Interestingly enough, the police - in Canada - do have a duty to protect individuals
R. v. Nixon (1990) B.C. Court of Appeal
Of course, us poor oppressed Canadians can’t have guns all willy-nilly, and people don’t get killed as a result of violent crimes involving guns, and…
…wait a minute…
Ya don’t figure those things are in any way related do ya?
Ahh, never mind - you Americans are too busy being ‘free’ to worry about silly little bits of logic like that…
z
Tejota:
In response to Spoofe’s post:
What’re you, God? You happen to know for certain the objective realities of all the respondents of that [Kleck] survey?
All studies? All thirteen of them? Unless you’re accusing the authors (Kleck & Gertz) of incompetence, or even worse, then don’t you think that they would have tailored their survey in such a way as to try to reveal, and then mitigate, this bias? Profs. Kleck and Gertz certainly seem to think so.
What did you do, C & P directly from Tim Lambert?
Barbarian:
I’m also sure that it’s quite atypical. Citing the atypical and trying to portray it as “normal” and therfore justifiable is…well, you know.
But you’re right in one respect: a firearm is not a defense, at least in the passive sense of the word. Mere possession confers no protection. It is a defense in the active, though. Once brought into action (from holster to hand is an eternity of difference, an infinity of possibility).
You know, the way a CIWS (Close In Weapon System; the “R2D2” anti-missile guns on modern warships) is an active defense.
Minty Green:
And exactly how many criminals did you interview to arrive at this conclusion?
And if you were at all familiar with the mechanics of self defense, you’d not be posting little nuggets concerning frequency of carry among CCW permit holders. I am a CCW permit holder, and I rarely carry at all. Only when my business takes me into high crime areas. And when I do carry, I am consciously aware of my environment, and think tactically about darkened doorways, alleyways, and such, avoiding them to preclude ambush. I conduct my business and leave the area.
As Glitch has admirably pointed out time and again, no weapon, no physical capability, no style of unarmed combat, will help you if you are not mentally alert to your surroundings, in order to spot trouble early, and are mentally prepared to act, quickly, decisively and effectively (as provided by training) to defend yourself. A firearm isn’t a magic wand to make the world, or even just the bad guys, dance to your tune.
Carrying just because you wanna strap on a gun is frowned upon (but allowed) by the State of Texas, and strongly discouraged in the classroom instruction one is required to take in order to obtain a CCW permit. Besides, the damned things are uncomfortable, and a nuissance. But it sure beats being dead, in a hospital with a busted skull, or bent over a trashcan with your pants around your ankles. The instructors emphatically suggested further self-defense training was truly necessary to effectively defend oneself if attacked.
Amd BTW, my info is gleaned from other CCW permit holders and instructors (quite a few of them are police officers, or retired POs. I’m sure they’re really in a hurry to put themselves out of work by training and arming the entire populace). It’s a gun thing.
We gun nuts like to hang out at Militia HQ, swill beer, chaw tobaccy, hoss whup mud people, keep our wimmin’s from gettin’ uppity notions (like readin’, writin’ and wearin’ shoes), and burn crosses, all under the glory of the ol’ Stars-and-Bars.*****
Careful there, Kimstu:
True enough in essence and in fact, as of now. But at what point does one become the other? Who draws that line, and on what basis? Does Barbarian draw that line after an incident in the paper convinces him (her?) that such incidents are normal? Or does Tejota, after reading incomprehensible (to anyone other than another statistician) studies and swearing before Congress to God and sonny Jesus that they’re absolutely true and correct?
Having read Kleck/Gertz, Hemenway, Cook & Ludwig, Lambert, et. al., I can only conclude that I need at least an Associates in Statistical Math just to understand what they’re trying to say, much less glean any comprehensive meaning out of their works. Duelling statistics by people who really don’t understand them is a meaningless debate.
As is usually the case in such instances, I’d not be suprised to find that the truth lies somewhere in between, as you, and Mr. (Prof.? Dr.?) Tom W. Smith points out.
As to the OP:
citing a 33 y/o State Supreme Court ruling of a 42 y/o crime is hardly the ringing pro-gun argument I’d try to make. Notwithstanding the gun control laws of 1950’s New York City, Miss Riss’ assailant was a stranger hired by the loon who was stalking and threatening her. It is doubtful that, even if she had had a firearm, that she would have been able to successfully defend herself; the threat that isn’t recognized can hardly effectively be defended against.
In more recent times, various Police agencies (or more accurately, the elected politicians who appoint police officials and set policy) have become more responsive to the public’s concerns, taking a more proactive stance, as the law allows, to protect, or at least help protect, individual citizens.
Of course, a restraining order is little defense against a knife, a bullet, or a cupfull of lye tossed in one’s face; but the ability of individual citizens to look to their own safety is still more attractive that dialing 911 and hoping and praying that someone, anyone shows up in time to save you.
Even if only 1 in a thousand lives were saved in a criminal attack by a firearm, then it would still be worth it.
The mathematics are certainly debateable, but the underlying morality is no different from the liberal paean that guns must be outlawed and confiscated because it might save even one life that might otherwise not have perished.
Zoony: yes, they are related. AS is myriad other social factors. Pointing your finger at one and saying “A-HA!” is moronically simplistic. Go play some more hockey, and for God’s (and our’s) sake, wear a fucking helmet.
ExTank
“Mostly Harmless” 
*[sub]for the sarcasm impaired: this paragraph is emphatically untrue[/sub]
Motto of the Los Angeles Police Department: " * ** To Protect ** * and Serve".
There ya go. Debate settled.
stoid
Heh heh heh…I’m sure the numerous African Americans and Latinos in LA are comforted by this motto.
Marc
Yes, but to protect and serve whom, dear Stoid? Society at large, or the individual members of that society?
[QUOTE]
*Originally posted by ExTank *
**I’m also sure that it’s quite atypical. Citing the atypical and trying to portray it as “normal” and therfore justifiable is…well, you know. **
I don’t know how ‘normal’ it is… but it seems to be a unfortunately regular occurence for a firearm to be used against family members.
I listen to scanners overnight-- it’s my job. I don’t have recorded stats, but here in Vancouver it seems like the majority of gun use is by criminals attacking other criminals eg. shots fired as people raid a marijuana grow op.
Number 2 in frequency is family member terrorizing the rest of the household with his firearm. I hear this call about once a week in a city of less than 2 million people
I never hear of anyone stopping a crime because they own a firearm.
**
You know, the way a CIWS (Close In Weapon System; the “R2D2” anti-missile guns on modern warships) is an active defense.**
I’d like to see someone shoot a gun out of someone’s hand 
As Glitch has admirably pointed out time and again, no weapon, no physical capability, no style of unarmed combat, will help you if you are not mentally alert to your surroundings, in order to spot trouble early, and are mentally prepared to act, quickly, decisively and effectively (as provided by training) to defend yourself. A firearm isn’t a magic wand to make the world, or even just the bad guys, dance to your tune.
Hear, hear! Best example of this I’ve seen in fiction is in Heinlein’s ‘Tunnel through the sky.’
Even if only 1 in a thousand lives were saved in a criminal attack by a firearm, then it would still be worth it.
But what if 2 kids die because they were playing with said firearm?
The study I really want to see would be:
number of innocent victims v. number of crimes stopped
And I’m gonna leave it at that. It’s the weekend. Go enjoy it, and don’t shoot nobody
Half of that study (estimated number of crimes stopped) has been presented. Numbers of innocent victims are most likely available (it being 3:00 AM, I don’t want to dig around fedstats.gov to find the numbers). You simply compare those two numbers.
As for both numbers being researched at once, using the same criteria for judgement, I’m not aware of any study being done - which doesn’t mean it doesn’t exist. Either that, or I’m aware of one, but like I said, it’s 3:00 AM, and my brain’s probably just not recalling it.
None. But I stand by the statement. There is simply no perception among the Texas populace at large that people are actually carrying concealed weapons, a point that even you admit is grounded in the cold, hard facts:
Which is precisely the point I made above about my friends with concealed carry permits. If you ain’t armed, that gun ain’t defending squat.
Huh. Remember that Luby’s that got shot up by the loonie back in '92? How all the pro-gun folks argued so vociferously that if one of the poor victims had been able to carry, the crime would have been stopped in its tracks? Nonsense, because nobody actually carries the damn things on even a semi-regular basis. Thanks for making my point so effectively, ExTank.
Relax, man. I’m a gun owner myself, and so (obviously) are a bunch of my friends. Hell, I might have bumped into you down at Red’s a couple weekends ago. I just think the argumentss of the pro-gun crowd tend to be exaggerated beyond all reality. Case in point: the oft-repeated OP argument that the police have no duty to protect citizens. Buttressed, I might add, by a single case from New York in 1968, which stands for the simple proposition that the police will not be liable in tort to a crime victim who gets hurt when the police aren’t there. Color me unconvinced.
ExTank replied to me: * Careful there, Kimstu: “A restriction of a right is not the same thing as an infringement of it!”
True enough in essence and in fact, as of now. *
Praise the Lord, a gun advocate who understands this point! Now could you please explain it to SPOOFE?
But at what point does one become the other? Who draws that line, and on what basis? Does Barbarian draw that line […] Or does Tejota […]?
A very good question. There’s no doubt that certain kinds and degrees of restriction can be sufficiently crippling to count as rights infringements (e.g., poll taxes and voter literacy tests, etc.). But all rights are defined so as to balance individual liberty against other compelling interests of society. It seems to me that in this case (though IANAL), society’s particular “compelling interest” is in being able to make sure that a person attempting to buy firearms is not legally unfit to do so. The extent of the necessary restriction, therefore, ought to be set according to the time requirements for conducting a truly effective and thorough background check. Enforcement officials claim that that time scale is on the order of thirty days, which doesn’t sound implausible to me (although I see no reason why we couldn’t work on developing technology and record-keeping improvements to shorten it in the future). Therefore, a thirty-day waiting period is currently a reasonable restriction, not an infringement.
So no, SPOOFE, this is by no means a “semantics game”. There really is no such thing as an absolute or unrestricted individual right. All rights are restricted by the needs of other compelling interests in society, and no rights-bearer is permitted to ignore those interests in demanding an expanded interpretation of his/her own right. It is far too simplistic to try to reduce the question to a mere assertion such as “If I CAN’T bear arms, for any significant length of time, then my rights have been infringed.” Nope; they’ve been restricted, sure, but if society and the law have defined those rights to include such restrictions on the basis of another compelling interest, then they’re not being infringed. For example:
SPOOFE: *There’s a difference between making sure psychotic murderers aren’t allowed to have guns and keeping guns out of the hands of law-abiding citizens. *
Exactly. And the difference is that after background checks are run on both of them, the law-abiding citizen is free to purchase a gun and the psychotic murderer isn’t. That’s because the law-abiding citizen possesses the right (although not an unrestricted right) to keep and bear arms, whereas the psychotic murderer has forfeited his. Nobody’s rights are being infringed in either case. Sounds like a pretty good system to me.
(And anyway, aren’t gun advocates always railing about the importance of limiting criminal access to guns rather than banning the guns themselves? I would think that the more you see the main issue as one of preventing criminals from getting guns, the happier you’d be to see thorough background checks in place to screen out the criminals.)
Back to ExTank: *Having read Kleck/Gertz, Hemenway, Cook & Ludwig, Lambert, et. al., I can only conclude that I need at least an Associates in Statistical Math just to understand what they’re trying to say, much less glean any comprehensive meaning out of their works. Duelling statistics by people who really don’t understand them is a meaningless debate. *
Yes, but duelling statistics by people who do understand them can contribute a lot of clarity to the debate. Maybe it’s just the aftereffects of my math major, but the impression I got from the Kleck/Hemenway/Smith discussions was not so much that the statistics are terribly difficult as that not enough data is currently available. That is, as Smith points out, we don’t really know what people’s biases are in reporting DGU, we haven’t seen a properly selected sample space, etc. etc. More study is needed, unquote.
*As is usually the case in such instances, I’d not be suprised to find that the truth lies somewhere in between, as you, and Mr. (Prof.? Dr.?) Tom W. Smith points out. *
(Dr. Tom Smith, Director of the General Social Survey of the National Opinion Research Center at the University of Chicago.) Neither would I, but the results are hard to call at present, which is why in my original post I said that the numbers that disputants throw around so casually should be regarded very cautiously. I personally don’t think the evidence can currently support a stronger statement than, say, “Studies indicate that the number of annual DGUs in the US may be less than 100,000 or as high as a few million.” The truth may well lie somewhere in between, but that doesn’t mean we’d be justified in taking the average of the low-end and high-end estimates and thinking we’ve somehow achieved a more reliable number! Of course, that’s the math major in me coming out again.
The day George Hennard drove a pickup truck through the front of the Luby’s cafeteria in Killeen, Texas.
He killed 23 people, wounded 20 more, and took his own life.
I was eating lunch with my section sergeant at Denny’s less than a block away when it happened.
Talk about the goose walking over your grave!
Minty: strictly IMHO, there is no effective defense against totally random violence. No one (well, hardly no one) can maintain the level of vigilance necessary to spot a totally random nut-job. The Israelis probably have the most experience with random violence, and even they are largely helpless except for retaliatory strikes.
But your average citizen is probably aware of when they are entering a higher-risk area (can you say south Dallas? Oaklawn?) and adjust their habits accordingly. At such times, I gladly endure the inconvenience. I’ve been scoped several times, as I’m not the tallest or baddest-looking mo’ fo’. Turning slightly to reveal the holstered pistol at my side has probably convinced at least two people, and one group (4-5) to leave me alone.
Maybe they wouldn’t have done anything in the first place; I’m not a mind reader. I’m also not a frothing paranoiac grabbing my gun and blasting away when someone looks at me funny. Funny thing, neither are the CCW permitees in Texas (do you sense a “yet” coming?).
I was aware of my environment; of the people in it, and their actions. My awareness most probably showed in my demeanor. That’s about the best that can be hoped for against the average criminally-minded type, until they decide to act.
Barbarian: you obviously weren’t watching The Learning Channel last Saturday, where one of those reality video shows showed exactly that very thing. A loopy citizen was threatening a cop from a second story window with a handgun. I was just saying to the other Dopers (UncleBeer, thinksnow, Falcon, Jophiel, Wierddave, I’m definitely forgetting someone) “I have a scoped Winchester [rifle] that’d take that pistol clean out of his hand”, when a police sharpshooter took the pistol right out of the loony’s hand with a well-placed shot.
Hey Tank, good to see you. I was hoping you would stop by.
No, of course not. I didn’t phrase that very well, which Is why I re-stated the same idea in different terms later. It goes without saying that there is and objective reality, and that some fraction of the respondants to the study gave mistaken or outright lying responses to the survey.
One problem here is that people are more likely to mis-report that they used a firearm in self defense than that they didn’t. Human nature just works that way. Use of a gun in self defense is a memorable event, so it isn’t very likely that someone would use one would then forget when the survey comes around. On the other hand, It IS quite a bit more likely that who has a gun but didn’t use it defensively will say that he/she did.
Studies show that if you ask someone if they did something ‘within the last year’ and really they did it 18 months ago, they are likely to say yes anyway. This further inflates the positives.
So the number of false-positives is likely to be larger than the number of false negatives. And in the Kleck study, this is really harmful because false positives have a dramatic effect on the result while false negatives don’t. (This is a normal statistical effect seen whenever a survey tries to measure a rare event).
It’s debatable whether or not they think so. But it isn’t debatable that they havn’t.
So, yes. I am accusing Kleck & Gertz of either incompentence or deliberate bias. Their numbers just don’t stand up to analysis by an honest statistician. They not only didn’t control for self-reporting bias, but they encouraged it by gathering most of their data from the west and south. No, I’m not saying that the west and south lies more. Only that it biases the study by inflating the gun use figures, which are then extrapolated to the country.
I personally believe that they don’t want and honest survey, but even if you assume that they are honest, then all you learn is that they aren’t very good statisticians. Here is the analysis of their work that shows the problems. (This is the same as Kimstu posted earlier) The math is a bit dense, but even if you skip over it and just read the text, the ideas come thorough.
Their numbers are almost certainly inflated, easily 10 or even a 100 times larger than reality. Now, there isn’t any way to know how far off they are, but let me give you some examples (taken from the above link).
Lets say that 10 people in a thousand don’t respond accurately. Some who used guns said no, some who didn’t said yes. (We know, from other studies that 10 inaccurate answers out of 1000 is fairly normal). 10 errors in 1000 makes the Kleck study wrong by as much as a factor of 4, which is pretty bad. Their lower bound of 700,000 becomes less than 200,000, while the upper bound of 2.5 Million is virtually unchanged.
Ok, now assume that we get only 3 more mistaken answers, 13 out of 1000 answer wrong. Now the Kleck study is wrong by a factor of as much as 10. That is, the lower bound becomes 70,000. With only 14 mistaken anwsers the lower bound becomes 0!! Based only on the data that Kleck gathered, it is possible to set an upper bound with high accuracy, but impossible to set a lower bound at all without assuming that peoples answers to a survey are nearly always accurate!
When measuring rare events, the lower bound is always MUCH more sensitive than the upper one.
So, let me ask you. How many NRA members out of 1000 whould be willing to lie on a survey in order to promote the case for defensive gun use? Do you honestly believe that that number is less than 1 in 1000? 1 in 100? I’d personally guess at somewhere between 1 in 100 and 1 in 10.
And that is why the Kleck study is toilet paper. Basically they have shown that defensive use of firearms occurs somewhere between 0 and 2 million times a year. somewhere between an unreasonably small number and and absurdly large one.
No, I havn’t read Lambert recently.
If the NET savings is one in a thousand. Then I’d agree. But to get an actual ‘net safty factor’ for guns you would take defensive use then subtract loss of life by accident. Then somehow adjust for the fact that a major source of guns used by criminals is to steal them from legal owners, if you take guns away from citizens, then you stop a major source of supply to criminals.
So far no study has even come close giving us that number.
ExTank, you’ve always seemed to me to be one of the most mature and reasonable posters in favor of defensive gun use. I don’t have much chance, for instance, of changing SPOOFE’s mind until he stops thinking with his testes. But I think I could change your mind if I could get you to understand enough of the math to know who is playing straight and who isn’t in those studies.
p.s. I always seem to be popping into gun debates on the anti-gun side, which bothers me a bit. Because I’m only really anti-handgun and anti-concealed-weapon. I’d love to someday see a gun debate where the nuts who are making spurious and emotional arguments are banned and we can talk rationally about what an optimal gun policy would look like.
tj
Tejota: *I’d love to someday see a gun debate where the nuts who are making spurious and emotional arguments are banned and we can talk rationally about what an optimal gun policy would look like. *
Oooh, could I play too? I’d be very interested to see such a discussion! Maybe we could start it off with a stiff barrage of complicated statistical talk and get down to more general proposals once we’d frightened the nuts away.
On that note, that was a good summary of Hemenway’s objections to Kleck/Gertz, but there are a few counter-arguments to the counter-arguments. (By the way, I originally said that the link that you and I both posted also contained Kleck/Gertz’ response to Hemenway and Smith’s overview of both of them, but in fact they’re at different URLs—sorry about that.) Here’s the full list of links:
The original 1995 Kleck/Gertz article
Hemenway’s critique
The Kleck/Gertz response
The Smith summary
The major problems that led Smith to consider that the whole issue remained as yet undecided:
It’s not clear exactly what the “social desirability effect” would be, or how strong. Yes, Hemenway makes a good point, as you note, that people who like guns would be apt to exaggerate their DGU value, giving a strong trend toward overestimation. But K&G counter that most respondents’ DGUs were actually part of illegal activity, which they would be shy of reporting, leading to underestimation. Smith thinks this objection is rather implausible but notes that there isn’t actually any reliable data on how respondents do regard their DGUs.
Far lower estimates of annual DGU (<< 100,000) are extrapolated from the National Crime Victimization surveys; but these are likely to be too low, since they are derived only from victims’ reports of crimes and don’t include every type of crime, nor do they specifically probe for DGU.
And another important question, IMHO, is the type and frequency of the different crimes for which DGUs are claimed to be a deterrent. Just because you used a gun to deter a criminal doesn’t mean that you needed a gun to deter a criminal. As minty green commented, there are many situations where you could scare off a potential trespasser or vandal by yelling “Hey, get away from my bike!” or “Get off my lawn or I’m calling the cops!” just as easily as by waving a gun at them. These are definitely not in the same league as situations where you use a gun to face down a would-be rapist or shoot an armed intruder. We need much better info about what kinds of defense guns are actually being used for, as well as how often they’re being used, before we can make intelligent policy on gun ownership for self-defense.
Thanks for the rest of the links.
It was the Kleck reply that convinced me that Kleck/Gertz are liars rather than fools.
The first third of that reply is an ad-hominum attack on
Hemenway. They also claim that Hemenway ‘libelled’ them. But he did nothing of the sort. He never once implied that the cooked their data, only that they interpreted it wrong. In my experience, the only people who think they are being called a liar when someone points out their mistakes are those who actually are liars.
Also, they claim that DGU’s are underreported because most are connected with crimial behavior!! Smith points out that their data doesn’t support that idea. And even if it did, the mathematical criticism of their conclusions is unaffected.
Not to mention that DGU’s by CRIMINALS aren’t something we need to be supporting.
Kleck/Gertz perform a little sleight-of-hand in their criticism of Hemenway’s substantive point by arguing that the number of false positives and false negatives cancel each other. When they give an example of this, they make the number of false positives and false negatives exactly cancel. and show that their results are unaffected. Well duh!. Of course if they exactly cancel the result is unaffected!! But if the number of false positives is only SLIGHTLY larger, then their results go out the window.
The have only one (potentially) valid point to make in this sea of mis-information and personal attacks. The point out that Hemenway never mentioned a NIJ study that they claim produced figures similar to theirs. Their implication seems to be that Hemenway didn’t mention it because it shot his argument to pieces. He could just as easily thought it to be irrelevant or broken in the same fundamental way as Kleck/Gertz. I imagine if it was STRONG support for them, they would have done more than mention it, but I havn’t looked at it (yet).
All in all, their reply a classic example of the fact that liars use statistics.
tj
I’m sorry that evidence scares you.
That’s right Spoofe. They were so “scared” by your evidence that they discussed and analyzed it in specific detail, then found it duplicitous and/or inadequate.
[Cartman]
Man, you guys are a bunch of pussies.
[/Cartman]
Speaking on behalf of simplistic morons, I’m pretty sure that my mildly facetious comment was not an ‘a-ha’, just some non-statistically grounded food for thought.
I will forever be amazed at the mentality that is convinced that possession of a firearm is the answer to the ineffectiveness or indifference of a police force whose interest in public safety (at least in the U.S.) is limited to the decimal place in their paychecks.
While it may be the opinion of the courts that the police owe no duty to protect your more hapless countrymen (or at least those too dopey to have the sense to have their own shootin’ irons), I believe - without statistical backing - that you’d have a hard time selling that argument to those who are under the impression that they are responsible for law enforcement (poor dumb bastards).
Tank, I gotta admire your almost evenhanded treatment of some of the issues raised, and I am likewise completely in line with the necessity of awareness as the first tactic.
I hafta say that I am disappointed that you trot out the ‘slippery slope’ defence to drawing lines between restrictions and infringements. All that amounts to is the sticking point for those who realize their arguments may be more tenuous than they thought.
And wait just a damn minute! Wearing a helmet is an infringement of my God-given right to be concussed if I so choose. And while it may not be Constitutionally enshrined, it wouldn’t look out of place as an amendment…
z
Isn’t there just a little bit of teenage fantasy in the idea of heroically defending one’s babe from a “fate worse than death”? Or standing alone against the hoards of ill doers? 
C’mon, guys.
I live in a 'hood that would scare the shit outta most people, day or night, and I haven’t yet had the need to defend myself beyond looking mean. And I’ve lived in worse. Much worse.
Somebody did steal my Pep Boy’s hub caps once.
Fifteen bucks for the set. I was devastated.
I do do pretty much as Ex Tank suggests, awareness and all that. But I’m never armed.
Don’t have a wife and kids now. Maybe that’s it.
Peace,
mangeorge
I will forever be amazed at the amount of bleeding-heart rhetoric that is spewed by people who don’t like guns.
Please educate me, O Wise One, how a gun, in the hands of someone properly trained and educated in it’s use, can not possibly be an effective defense against one who wishes you harm.
**Tejota:**before I get started, I’d be interested in hearing your logical objections to specifically handguns and concealed carry.
From the Hemenway critique (III. The Kleck-Gertz Survey)
How does this method yield a preponderance of respondents from the South and West? Considering that, demographically, these areas probably have higher rates of gun ownership but a much more sparse population. Oh, that’s right: they fabricated results.
He [Hemenway] then turns around and says just a few paras. later:
If it was a random digital-dial mechanism, how can he claim that a "high percentage of initially selected homes (implying that the interviewed were less-than random) weren’t interviewed. 5,000 people are 5,000 people. If it takes 15,000 phone calls to get 5,000 responses (10,000 “busy” or “no comment” and 5,000 respondents) how does that significantly change the results?
And while on the subject of numbers of respondents, NCVS surveyed “50,000 housing units” over a longer period of time. While kudos to them for using a higher number over alonger period for sampling purposes, what’s the statistical difference between interviewing 50,000 “people” and 50,000 “housing units.” Not everyone in a “housing unit” will own or otherwise handle a gun (my step-dad still has his service revolver, but my mother refuses to even look at it.) So how does my mother answering the phone during a K-G style survey skew the results than if my step-dad answered? If there were a DGU in their houshold (by my step-dad) but my mother answered “NO”, wouldn’t that be a fasle negative?
It seems to me that interviewing everyone in a given household, and then counting DGUs by household, would skew the results downward artificially, as not everyone in a given household has access to the owner’s (head of household’s) firearm; and that not everyone in a household will have a DGU.
If your average household has a mother, father and two kids, how does the inclusion of the kids (not likely to be involved in a DGU) contribute to an accurate count of DGUs?
Silly me. And here I had been told (right here on SDMB!) by another trained mathematician knowledgeable of statistics and survey methodology that just about any 1,500 respondents in a random dial-up survey (assuming good regional diversity) will yield a statistically representative sample of the population. The issue arose over a political poll (a CNN/Gallup, IIRC) where I essentially asserted that the poll was skewed by a preponderance of respondents in urban/metropolitan areas. That segued into survey methodology, and the statement above.
Only adult in the household? Or only adult in the household at the time of the call? Adults? Households? WTF? Who lives in “households”? Correct me if I’m wrong, but I was generally under the impression that “people” (you know, “individuals”) lived in “households.” That the “national population” was comprised of “people” as, in 9 points out of 10, the Bill of Rights guarantees the rights of “people,” not “households.”
Since several people here claim to be familiar (if not experts) with statistics, I await an explanation. In English, please.
Considering the contradictory claims of the Hemenway critique, K-G may not be the only ones fabricating results.
Approximately the same percentage of HCI members who routinely trot out Kellerman’s flawed “43 Times” figure or HCI’s seriously cooked “13 Children A Day” [die from handguns] figure.
On the surface, the NCVS seems to have used a more in-depth survey method (more people, longer period). My real problem is that it was conducted on behalf of the most antigun administration in the history of this country, an administration that showed unabashed bias towards HCI and like groups, and was downright hostile to the NRA and pro-gun groups; an administration that jumped all over “43 times” and “13 children a day” and danced on the graves of every fresh victim to shamelessly use emotion rather than reason to promote their antigun agenda.
So it seems that pro-gun groups and surveys aren’t the only ones capable of bias, emotional appeals and outright fabrication.
As Prohibition and the War on Drugs have amply shown (and I have stated repeatedly ad nausea) that any proscribed commodity not available domestically will be provided for through foreign sources. I can easily see the cocaine cartels in Colombia getting into the gun smuggling business as well shortly after any type of firearm is banned.
Zoony: not being Canadian, I couldn’t care less what your constitution guarantees you a right to; considering that this is a debate about American gun control, your Canadian POV is largely irrelevant. If you want to start a thread about gun-control on the international scale, then by all means, I will listen and debate respectfully with you on that issue (that is, until you say something really stupid; then I will slam you mercilessly. Nothing personal )
Oh, and the “slippery slope” really depends upon your POV. Standing at the top 'o the heap looking down, the slope probably doesn’t seem so slippery. When you have a dog in this fight then maybe you’ll look slightly askance at every new gun-control scheme that comes along.
I may not toe the NRA party-line, but I am in agreement with them when they say that gun control does not equal crime control. Not being a sociologist or criminologist, I can’t speak with authority on the root cause of crime or criminal behavior, but I strongly suspect that the issue has more to do with civil rights, economic disparity, educational disparity, (perceived and/or actual) disparity in employment opportunity, conflicts of subcultures (or cultural isolation), and so on. I feel that addressing and resolving these issues will go a lot further towards reducing violent [firearm] crime than gun control or bans.
I am also in agreement with the NRA when they say that a more routine and evenhanded enforcement of existing gun-control laws is in order before the efficacy of the existing laws can be accurately assessed. As each unenforced law fails to reduce violent [firearm] crime, each new law, further restricting law-abiding gun owners, making us jump through more-and-more bureaucratic and administrative hoops, redirects the onus away from criminals and onto law abiding citizens.
And let’s face it: there is a gun culture in America. It is composed of hunters, police, military, target shooters, recreational shooters, and collectors.
Liberal/urban dwelling types may not like it; they may find it barbaric, unseemly, or distasteful. They may see a bunch of “gun worshiping maniacs,” “racist rednecks,” “hick goobers” “beered-up bubbas” or “limp-dicked wannabes, stroking their substitute phalli.” So be it. That’s their opinion, and their welcome to it.
I personally don’t care for the “Gangsta” Rap culture, or the Hollywierd culture that produces and promotes extremely violent films glorifying criminal behavior, and then turns around and attacks gun rights and gun owners. But I sorta respect that 1st Amendment thingy in our BOR. So all I ask is: please respect ours.
I often find it amusing (and insulting) that people who have an avowed dislike of guns, who have never owned a gun, never fired a gun, know nothing about their operation, their myriad types, or even of the many existing different gun-control laws currently on the books pipe up with ad hominem rhetoric about gun owners and the “gun culture.” I was recently reminded of a post I composed about a year ago, addressing this very issue. It can be found here, in a post dated 03-17-00, at 5:43 PM.
Hello. Meet the true “gun culture”; you’re talking to them.
Now, to some commonsense suggestions and restriction I think can go a long way towards reducing the number of firearms and handguns hitting the street, and towards reducing further the number of accidents (which are pretty damned low anyway) without placing undue burden upon law-abiding gun owners:
[ul]
[1.] Don’t ban private transactions, but “flag for investigation” any non-licensed person engaged in regular/routine purchases from a licensed dealer. A person buying several guns a week, every week, may just be building an extensive collection. They may also be reselling on the side to others who cannot legally purchase a firearm. A quiet look-see at these individuals may help reduce “strawman” buyers.
[2.] Mandatory training. Compile a diverse list of groups and organizations engaged in basic firearm safety training, and make the sale of a firearm (at a licensed gun store) contingent upon producing a current “card” from one of these organizations showing completion of a basic safety course. Also: make it so the card must be current while in possession of a firearm. The police have no right or obligation to check up on these, but may legally ask to see one if, for some reason, they enter a person’s home while responding to a call (say, responding to a 911 call, or somesuch). Make a first-time offense a fine (say $500 dollars or equivalent hours community service) and violators subject to subsequent follow ups. Multiple offenses (three-strikes-you’re-out?) could be made a minor felony, to revoke their right to keep and bear, as they have demonstrated themselves to be unsafe.
This is as close to the “classic” mode of “licensing” as I dare skirt, and would adamantly oppose (to the point of civil disobedience!) any further government control of a licensing process.
[3.] Mandatory criminal (and possibly civil) liability for inadequate storage, in the event of theft or accidental shooting. Be common sensible about it; if you go away for vacation, and come home a week later to find an empty house (including the 600 pound gun safe in the basement), then no liability would ensue. But leaving a loaded firearm in a dresser or nightstand drawer, or in a shoebox in the closet, is unacceptable.
[4.] Self and Dealer Registration. Require all dealers to maintain a written record, by make, model, caliber and serial number, of all firearms bought and sold. The ATF form 4473 should suffice. The form must be maintained for the life of the firearm, inasmuch as possible. Also create a form for owners to keep and maintain as well. Make the forms for this “free access” from any gun store, the post office, city/county offices, police stations, even downloadable from the internet. Owners keep and maintain these records, and will be required to note the name, address and particulars of any private transaction on the form, including dates and such.
If any law enforcement agency has any question concerning a firearm, the dealers and owner will be required to produce the form for that firearm. This will aid law enforcement in the tracing of firearms used in crimes, and provide an overall better picture on the flow of firearms from legal to illegal hands.
This is as close to the “classic” mode of “firearm registration” as I dare skirt, and would adamantly oppose (to the point of civil disobedience!) any further government control of a registration process.
[5.] Mandatory Safety Training in both public and private schools. At grade, middle and high school levels. This will help teach kids that guns are not toys; that, unlike Hollywierd, when a person gets shot, it has real and devastating real-life consequences. Have real-life victims of gun violence (preferably people who are permanently crippled or maimed) come in for guest lectures, and some extremely realistic film footage of victims of gun violence (like the gross-out driver’s-Ed. film). Touch upon basic operating principles with non-op firearms so that if a kid finds a gun, they don’t accidentally discharge it through ignorance.
[6.] A drive to encourage firearms manufacturers to design and produce firearms with keyed internal locking mechanisms, and easy-disassembly designs to allow removal of firing pins. [I personally don’t put too much stock or reliance in mechanical safeties; the best safety is a trained, conscious human being who thinks safety, **acts safely** and is therefore **safe.** But to err is human, so I concede this occasional necessity. -ExT.][/ul]
I think that this is enough debate material for now.
ExTank, you da man (and so are Tejota and minty green, IMHO). This is the best gun debate I’ve ever seen or heard of. On to some serious intelligent responses to your serious intelligent points:
How does this method [in the Kleck/Gertz survey] yield a preponderance of respondents from the South and West?
Hemenway’s claim about K&G oversampling respondents from the South and West (and also males) is made, according to his footnote #20, on the basis of the original K&G article, p. 161. Going there (and briefly thanking heaven for descriptive markup in electronic text encoding that includes page numbers :)), I spotted the sentences that I think Hemenway is referring to:
So K&G deliberately oversampled in the south and west, for the sake of getting more raw data on DGUs. Admittedly, they do say that “Data were later weighted to adjust for oversampling”, but that leads us to the question of their data weighting mechanisms, which are part of what both Hemenway and Smith were objecting to on methodological grounds in their paper.
*He [Hemenway] then turns around and says just a few paras. later [? earlier?]: “The article states that when a person answered, the interview was completed 61% of the time. [16] But what happened when there was a busy signal, an answering machine or no answer? If no one was interviewed at a high percentage of the initially selected homes, the survey cannot be relied on to yield results representative of the population.”
If it was a random digital-dial mechanism, how can he claim that a "high percentage of initially selected homes (implying that the interviewed were less-than random) weren’t interviewed. 5,000 people are 5,000 people. If it takes 15,000 phone calls to get 5,000 responses (10,000 “busy” or “no comment” and 5,000 respondents) how does that significantly change the results? *
Hmm, this is not addressed very clearly in any of the articles, but I don’t think the phrase “initially selected” means that the selection wasn’t random. It seems to mean that if the pollsters didn’t get an answer at one number, they moved on to another. There’s nothing wrong with that except that it tends to bias your sample space against the people who aren’t home at the time you were calling: e.g., if you call during the day, you end up underrepresenting the households where everybody spends the working/school day outside the home. The way to get a more “random” sample (and of course, it’s still biased towards the folks who have phones, anyway) is to pick your 5,000 phone numbers by true random-digit selection and then to keep calling those 5000 numbers till you get them.
*[Hemenway:] “Interviewers do not appear to have questioned a random individual at a given telephone number, but rather asked to speak to the male head of the household. [17] If that man was not at home, the caller [Page 1434] interviewed the adult who answered the phone. [18] Although this approach is sometimes used in telephone surveys to reduce expense, it does not yield a representative sample of the population.”
Silly me. And here I had been told (right here on SDMB!) by another trained mathematician knowledgeable of statistics and survey methodology that just about any 1,500 respondents in a random dial-up survey (assuming good regional diversity) will yield a statistically representative sample of the population.*
But but but—that assumes 1500 random respondents, yes? If the survey is deliberately selecting for male “head of household”, and querying another adult only when that male isn’t available, obviously the results are going to be way skewed towards male heads of households, not equally representative of the adult population overall.
*[Hemenway:] “The 2.5 million estimate is based on individuals rather than households. [19] But the survey is randomized by dwelling unit rather than by individual, so the findings cannot simply be extrapolated to the national population. Respondents who are the only adults in a household will receive too much weight.”
Only adult in the household? Or only adult in the household at the time of the call? Adults? Households? WTF? Who lives in “households”? Correct me if I’m wrong, but I was generally under the impression that “people” (you know, “individuals”) lived in “households.” That the “national population” was comprised of “people” as, in 9 points out of 10, the Bill of Rights guarantees the rights of “people,” not “households.”
Since several people here claim to be familiar (if not experts) with statistics, I await an explanation. In English, please. *
I’ll do my best! Okay, the technique of the survey was to pick 5000 random households, because generally speaking, 1 household = 1 phone number (although these days what with the kids having their own phone and the extra line for the fax machine and the home office and so forth you can’t count on that, but I digress). However, once a household was contacted, the respondent was asked if anybody in the household had had a DGU. Now if there’s no other adult in the household, “yes” or “no” for the respondent is “yes” or “no” for the survey: on to the next household. But, if there are other adults in the household, that effectively gives the respondent “another chance” at a positive response for a DGU. Since it’s not reliably true that “1 household = 1 individual”, you can’t extrapolate the results to individuals in the population at large.
See, imagine you decided to take a random survey of the population to estimate DGUs, by asking survey questions of everybody who comes into the post office one morning. Assuming that that sample is representative of the population as a whole (which I’m sure it isn’t, btw, but let’s leave that aside for now), you would be justified in concluding “x% of this representative sample report a DGU, so it is likely that x% of the population at large have also had a DGU.”
Okay, but now imagine that for, say, a quarter of the survey respondents, you don’t just ask about their own experience, but say “Well, how about your brother? Did he have a DGU? Or your father? Or your uncle?”, and if the respondent says “yes” to any of those, you mark it down as a positive. All of a sudden, those respondents are not really individuals anymore, they now essentially represent clusters of individuals, giving them a greater chance at a positive answer. That’s the problem with randomizing your data collection by dwelling place but extrapolating your results to individuals. There, did that help? Was it English?
Considering the contradictory claims of the Hemenway critique, K-G may not be the only ones fabricating results.
Mmm, I don’t think so: it’s certainly very difficult to keep track of all the different factors properly, and I think Smith is probably right that neither K&G nor Hemenway cover all the bases in a truly conclusive fashion, but I don’t think the Hemenway claims that you mentioned are actually contradictory.
*“So, let me ask you. How many NRA members out of 1000 would be willing to lie on a survey in order to promote the case for defensive gun use? Do you honestly believe that that number is less than 1 in 1000? 1 in 100? I’d personally guess at somewhere between 1 in 100 and 1 in 10.”
Approximately the same percentage of HCI members who routinely trot out Kellerman’s flawed “43 Times” figure or HCI’s seriously cooked “13 Children A Day” [die from handguns] figure. *
May well be, but those ain’t the numbers we’re talking about here: I’d be happy to see a sober analysis of misstatements from the gun-control side on another thread, but the subject now is guns and self-defense and the accuracy of the estimates concerning them.
*On the surface, the NCVS seems to have used a more in-depth survey method (more people, longer period). My real problem is that it was conducted on behalf of the most antigun administration in the history of this country, an administration that showed unabashed bias towards HCI and like groups, and was downright hostile to the NRA and pro-gun groups; an administration that jumped all over “43 times” and “13 children a day” and danced on the graves of every fresh victim to shamelessly use emotion rather than reason to promote their antigun agenda. *
That’s all very well, but as Smith points out, the NCVS survey was actually conducted by the Bureau of the Census for the Bureau of Justics Statistics, and it continued the BJS’s project of data collection on this same subject that has been carried out by them since 1973. If you have specific evidence that the 1992–1994 survey “on behalf of” the Clinton administration is unreliable due to having been tainted by gun-control propaganda, feel free to produce it; otherwise, I don’t think vague allegations of some kind of “guilt by association” are going to cut it.
*Tejota: “Then somehow adjust for the fact that a major source of guns used by criminals is to steal them from legal owners, if you take guns away from citizens, then you stop a major source of supply to criminals.”
As Prohibition and the War on Drugs have amply shown (and I have stated repeatedly ad nausea) that any proscribed commodity not available domestically will be provided for through foreign sources. I can easily see the cocaine cartels in Colombia getting into the gun smuggling business as well shortly after any type of firearm is banned. *
Um, though I don’t advocate total bans of firearms, I don’t think your analogy holds up. To say that legal prohibition won’t eradicate use of the prohibited item is not the same thing as to say that legal prohibition can’t reduce it, as in fact drug prohibition does reduce drug use (although not as much, IMHO, as drug prohibition plus an intelligent treatment-and-rehabilitation plan would do, instead of the meaninglessly punitive War on Drugs we’ve got now). Certainly, because of the laws of supply and demand, illegal suppliers will always expand to help fill the vacuum caused by removal of legal suppliers; but because they’re illegal, the job can be made much harder for them. It was more difficult, even for many criminals, to obtain alcohol during Prohibition than it is now; it would be more difficult, even for many criminals, to obtain guns if all guns were banned than it is now. I don’t say that that means it would be good to ban all guns, but we won’t make better policy just by ignoring the realities of the situation.
*I may not toe the NRA party-line, but I am in agreement with them when they say that gun control does not equal crime control. Not being a sociologist or criminologist, I can’t speak with authority on the root cause of crime or criminal behavior, but I strongly suspect that the issue has more to do with civil rights, economic disparity, educational disparity, (perceived and/or actual) disparity in employment opportunity, conflicts of subcultures (or cultural isolation), and so on. I feel that addressing and resolving these issues will go a lot further towards reducing violent [firearm] crime than gun control or bans. *
I’m behind you all the way on this one. I think that many gun advocates greatly damage their credibility by maintaining aggressively punitive positions on criminal justice. Pace people like TurboDog, vicious treatment of criminals does not produce an appreciably safer and more just society (see, e.g., the recent Harper’s article about the horrific Sheriff Joe Arpaio of Phoenix, AZ), and gun advocates who support vicious treatment of criminals just pump up the “wacky gun nut” stereotype.
*Now, to some commonsense suggestions and restriction I think can go a long way towards reducing the number of firearms and handguns hitting the street, and towards reducing further the number of accidents (which are pretty damned low anyway) without placing undue burden upon law-abiding gun owners: *
All these sound good to me, particularly #3, and with reservations about #5. “Mandatory” gun safety training in schools is too draconian for the numerous parents who prefer that their children not be involved with guns to any extent whatsoever, not even in safety training. After all, if parents can opt their kids out of sex ed on account of moral objections, they should be able to do the same with gun ed.