Debunking the '3 times more likely to be the victim' myth -- reprise

[X_Digger]

Figured I'd drag this one from the archives, original discussion here- http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=221485

We've often heard the statement that just having a gun in a house makes one 2.7 times more likely to be the victim of gun violence.

The root of this statistic is a 1993 study published in the New England Journal of Medicine by Arthur Kellermann et al

http://content.nejm.org/cgi/content/full/329/15/1084

The conclusion reads in part:

Despite the widely held belief that guns are effective for protection, our results suggest that they actually pose a substantial threat to members of the household. People who keep guns in their homes appear to be at greater risk of homicide in the home than people who do not. Most of this risk is due to a substantially greater risk of homicide at the hands of a family member or intimate acquaintance.

Here are the problems that many have noted-

The authors note that "One or more guns were reportedly kept in 45.4 percent of the homes of the case subjects," This implies that no guns were kept in 54.6% of the households. No study was made of how many were killed by guns kept in the home versus those brought in by a perpetrator.

Five years later, Kellerman revised himself (quoting http://www.guncite.com/gun-control-kellermann-3times.html ):

A subsequent study, again by Kellermann, of fatal and non-fatal gunshot woundings, showed that only 14.2% of the shootings involving a gun whose origins were known, involved a gun kept in the home where the shooting occurred. (Kellermann, et. al. 1998. "Injuries and deaths due to firearms in the home." Journal of Trauma 45:263-267) ("The authors reported that among those 438 assaultive gunshot woundings, 49 involved a gun 'kept in the home where the shooting occurred,' 295 involved a gun brought to the scene from elsewhere, and another 94 involved a gun whose origins were not noted by the police.") (Kleck, Gary. "Can Owning a Gun Really Triple the Owner's Chances of Being Murdered?" Homicide Studies 5 <2001>.)

Secondly, no correlation was made between "independent" factors that actually may have been factors related to each other- they treated illicit drug use, having an arrest record, living alone or not, renting, having a gun, and a history of domestic abuse as independent variables without any relationship to each other. No collateral multivariate analysis was performed. The correlation to each control was not predicated on other factors, just gun ownership. They gave the same weight to a gun death in a household with someone with a previous arrest as to a gun death in a household where an intruder brought their own gun to a home invasion and shot the occupant (each weighting was independent, not cumulative). No correlation was explored for similar situations with the only difference being gun ownership.

Thirdly, there were significant differences between the study participants and the control. There was a 30% difference between home ownership vs renting between subjects and control, and a 15% difference in living alone or not. Only 48% of the control subjects were interviewed in person. Never mind that there were more users of illicit drugs, alcoholics, and persons with a history of violence in the households of the case subjects than in the households of the controls.

Finally, correlation doesn't equate to causation. They state in one place, "keeping a gun in the home was strongly and independently associated with an increased risk of homicide". "Associated with", not "causally related to". The possibility of why a gun was kept in the home was not explored nor accounted for- so a person who lives in a high crime neighborhood who may already be at higher risk of homicide death was treated the same as a person shot in a "nice" neighborhood.

Further reading (some are related to Kellermann's previous work on the subject, just to show how tortured his conclusions are):

http://www.nationalreview.com/kopel/kopel013101.shtml

http://guncite.com/gun-control-kellermann-3times.html

http://www.reason.com/news/show/30225.html

http://www.guncite.com/journals/tennmed.html

http://www.guncite.com/kleckjama01.html -- this one is an especially good article published in the Journal of the American Medical Association

quoting the above:

This association was at least partly attributable to confounding factors that are known to be strongly associated with both gun ownership and homicide victimization, such as dealing in illicit drugs (but not drug use) and membership in a street gang. Either of these confounding factors alone is associated strongly enough with gun ownership and homicide victimization to produce a spurious odds ratio of 2.8,<14> and neither factor was controlled by the researchers. Indeed, most factors that increase the risk of homicide victimization in a way that is evident to the subjects are likely to also motivate some of them to acquire a gun for self-protection.<15> Thus, a positive gun-homicide association is expected even if gun possession had no impact whatsoever on homicide risk.

-------------------------


Here are some other threads discussing problems with methodology w/r/t studies of this kind:

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=256837

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=259353

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=263399#263698

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=275068

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=275068

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=290313#292405

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=314650

[spin]

1. Thanks. Your post should help clarify this issue. (n/t)

[rfranklin]

2. Well, that proves it...guns are a safety device...

sounds logical to me.

[jazzhound]

12. Well, your post proves it.......

...........you have absolutely nothing worthwhile to contribute to this conversation.

[Remmah2]

29. Sort of like condums.

You never know when you might need one.

[rfranklin]

30. That's a conundrum to me...

That's the long and short of it!

[SteveM]

36. Well, they are safe devices in my household. Thanks. nt

[wtmusic]

3. Cut and paste it 100x if you like; it's not a valid criticism.

I am pulled to the earth because of gravity? Nonsense - correlation doesn't equal causation. :silly:

The tortured logic and justifications offered by the NRA are akin to those offered by addicts, and are only interesting as case studies for the psychology of fear.

[X_Digger]

4. What's invalid? Please, point out the error in my logic.

How is it that studies that compare dissimilar groups of people yet draw conclusions based on that comparison are valid?

Why is it that no other researcher outside of a small circle of academics have reproduce any of the studies?

Why is it that when these same researchers repeat a study, they can't find the same level of 'risk'?

And finally, why is it that you never actually have anything to refute any of these arguments? http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=118&topic_id=293219#293320

[wtmusic]

6. How is it that studies that compare dissimilar groups of people

yet draw conclusions based on that comparison are valid?

Researchers draw conclusions from studying dissimilar groups all the time.

"As compared with the controls, the victims more often lived alone or rented their residence. Also, case households more commonly contained an illicit-drug user, a person with prior arrests, or someone who had been hit or hurt in a fight in the home. After controlling for these characteristics, we found that keeping a gun in the home was strongly and independently associated with an increased risk of homicide (adjusted odds ratio, 2.7; 95 percent confidence interval, 1.6 to 4.4). Virtually all of this risk involved homicide by a family member or intimate acquaintance."

Why is it that no other researcher outside of a small circle of academics have reproduce any of the studies?

It might have something to do with the fact that conducting a study of this magnitude requires a lot of funding, and: "Kellermann’s findings have been linked to the June, 1996 Republican-led decision of the U.S. House Appropriations Committee to strip US$2.6 million from the budget of the National Center for Injury Prevention and Control at the U.S. Centers for Disease Control – the exact amount previously set aside for NCIPC/CDC research into the causes and effects of firearm-related death and injury."

Wayne LaPierre and the NRA successfully used money and intimidation to stifle any research which might portray (even if accurately) risks associated with handgun ownership. Do you want to tell the truth or have a career?

Why is it that when these same researchers repeat a study, they can't find the same level of 'risk'?

You're in the wrong venue; you should be working in a laboratory. No consecutive studies in public health ever show the same levels of risk and they don't attempt to. The point is to draw broad conclusions from many smaller independent ones, and Kellerman's 1993 study was a decisive confirmation of the 1986 one:

"These results confirmed the 1986 finding that, in the net, a firearm in the home represents a greater risk overall than the protection it may offer against intruders, either indirectly or by discouraging potential assaults. Kellermann noted that the study demonstrates the pervasiveness of domestic assault, as compared to better publicized crimes such as home invasion, but continued to stress the role of handguns in increasing the lethality of such assaults."

http://en.wikipedia.org/wiki/Arthur_Kellermann

Regarding the NRA and their manic attempt to quash anti-gun research: "The final appropriation language included the following statement: “none of the funds made available for injury control at the Centers for Disease Control and Prevention may be used to advocate or promote gun control”. These words appear in every CDC grant announcement to this day." A question for you: if handgun ownership was so safe, why wouldn't the NRA welcome a study which vindicates it?

[X_Digger]

7. And yet..

They didn't control for neighborhood crime rate between case and control (delta between case and control was 45 blocks for 85%), relied on self-reporting of previous criminal history and gun ownership, and performed no cumulative regression.

Check table 4 in the results section of the study. Each of these items independently had more of an impact on being shot than having a gun in the home, according to Kellerman- living alone (AOR 3.7), home rented (AOR 4.4), domestic violence (AOR 4.4), illicit drug user (AOR 5.7); the rest are gun ownership (AOR 2.7) and previous arrest (AOR 2.5).

Confounding factors that have a cumulative effect were not accounted for; ie, in suicide research, two of those are often cited- living alone and renting. The AOR for these co-variates are cumulative in major suicide studies- If the AOR for living alone is 1.0 and the AOR for renting is 1.0, the AOR for both is NOT 2.0, it's usually on the order of 4.0 - 6.0, depending on study.

They even acknowledge that their data is not conclusive in the study- "Living in a household where someone had previously been hit or hurt in a fight in the home was also strongly and independently associated with homicide, even after we controlled for the effects of gun ownership and the other four variables in our final model (adjusted odds ratio, 4.4; 95 percent confidence interval, 2.2 to 8.8)" But even after that analysis, they retained data from homes with domestic violence, even though its impact was larger than gun ownership. (The logical thing to do on finding this would be to go back to the pool of 600 and find another match without domestic violence- otherwise, you have no way of knowing which factor, if either, is determinant.)

Of course, they bring up the same possibility I did re correlation and causation- "Third, it is possible that reverse causation accounted for some of the association we observed between gun ownership and homicide... Finally, we cannot exclude the possibility that the association we observed is due to a third, unidentified factor."

Gee, fancy that. LOL!

You're in the wrong venue; you should be working in a laboratory. No consecutive studies in public health ever show the same levels of risk and they don't attempt to. The point is to draw broad conclusions from many smaller independent ones, and Kellerman's 1993 study was decisive confirmation of the 1986 one:

"These results confirmed the 1986 finding that, in the net, a firearm in the home represents a greater risk overall than the protection it may offer against intruders, either indirectly or by discouraging potential assaults. Kellermann noted that the study demonstrates the pervasiveness of domestic assault, as compared to better publicized crimes such as home invasion, but continued to stress the role of handguns in increasing the lethality of such assaults."

(emphasis added)

Let me repeat that again.. 1986 AOR? 43 (not 4.3, 43) -- 1993 AOR? 2.7

Unless you think that the relative risk dropped by a factor of fifteen, there's obviously something wrong with the methodology.

Yeah, "broad" fucking "conclusions", all right.

Kellerman didn't study risk versus protection. At no point did the survey ask if a gun had been used to defend either the case or the control. That's bald assertion not studied, not backed up by data. Of course, that's wikipedia's text, surely not Kellerman, right? Not really.. "We did not find evidence of a protective effect of keeping a gun in the home, even in the small subgroup of cases that involved forced entry." -- except that their only criteria for protection required that someone be killed. (see case selection criteria.)

Now, here's one for you- please explain to me how having a gun in your home makes you more vulnerable to being shot by someone else's gun.

Is that the gunz as bacteria on a doorknob theory, again?

A subsequent study, again by Kellermann, of fatal and non-fatal gunshot woundings, showed that only 14.2% of the shootings involving a gun whose origins were known, involved a gun kept in the home where the shooting occurred. (Kellermann, et. al. 1998. "Injuries and deaths due to firearms in the home." Journal of Trauma 45:263-267) ("The authors reported that among those 438 assaultive gunshot woundings, 49 involved a gun 'kept in the home where the shooting occurred,' 295 involved a gun brought to the scene from elsewhere, and another 94 involved a gun whose origins were not noted by the police .") (Kleck, Gary. "Can Owning a Gun Really Triple the Owner's Chances of Being Murdered?" Homicide Studies 5 <2001>.)

[wtmusic]

8. You didn't answer my question.

If handgun ownership was so safe, why wouldn't the NRA welcome a thorough, complete study which vindicates it? :shrug:

[PavePusher]

9. Run one and put it up for peer review.

The NRA isn't going to stop you. Why are you trying to shift some invented "blame" to them?



By the way: http://www.cdc.gov/mmwr/preview/mmwrhtml/rr5214a2.htm

The best they could come up with? "Insufficient evidence" to prove any conclusion, pro- or anti-gun.

Hmmmmm.....

[wtmusic]

13. The NRA has done a very effective job of stopping researchers

by cutting off funding. Have $2.6M for me? I'll give you (another) great, objective gun study.

Of course that's the best the CDC could come up with - they're legally prevented from coming up with anything that "may be used to advocate or promote gun control".

A wonderful environment for science. :eyes:

[PavePusher]

20. I'm sure a private entity can run any study they want...

and the NRA can do nothing about it.

If you speak of public funding, well, I'd have to say I'd be O.K. with it as long as there was no politics involved. Good luck on that one.

Then again, I also think that funding a study that could ultimately lead to the curtailment of a fundamental Civil Right is probably a really bad idea.

[wtmusic]

22. No, the truth is you do want politics involved...YOUR politics.

The NRA has already foundered scientific method by cutting off any possibility of a certain result.

Your rights of gun ownership are significantly curtailed right now. There are two possibilities:

1) We open the doors for all Americans to all rights to bear any arms
2) We decide we must impose certain limitations based on what...hunches? Wouldn't a scientific approach be preferable?

[PavePusher]

24. Your assumptions of my positions are erroneous.

Unless you are finished making such assertions, this conversation has reached its conclusion.

[wtmusic]

25. I apologize.

Now back to my last post. Which is it, #1 or #2?

[Name removed]

34. Deleted message

Message removed by moderator. Click here to review the message board rules.

[hack89]

45. Is the Brady campaign intimidated?

or just irrelevant?

[X_Digger]

32. No, I ignored your red herring.

We're talking about the validity of Kellerman's "study". If you want to talk about other subjects, start an OP and we can talk there.

Back on point, eh?

[SteveM]

37. If NRA ran a "thorough, complete study which vindicates it," would you believe it?

[Euromutt]

11. "Kellerman's 1993 study was decisive confirmation of the 1986 one"

This is about as close as I've ever come to using that despicable ROFL smiley. The validity of scientific research is largely dependent on the ability of others to replicate the findings. It doesn't mean a goddamn thing when a researcher replicates his own findings, because if his methodology was flawed the first time round, and he failed to notice it then, he's the very person least likely to notice those flaws the second time around. The scientific method was developed to compensate (as best as possible) for the possible failings of individual researchers (who, being human, may fall prey to confirmation bias and the like) by having others cast a critical eye over their work. Try looking up "Jacques Benveniste"; he's an excellent example of a scientist who replicated his own findings numerous times, but failed to actually do so when anybody outside his own team was actually watching.

I'm going to refer here to an older post of mine (which X_digger already linked to in the OP), in which I expounded at some length to the problems found with public health research on firearms in general; see here http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=118&topic_id=290313&mesg_id=292405

Kellermann's work is merely the most egregious example of the genre. After his 1986 study was thoroughly savaged, he insulated himself from criticism in his 1993 study ("Gun Ownership as a Risk Factor for Homicide in the Home") by refusing to share his research data. What is even more remarkable is that the NEJM published the study, even though Kellermann had not deposited his research data with the journal. In other words, neither the journal nor its peer reviewers were in a position to verify that the study's findings were actually supported by any data, let alone valid data, and yet they approved it for publication anyway. This, combined with the same pass they gave John Sloan in 1990 on his study "Firearm Regulation and Rates of Suicide," is--or at least, should be--an indelible stain on the reputation of the NEJM as a reputable journal, even more than The Lancet's heel-dragging over retracting Wakefield's fraudulent MMR study. At least Wakefield had to cook up some fake data; they didn't let him get away with not showing any.

[jazzhound]

14. "Kellerman's 1993 study was decisive confirmation.......

.........of the 1986 one."

For Christ's sake! I hired a coyote to guard my hen house, and now I come to find that he's been eating my hens!! I just can't figure out where I went wrong!!!

[wtmusic]

15. Kellerman's 1993 data has been public for thirteen years.

It's been hashed over time and again without any significant challenge to his methodology or findings.

Fail.

[jazzhound]

16. All I can say is.........wow.

You're perfectly alright with a "scientist" who debunks his own findings!!

http://rkba.org/research/suter/med-lit/benefits.html

[Name removed]

17. Deleted message

Message removed by moderator. Click here to review the message board rules.

[jazzhound]

18. Do you actually want to go on record as denying

that Kellerman dramatically revised his own "study" downward from the "43 times" claim to the "2.7 times" claim? Really?

[wtmusic]

21. Of course, because he didn't do that at all.

The 1986 study concluded:

"Every time a gun is used in self-defense, it is 43 times more likely to be used in a homicide, suicide, or accidental shooting."

The 1993 conclusion was:

"When a firearm is kept in the home, there is a 2.7 times increase in risk of homicide."

Keerist...debating you people is like shooting fish in a barrel.

[jazzhound]

27. "The 1986 study concluded: Every time a gun is used

in self-defense, it is 43 times more likely to be used in a homicide, suicide, or accidental shooting."

False. This is the conclusion of the 1986 study:

Conclusions:
Ratio of killed by stranger to killed by person known = 12:1
Ratio of accidental deaths to self-protection homocides = 1.3:1
Ratio of criminal homocides to self-protection homocides = 4.6:1
Ratio of suicides to self-protection homocides = 37:1
Ratio of suicides, criminal homocides, and accidental deaths
to homocides for self-protection = 43:1

So the truth of the matter is that Kellerman was comparing homicides to homicides --- not simple defensive gun uses to homicides. This is important because he's willfully distorting his results by comparing the burglar body count to friends and family members killed by guns in the home fully aware that burglars are infrequently killed by homeowners through defensive gun uses. EDITED TO ADD: As a sidenote --- how is it that Kellerman seems to think that the purpose of defensive gun use is to kill burglars rather than to defend the homeowner from assault? (Where's a loving liberal when you need one!!)

And yes -- still there is difference in the stated conclusions. Given the justifiable beat-down he received for his '43 times' "study", it's pretty obvious to me that he was trying to walk back with his '2.7' "study". Your mileage may vary.

Care to address the critique of the 2.7 "study" furnished by X_Digger?

[Callisto32]

35. That was beautiful, Jazzhound.

Basically, all that 43:1 crap shows is that violent people are more likely to kill people than normal folks.

[SteveM]

38. His ammo has run out; the barrel still teems.

[jazzhound]

19. If you want to honesty attempt to refute the points made

by Dr. Suter, we can have a discussion. Dismissing Suter's points based on the particularly webpage where they were posted is dishonest on it's face.

Regarding Kellerman's willful dishonesty, I believe this pretty much sums it up:

The true measure of the protective benefits of guns are the lives saved, the injuries prevented, the medical costs saved, and the property protected -- not the burglar or rapist body count. Since only 0.1% to 0.2% of defensive gun usage involves the death of the criminal, <10> any study, such as this, that counts criminal deaths as the only measure of the protective benefits of guns will expectedly underestimate the benefits of firearms by a factor of 500 to 1,000.

[wtmusic]

23. Who the hell is Edgar Suter?

He's some doctor with no research qualifications. Here's some fun stuff if you want to bother.

http://scienceblogs.com/deltoid/1995/11/international-00024.php

Next...

[jazzhound]

26. Who the hell is Arthur Kellerman?

You want to dismiss Suter as a gun lobby shill? Alright. Try dismissing lifelong liberal Democrat Gary Kleck --- the only criminologist who has won the highest honor from the American Society of Criminology (M. Hindelang Award) on the subject of gun control. Pages 56-62 deal with your champ --- y'know the guy who believes that comparing the burglar body count to the number of friends and family members killed by guns is an honest measure of the value of firearms in home defense.

http://books.google.com/books?id=xJ3Y2-CHYfMC&pg=PA61&lpg=PA61&dq=kleck+kellermann&source=bl&ots=sBP0LmKhVF&sig=p8X2FB8GslXZsNWGTL2atjOEnds&hl=en&ei=lelkTJWBIIfEsAO-0qSwDQ&sa=X&oi=book_result&ct=result&resnum=5&sqi=2&ved=0CCQQ6AEwBA#v=onepage&q=kleck%20kellermann&f=false

[Euromutt]

31. Not really

Yes, after four years (and much badgering), Kellermann grudgingly released a partial dataset, that is now no longer available. This is where it used to be: http://www.icpsr.umich.edu/cgi/archive.prl?study=6898

It's been hashed over time and again without any significant challenge to his methodology or findings.

What an... interesting definition of "significant" you must have. I get the impression that in your vocabulary it means "that I would not prefer to ignore as inconvenient."

In response to Kellermann's 1993 study, a class of students taking a statistics course wrote a letter to the editor of the NEJM, asking:

Finally, the authors also state, "One or more guns were reportedly kept in 45.4 percent of the homes of the case subjects." This implies that no guns were kept in 54.6 percent of the homes of the case subjects. In how many of the homicides was the victim killed with a gun that was kept in the house rather than a gun that was brought to the house by the perpetrator?

Kellermann never provided a direct answer to that question, and the partial dataset he eventually released did not contain anything that could point to an answer. It is entirely possible that Kellermann cum suis, in their eagerness to produce findings that supported their predetermined conclusion, completely overlooked the need to check whether the firearm used to commit the homicide was actually one kept in the household in question. (If so, this is rather remarkable, given that in his 1986 study, the "perpetrator" listed in three of study cases who were homicide victims was "Seattle PD"; since police tend to bring their own guns, rather than relying on being able to borrow a firearm kept in the household with which to shoot an occupant, the likelihood is negligible that these three cases were killed with a firearm kept in the household. On the basis of this experience, Kellermann should have been ready to check for the distinction.)

As the Mathematical Statistics 460 class at St. Louis University pointed out, the study's conclusion asserted that "'In the light of <other> observations and our present findings, people should be strongly discouraged from keeping guns in their homes,' implying that a gun in the home is a causal factor in homicide <...>." It severely undercuts this conclusion that the researchers made no apparent effort to determine the origin of the firearm involved in the shooting (or if they did, they concealed it).

But ultimately, the issue with Kellermann's work--as with all public health "research" into firearms--is that it consists entirely of retrospective studies. To quote Dr. David Gorski, MD, PhD (research oncologist and managing editor of sciencebasedmedicine.org):

<...> here's one thing to remember about retrospective studies in general. They often find associations that later turn out not to hold up under study using prospective studies or randomized trials or, alternatively, turn out to be much weaker than the retrospective study showed.
http://www.sciencebasedmedicine.org/?p=2962

For this reason, retrospective studies are considered to be useful for generating hypotheses for further research, but in and of themselves, they have little value as scientific evidence of anything. Any positive findings have to be validated by cohort studies and/or prospective studies before they start to become interesting.

So if nobody's tried very hard to produce evidence contrary to Kellermann's findings, that's because it's not worth the effort. It's just another retrospective study that found a correlation, like dozens of others that get published every month and never pan out to anything bigger. That is, scientifically; the importance of Kellermann's work consists entirely of the PR value to the gun control lobby.

[PavePusher]

5. I don't think you actually addressed anything in the O.P.

Perhaps you could present some evidence to refute it?

[Euromutt]

33. I think you need a remedial course in basic logic

I am pulled to the earth because of gravity? Nonsense - correlation doesn't equal causation. :silly:

Try looking up the terms "necessary condition" and "sufficient condition"; the Wikipedia article isn't bad: http://en.wikipedia.org/wiki/Necessary_and_sufficient_condition
Statistical correlation is a necessary condition for establishing a causal relationship, but it is not a sufficient condition. The correlation that, when you release an object, it falls to the ground could just as readily be interpreted as evidence that angels/spirits/whatever push the object down. Every time, without fail. If you're a dedicated pseudoscientist, you could even present the fact that when objects are released, they fall to the ground, as evidence of the existence of angels/spirits/whatever (who else would be able to always be there to push the object down but a supernatural entity?) because why should you let the fact that's a completely circular argument stop you?

More seriously, correlation not being a sufficient condition to establish causation is a cornerstone of the scientific method, and you cannot disparage it without discarding the scientific method entirely. Otherwise, you would have to accept that, every time someone went to a faith healer or a naturopath or a witch doctor or some other form of quack with a headache (or cut out the middleman and prayed directly to God, gods, or spirits) and that headache went away, it must have been the quack or the invisible buddy who cured the headache.

The methodology of public health research concerning firearms, such as Kellerman's, is pretty similar: dude had a gun, dude got killed, therefore having gun got dude killed. Post hoc ergo propter hoc.

[jazzhound]

10. The "43 times" fallacy becomes the "2.7 times" fallacy

When Arthur Kellerman was soundly rebuked for his assertion that a gun in the home was 43 times more likely to kill a friend or family member than to kill an intruder, he ultimately "revised" his figure to 2.7 ---- essentially debunking his own "study"! The fact that pro-control supporters continue to quote Kellerman's "work" is either a reflection of their ignorance, dishonesty, or both. Since the object of defensive gun is not to kill intruders but rather to defend the occupants of the dwelling Kellerman's "study" was flawed from the outset. In the vast majority of defensive gun uses, the occupant not not only doesn't kill the intruder, but doesn't even fire the weapon. It's difficult to believe that Kellerman was unaware of this fact -- so his "studies" are in all probability willfully dishonest.

In one of his papers (don't recall which) Gary Kleck exposes multiple occasions when Kellerman knowingly concealed facts that ran counter to his assertion(s). Kleck was aware of this because he sat through meetings/conferences with Kellerman wherein these facts were presented.

http://rkba.org/research/suter/med-lit/benefits.html

[jazzhound]

28. Upon further reflection:

"In one of his papers (don't recall which) Gary Kleck exposes multiple occasions when Kellerman knowingly concealed facts that ran counter to his assertion(s). Kleck was aware of this because he sat through meetings/conferences with Kellerman wherein these facts were presented."

I could be wrong on this ---- might be thinking of David Hemenway.

Footnote: My previous incorrect spelling of Arthur Kellermann's name was unintentional.

[Hoopla Phil]

39. This thread needs a bump.

Do to the Kellermann farce making another appearance.

[friendly_iconoclast]

40. Kellermann had about as good a 'weapons dataset' as Bush did, pre-Iraq

And in both cases, people supported this horseshit because it was what they wanted to believe.

[jazzhound]

41. kick NT

[jazzhound]

42. kick NT

[jazzhound]

43. kick NT

[Hoopla Phil]

44. A classic thread worthy of kicking.

[X_Digger]

46. *bump*

Hemenway's blathering again, thought it'd be a good time to bump this.

eta: Hemenway's "works" suffer from the same faults as Kellerman- they've co-authored a few pieces with the same flaws.

[Bold Lib]

47. This thread needs a bump. Especially with the crap studies being reposted

again.

yup