinking & car crashes, etc.  Could you address why?

LRH: No reason. I picked ones I thought people would be surprised about. In fact I would guess that the ones you mention would be even larger. Let me know if you figure them out, please.

GP: Finally, what is the broader implication of the increased aggression you have observed? Juvenile crime rates have been generally declining over the last decade or so, have they not? With the amount of violent media out there, wouldn't you expect an opposite trend?

LRH: No, because there are so many factors that affect juvenile crime rates. For example, one of the biggest factors historically in crime rates is the state of the economy. For example, one of the highest murder rates in the past 120 years was during the Depression in the 1930's. Another important factor is the demographic composition of the population, e.g. - exactly how many youth are in a particular age range. Another important factor is the number of police and how crimes are reported. So, generally, one can't use gross population crime rates to infer much of anything at the individual level.

Please be aware that an implication of the body of research I reviewed is that the mass media and video games are very powerful teachers. Video games can teach wonderful things or they can teach bad things. It would be a shame if my review were used to indict all video games. There are many great games that teach kids very valuable things.

GP: Thanks to Dr. Huesmann for taking the time to respond and clear up the confusion.


Talking about confusion : don't forget how the LaRouchites covered this study, by saying that "violent video games were worse than AIDS" :

Well he seems like a fairly pleasant fellow.

Oh, and I'd add one more thing : I'm very glad Huesmann says CLEARLY that this is not an attack against all video games.

"LRH: No reason. I picked ones I thought people would be surprised about. In fact I would guess that the ones you mention would be even larger. Let me know if you figure them out, please."

0_o .....

@ Soldatlouis

Well, some special interest groups just need to be honest "enough". All you really need to say is "'y' caused 'x' amount of people to die"- there's no need to say "5 others things caused more deaths than 'y'", is there? When you withhold information that you expect few people to look through, you can say alot of things and not be caught "totally lying".

I'm kind of surprised that he would freely admit that the study was intended to be sensationalist.

What the hell is this? That piece of crap just admitted to his study being full of shit basically and it sounds as if he feels he did nothing wrong by essentially lying to the public. There were far more sensible and obvious things to list which was proven by GP's list. That admission alone removes any and all credibility he may have had.

Good interview.

That was the big response for me, too. The comment he made about wanting to surprise people, sounds like another way to say "sensationalism" to me.

NW2K Software
Nightwng2000 NW2K Software Nightwng2000 is now admin to the group "Parents For Education, Not Legislation" on MySpace as

We should round up all the gamers who're also smokers and see which they develop first, lung cancer or a personality disorder. That'd settle the controversy.

"No, because there are so many factors that affect juvenile crime rates."

Remember, morons: Correlation is not Causation. Junk Science FTW.

From the journal article itself: “Our response should be… to avoid exaggerating the dangers (which would destroy our credibility)”.

I teach a second year research methods in psychology course. A week before this study came out, we discussed how slope and correlation are related but not the same. In other words, given a 0.3 correlation between media violence and aggression, we don't have enough information to determine if that means that increased exposure increases the chance of becoming aggressive from 2% to 4%, or from 4% to 10%, or any other amount. It was amusing to read such a misuse of statistics in a peer-reviewed journal, especially since my class found the concept rather straightforward.

I'm still pretty confused about the significance of the threat part. Your examples and the Dr.'s example involve inputs that results in quantitatively negative consequences (smoking:lung cancer, drinking:car crashes, and obesity:diabetes), but the linkage of violence in the media:aggression doesn't conclude negative behavior... it assumes it.

So, what's the point. I understand he's saying that the interaction can be a great teacher, but he frames it as this ticking time bomb...

He even explains that violence (which would be the negative consequence of aggression) would have to include a number of potential attributes including a potentially recessed economy, higher youth numbers in a given range, and (based on this study) over-exposure to violent media.

All-in-all, this doesn't teach me much more than we could have inferred ourselves, but instead sensationalizes the topic for others to misuse the data...

Don't get me wrong, because the article I think is good if interpreted correctly. The author explicitly states the intended purpose... that we should use this as evidence to support the use of media as a tool for learning and understand the impact of the media distributed.

Wait... did he just basicly admit that the only reason video games came in at #2 on the list is because he didn't feel like including any of the other issues that are worse except smoking? wow....

*sigh* one of the first things they teach you in psych/soc is that 99% of the studies and research done in the field are essentially useless and bias.

I think that for some of the claims he made, he should have been much more careful about his choice of wording in the original release. This strikes me as very nearly the same exact tactic as the Eidos Kane 7 Lynch "reviews" after the Gerstmann debacle. It's so brazen it just leaves me speechless.

I'm really surprised that a Ph.D. would play so fast and loose with issues of correlation and causation. He readily admits that his study only examines correlation between violent games and aggression but then turns around and talks about the "effects" of the games. Unfortunately for him, he can't prove anything about the *effects* of media violence, only that is has *some relationship* with aggression.

I'll bet the correlation between age and wealth is extremely large as well; that doesn't mean that being old makes you rich! Age is not a "cause" and wealth is not an "effect," but there is a relationship between the two.

You can prove anything with statistics. 37% of people know that.

"It would be a shame if my review were used to indict all video games."

Of course it would. The study should instead be used to indict the kinds of video games the researcher believes have a negative influence. Of course, that tells us exactly nothing about whether the games being criticized are worthy of such criticism.

Huesmann and Bushman set out to show that violent video games have negative effects, and -- sure enough -- they "found" them.

They didn't find that video games have negative effects. They found that exposure to specific types of content (violence) increased aggression. They are begging the question that exposure to violence increases aggression, so will that lead to increased violence.

That's where this study could be used inappropriately.

Well first off, I'm not actually surprised by his revelations that this is a "sensationalist” study. It just goes to show anyone will do anything to get their name in the news now a days.

Secondly, was he thinking?

"They found that exposure to specific types of content (violence) increased aggression."

I'd call that a negative effect, at least according to the standard by which these studies are usually interpreted.

By which it will be used, more than likely, yea. But aggressive behavior is expressed in business, sports, law, and basically any huge number of actions.

We have the same conclusion though... this will be used inappropriately (more than likely).

It would be a shame if my review were used to indict all video games.

Unfortunately, that is exactly what happens. It happens with every "study" on video game violence. The things are taken out of context by the people touting them. But it is not only taken out of context by those people, but als othe people performing the studies, as is apparent by this guy's comments. He could have used other dangerous things, but didn't because they weren't sensational enough or would have lowered the sensationalism of the video game "effects."

The only problem is interactive media desensitizes you to itself, leaving normal media like TV and films to be doubly absorbed.

Even if you disregard what the majority think, violent media has done lil if anything to brainwash the public, once you take into fact thos in the higher echlions of society have always felt the common man did not understand his media (theaters/plays, books,song,music,radio,TV,comics,film,video games) , its merely part of human society that thos who think they know better push their ways on thos lower in the social structure.

Dose this mean we should let kids gnaw on mature mediums....yes and no...parents play a key role in training their kids for life do you throw them in jail for teaching their kids about wine or alcoholic drinks or porn if they are merely 13-20, where dose "trendy" social influence stop and parental parenting begin?

Of coarse we should not let kids get mature stuff directly
but there coems a time when parents however good or bad in society, society itself must let them parent.

Maybe violent people are predisposed to like violence regardless of the medium. Maybe these killers were already killers; it's no surprise that one who enjoys the act of killing might enjoy violent games.

I hate the role of junk science in America today. So many laws are getting passed on bullshit. Smoking bans, for one. So much of the science around smoking and the effects thereof is absolute junk. How can we decide that cigarette smoke causes cancer, when we don't really know what causes cancer. Modern science has NO explanation for cancer. They have corrolary evidence to use when embracing "risk factors," but they're not even sure why these risk factors contribute to malignant cell mutation.

The smoking bans are simply cases of a malicious majority using junk science to impose its will on the minority. And this is probably what's going to happen with games.

@ TheStripe

Spoken like a true chain smoker. ;)

The science behind smoking and cancer is not "junk science" It is heavily tested, studied and peer reviewed. They have tested teh chemicals in cigarette smoke on human and animal tissue and have documented the effects. It has been proven to be a strong contributer to cancer. Not everyone that smokes or who breaths secondhand smoke will develop cancer but enough do for it to be a valid risk.

E. Zachary Knight
Ya but it dose not automatically remove ones right to smoke and it dose not give a city the power to remove it from private establishments,hell the states are gettign in on the bans and thats a fcking crime.

2nd hand smoke outside or in a vented/filtered area is no worse than fcking car exsaghted off the street or grarges...

the rush to damnation is paved with golden thoughts........

See, see, see he said he PICKED things he thought would SURPRISE people.

GP: Also - among the threats listed I don’t see [major] things like obesity & diabetes, guns & homicide, drinking & car crashes, etc. Could you address why?

LRH: No reason. I picked ones I thought people would be surprised about. In fact I would guess that the ones you mention would be even larger. Let me know if you figure them out, please.

Wouldn't that statement right there invalidate his whole study if he is targeting one specific thing and not comparing it to others as mentioned? I personally would expect more guns and bullets to kill people than violent media.

I like what Simon said, "We should round up all the gamers who’re also smokers and see which they develop first, lung cancer or a personality disorder. That’d settle the controversy."

I'd volunteer, but I just quick smoking a month ago. Chantix is the best drug ever.

from the interview:
"So, generally, one can’t use gross population crime rates to infer much of anything at the individual level."

but he just got finished saying how the effect size was so big, like smoking to lung cancer. Certainly if the effect size is so large, we'd see an effect in gross population rates (crime or otherwise). there's plenty of gross population statistics tying smoking to lung cancer.

crime rates are dropping while the games industry sees record-breaking growth. if there really is a conection, there should be at lest a 'bump' in the charts.

someone's lining this Dbag's pockets... i fucking hate fucking shills.

So what he is saying is that "Violent media is second only to smoking as a threat to society, except for a lot of things that I either didn't know about or choose not to include in my paper as the ones I listed were more surprising."

Also, when asked if we should be seeing more crime because of widespread violent media:
"No, because there are so many factors that affect juvenile crime rates. For example, one of the biggest factors historically in crime rates is the state of the economy."

Now, correct me if I'm wrong, but isn't this what we have been saying all along? The causes of crime have not changed, all of the old staples are still present; poverty, addiction, racism, good old fashioned craziness and sociopathy. The sad thing is though, it's much less work to scapegoat emerging media forms for societies problems than it is to actually suggest and carry out work that would reduce and alleviate the problems that ACTUALLY cause crime. (example: The Sony vs. Cathedral debacle over Resistance Fall of Man. The church decided to chastise Sony instead of being outraged over the city's gun problem. Sony has an address you can write to and bombard with silly complaints letters, while working with a community to try and oust or crack down on dealers, thieves and thugs takes effort!)

"exactly how many youth are in a particular age range"

Isn't 'youth' a particular age range?

[...] 15. Game Politics  This article contains an interview with L. Rowell Huesmann and Brad Bushman on the linking between media violence and aggressive behavior with children.  L. Rowell Huesmann states at the end of his interview that their are many factors that lead to aggressive behavior in children, not just violence in the media. [...]
Forgot your password?
Username :
Password :


Which group is more ethically challenged?:

Shout box

You're not permitted to post shouts.
Matthew WilsonI am old school on this. I believe its a conflict of interest to have public sector unions. that being said, I do not have a positive look on unions in general.07/07/2015 - 3:59pm
TechnogeekWhat's best for the employee tends to be good for the employer; other way around, not so much. So long as that's the case, there's going to be a far stronger incentive for management to behave in such a way that invites retalitation than for the union to.07/07/2015 - 3:10pm
TechnogeekTeachers' unions? State legislatures. UAW? Just look at GM's middle management.07/07/2015 - 3:05pm
TechnogeekIn many ways it seems that the worse a union tends to behave, the worse that the company's management has behaved in the past.07/07/2015 - 3:02pm
james_fudgeCharity starts at home ;)07/07/2015 - 2:49pm
james_fudgeSo mandatory charity? That sounds shitty to me07/07/2015 - 2:49pm
E. Zachary KnightGoth, if Union dues are automatically withdrawn, then there is no such thing as a non-union employee.07/07/2015 - 2:38pm
Goth_Skunka mutually agreed upon charity instead.07/07/2015 - 2:33pm
Goth_Skunkyou enjoy the benefits of working in a union environment. If working in a union is against your religious beliefs or just something you wholeheartedly object to, dues will still be deducted from your pay, but you can instruct that they be directed towards07/07/2015 - 2:33pm
Goth_SkunkBasically, if you are employed in a business where employees are represented by a union for the purposes of collective bargaining, whether or not you are a union member, you will have union dues deducted from your pay, since regardless of membership,07/07/2015 - 2:32pm
Goth_SkunkIt's something that has existed in Canada since 1946. You can read more on it here: - 2:27pm
Goth_SkunkSee, we have something similar in Canada, called a "Rand Employee." This is an employee who benefits from the collective bargaining efforts of a union, despite not wanting to be a part of it for whatever reason.07/07/2015 - 2:22pm
Matthew Wilson@info depends on the sector. for example, have you looked at how powerful unions are in the public sector? I will make the argument they have too much power in that sector.07/07/2015 - 12:39pm
InfophileIt's easy to worry about unions having too much power and causing harm. The odd thing is, why do people seem to worry about that more than the fact that business-owners can have too much power and do harm, particularly at a time when unions have no power?07/07/2015 - 12:31pm
Matthew Wilsonthe thing is unions earned their bad reputation in the US. the way unions oparate the better at your job you are, the likely you want to be in a union.07/07/2015 - 11:33am
InfophilePut that way, "right to work" seems to have BLEEP-all to do with gay rights. Thing is, union-negotiated contracts used to be one of the key ways to prevent employers from firing at will. Without union protection, nothing stops at-will firing.07/07/2015 - 11:06am
Infophilehas an incentive to pay dues if they're represented either way, so the union is starved for funds and dies, unless things are bad enough that people will pay dues anyway.07/07/2015 - 11:02am
InfophileFor those who don't know, "right to work" laws mean that it can't be a condition of an employment contract that you pay union dues. That is, the right to work without having to pay dues. Catch is, unions have to represent non-members as well, so no one...07/07/2015 - 11:01am
MechaCrashUnexpected? Seriously?07/07/2015 - 10:55am
Mattsworknamejob they wanted without the unions getting involved. The problem is, it has some unexpected side effects, like the ones Info mentioned07/07/2015 - 8:49am

Be Heard - Contact Your Politician