43:1 is GOOD!

Status
Not open for further replies.

glummer

Member
Joined
Apr 20, 2006
Messages
699
Location
NY
There have been many analyses showing the flaws in the infamous Kellerman study purporting to show that a gun in the home is 43 times more likely to kill a household member than an intruder.

There is, however, one point which I have never seen addressed: even if Kellerman’s figures were accurate, 43:1 is a GOOD result, NOT a bad one.

The size of the apparent risk disparity depends heavily on the fact that the figures are not adjusted for exposure time.
But ANY risk factor in one’s household is more likely to affect a household member because of the enormously greater time of exposure, compared to an intruder.
Average exposure times for household members would measure in person-years; average exposures for intruders would probably be in SECONDS, or even fractions of a second, considering that many (most?) households don’t ever HAVE intruders.

Thus ALL (not just firearms) risks will be skewed toward higher household member deaths.

GUNS will actually show the BEST odds!

Consider the knives in your kitchen.
How often does anyone kill an intruder by using a knife from the kitchen?
Virtually never.
But in-home violence with such knives is quite common.
The Household_Member/Intruder death ratio is probably well into the thousands.

For other risk factors in the home, the odds become astronomical.
Consider your electric wiring, natural gas supply, prescription drugs, the rope in your garage – what risk do they pose to an intruder?
When was the last time a rapist died from a fall in his victim’s bathtub?

Many of the risk factors in the home have likely odds in the range of MILLIONS to one.

At ONLY 43 to 1, firearms would have probably the BEST odds of intruder death of anything one can own. (Not surprisingly, since they are the only common object DESIGNED to kill intruders.)

Thus, when the Anti’s use the 43 to 1 argument, it is not even necessary to debunk the figures – we can point out that everything else is much, much WORSE.
 
That's not a bad analysis, and would probably be worth carrying forward.

In the meantime *rolls 20-sided die* I'd like to cast a "Summon Tim Lambert" spell.
 
glummer...spot on. in statistics/probability classes one thing that was learned fast is that you can manipulate data to justify virtually any desired outcome: stretch some parameters, filter out things that upset your desired outcome, etc. it is so obvious a truism it sometimes takes a genius to re-state it.
 
All types of violence, whether murderous or assaultive, is statistically overwhelmingly recieved from someone close to the victim rather than from strangers. That 43:1 ratio is meaningless because most violent folks hurt the ones closest to them (family, household members, significant others, etc.) So, it should be expected that guns (or any other type of weapon) will be used on those close by more often than on those not nearby. For example: Per capita, how many husbands slap their wives around compared to how many men slap strangers, co-workers or neighbors around? How many guys beat their wives per capita compared to how many people, per capita, have some stranger crawl in their window? Family violence is relatively common per capita compared to actual break-ins per capita.
 
woad yurt.......thank you for bringing in another interpretation of 'facts & figures' which make glummer's statement both more powerful to individuals & less powerful the those ( as yet ) unaffected. reminds me of when my dad had 3x bypass & a valve~~i asked the doctor about his chances ( knowing that it was about 80% ) and the doctor said sternly 100%--he will live or he will not! makes you to think, in a one off situation, you cant have 50/50 or 80/20...you have one or the other...period.
 
Excellent analysis, and a way to debunk the supposed impact of that statement.

But I still like TallPine's too. I use the "people with stairs in their homes are much more likely to die from falling down stairs in their homes than people with no stairs in their homes." Use it with pools, knives, chemicals, whatever.
 
Yeah, it's sorta like people who own motor vehicles are much more likely to drive drunk.

Heh. That's sort of common sense, eh? If a drunk doesn't have a car, then he must be statistically less likely to drive while intoxicated. Just like a home owner without a gun is statistically less likely to shoot an intruder. I wonder how that affects the stats on knives and clubs. I wonder if they'd be affected proportionately. Hmmm.

But a little more seriously, this is why we take cars (or driving privileges anyway) away from convicted drunk drivers, and take guns away from convicted felons. Innocent until proven otherwise, and these folks proved they can't be trusted.

And back on topic... I find myself wondering how this particular analysis managed to be missed for this long. Brilliant.


-T.
 
That all makes perfect sense. We all know that facts and statistics are on our side in this fight. The person I'll hail as a genius is the one who can figure out how to convince the antis that facts are a better basis on which to make important decisions than are emotions. That, and how get the mass media exposure necessary to convince the sheepish masses that we are right.
 
Great analysis. We all know that facts and statistics are on our side in this war.

The person I'll hail as a genius is the one who can figure out how to convince the general Anti population that facts are a better basis on which to make important decisions than are emotions. That, and how to get the mass media exposure needed to convince the sheepish masses that we are right.


Edit...
After I hit the submit button, I thought my post came across as sarcastic toward glummer's analysis. I didn't mean it to be. I was just expressing frustration at the fact that no matter how right we are, no matter how heavily the facts are on our side, the antis always seem to get their flawed message to the greatest number or people.

Glummers analysis is great. We all know that the facts are OUR friends, most of us, myself included, never thought of it and never thought to look at the data in the way glummer did.
 
The only problem with the house hold knives theory it is probably false. I would say almost every single household in America has kitchen knives. So if you take the number of homes with kitchen knives and their use in domestic violence, protection, premeditated, etc., and compare that to the far less number of house holds with firearms, and I think you will have a smaller ratio for the knife figure. The simple reason is that there are more good, non-violent people in this country than there are those with serious domestic problems, and there are fewer houses with guns than there are knives.

I realize there are an estimated 80 million gun owners, but that doesn't necessarily mean 80 million households.
 
The data - presented in short form at http://guncite.com/gun_control_gcdgaga.html - indicates that the total number of deaths was 389 and of those, 333 were suicides. Nine deaths were justifiable homicides, and 389/9 ~ 43 which is where the 43:1 ratio comes from.

But if 333 of 389 were suicides, then maybe all Kellerman showed was that if suicidal people have guns, they might shoot themselves rather than drink Drano.

Yes, exposure time is skewed - your point is very well taken - but Kellerman's entire dataset and analysis is so flawed that it fails on its own.

Regards,
Andrew
 
No. the resut reported by the Kellerman repport is flawed for two reasons:

1. They only surveyed homicides. The intent was to find "risk factors", treating homicides like a disease. The findings aren't false, they are just presented in a deceitful manner. What is being compared is the risk of being killed by a gun to beng klled in general, not as most people believe, being killed as a result of owning a gun, since the number of household with guns and without homicides wasn't a part of the study.

2. Other risk factors were ignored. Let's say you are a drug dealer, or your dad or other family member is. There's a real high chance that: a. That family member will have a gun even if it is illegal to do so; and b. that person will die from a gun. Just a most people would expect an alcoholic to commit suicide or have a fatal accident. So what what the cause of the homicide? The gun, or a criminal background or history of drug addiction?

Analyzing most of this doesn't require a study. All gun accident fatalities in the US in a year are usually less than 100. At the same time there are about 7-9000 murders. So the risk of being murdered is almost 1000 times that of having a gun accident. If you figure the overall risk for having a fatal accident with a gun as a part of owning a gun, it's about .0000012:1 or about 1 in 8,000,000. And that doesn't even take into account the same drug and alcohol abusers and criminals that tend to have way more accidents than the average person. I had to figgure this out after a forum troll on another forum tried to make the pooint that by not owning a gun, you reduced your chances of being killed by a gun to zero. Even the Kellerman report didn't cover the possibility of the proximity of the gun being due to the fact that the murderer brought it there.
 
Last edited by a moderator:
Outlaws
The only problem with the house hold knives theory it is probably false. I would say almost every single household in America has kitchen knives
And virtually no one succeeds in killing an intruder with any of those knives.
I don’t see your logic here.
There are a fair number of knife deaths involving kitchen knives in the home (sometimes it’s even an intruder using the homeowner’s knife), and virtually NO intruder deaths involving those same knives.
The ratio HAS to be greater than 43:1.
 
ROMAK IV
What is being compared is the risk of being killed by a gun to beng klled in general,
I thought it was the ratio of household member gun deaths vs. intruder gun deaths. ??
 
The 43:1 number is also flawed because the vast majority of defensive gun uses do not result in the death of the miscreant. So one could argue that those who use guns are actually quite circumspect and rational, and do not automatically shoot dead any stranger that appears in their home.

Or I suppose one could argue that gun owners need to learn to shoot straighter and kill more burglars :D
 
And 1 in 5 can be bad …

Kellerman also has 41 criminal homicides and 9 justifiables – that is 50 attempted murders with an 18% death rate for the attacker.
Since the attacker presumably times the occasion to his advantage, and the victim will often not have his gun to hand, that is a pretty impressive defensive result.
 
Status
Not open for further replies.
Back
Top