# Studying Firearm Related Violence **URL:** \ **Category:** Great Debates\ **Created:** [June 4, 2014, 4:48am UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994 "2014-06-04T04:48:21Z")\ **Posts on this page:** 20\ **Page:** 3
**Author:** ![Bone](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bone/32/407_2.png) [@Bone](https://boards.straightdope.com/u/Bone)\ **Post date:** [June 6, 2014, 1:19pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/41 "2014-06-06T13:19:18Z")
> [@steronz](#): > > To reuse my hypothetical, if the 5,000 people who were armed each shot their armed mugger, we’d have 5,000 justified homicides, but the 5,000 mugging victims who did not shoot anyone because they were unarmed completely cancel those out, revealing no net advantage to guns. Yes, there are certainly some actual situations where guns save lives, but you can’t look at any one anecdote and add it to some column, because none of us are psychic. You have an odd way of evaluating whether there is a ‘net advantage to guns’. In your example, you have 5000 armed people who defended themselves from being mugged because they shot their armed mugger. You have 5000 people who are unarmed and fell victim to their muggers. You call this no advantage to having a gun? Not getting mugged seems like a pretty big advantage. > [@](#): > > Yes, we must disregard them, because they tell us nothing. It’s not a numerator, or a gross number, or however else you’re characterizing it. It’s a fantasy, pure speculation. Not sure what you’re saying here. Are you saying that we can’t know how many or the nature of DGUs and therefore they must be treated as zero? Or are you saying that DGUs are a fantasy and that they don’t occur? Or something else? > [@Really\_Not\_All\_That\_Bright](#): > > No. If you can show me a method other than self-reporting I’m all ears. > > Not very, so it’s probably a good thing that the surveys aren’t done by the DOJ. That, and the fact that the NCVS doesn’t ask whether the use was legal. Do you discount all scholarship or study that is based on interviews or questionnaires? It seems like the self reporting nature of the NCVS is singled out in this instance. Here is a description of some of the [characteristics of the NCVS](http://www.nap.edu/openbook.php?record_id=10881&page=103): > [@](#): > > The NCVS is an ongoing annual survey conducted by the federal government (i.e., the Census Bureau on behalf of the Department of Justice) that relies on a complex rotating panel design to survey a representative sample of nearly 100,000 noninstitutionalized adults (age 12 and over), from 50,000 households.\*\* To elicit defensive gun use incidents, the survey first assesses whether the respondent has been the victim of particular classes of crime—rape, assault, burglary, personal and household larceny, or car theft\*\*—during the past six months, and then asks several follow-up questions about self-defense. In particular, victims are asked: > Was there anything you did or tried to do about the incident while it was going on? > Did you do anything (else) with the idea of protecting yourself or your property while the incident was going on? > Responses to these follow-up probes are coded into a number of categories, including whether the respondent attacked or threatened the offender with a gun. (my bold) [That is question #40 of the survey. It can be found about 3/4 of the way down here.](http://books.google.com/books?id=pdTAddcCkC0C&pg=PA128&lpg=PA128&dq=%22Altogether,+how+many+times+did+this+type+of+incident+happen+during+the+last+6+months%22&source=bl&ots=PoxQK7upaH&sig=0ak6vKDB81V8ZgOUOEwOyUBqrtM&hl=en&sa=X&ei=j7uRU8OAIorS8AGT5YGwCg&ved=0CCAQ6AEwAA#v=onepage&q=%22Altogether%2C%20how%20many%20times%20did%20this%20type%20of%20incident%20happen%20during%20the%20last%206%20months%22&f=false) So only actual victims of a subset of crimes are the subject of these additional questions to determine if there was a defensive gun use. Ultimately if you are willing to dismiss both this study, as well as every other study that has been done showing much higher values, would any evidence be satisfactory? If so, what would it look like? ---
**Author:** ![Really\_Not\_All\_That\_Bright](https://avatars.discourse-cdn.com/v4/letter/r/e8c25b/32.png) [@Really\_Not\_All\_That\_Bright](https://boards.straightdope.com/u/Really_Not_All_That_Bright)\ **Post date:** [June 6, 2014, 1:22pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/42 "2014-06-06T13:22:11Z")
I don’t know. Hence the “I’m all ears” bit. > [@Damuri\_Ajashi](#): > > By self reporting do you mean crackpots running up to researchers and saying “I’ve got a story to tell you” or answering questions in a survey that is conducted the way that surveys are supposed to be conducted? What is the DoJ doing wrong? how should they be conducting these surveys or do you think there is some better method of determining the right numbers? You seem to have misread the post you were responding to: “if you can show me a method other than self-reporting I’m all ears”. > [@](#): > > Fine, the interviews are conducted by the census bureau and administered by the DoJ (you aren’t suggesting that the census bureau doesn’t know how to conduct surveys, do you?). And are you saying that tens of thousands of people are admitting illegal firearm use to the government (whether census bureau or DoJ)? I am not suggesting anything about the census bureau’s surveying ability. I am pointing out an error in your post. And yes, there is no reason why they would distinguish between legal and illegal firearm use. ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [June 6, 2014, 1:34pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/43 "2014-06-06T13:34:02Z")
> [@Bone](#): > > There are two propositions here that are separate. The first is whether the CDC should be doing advocacy work. I do not think they should. Do you believe it is the CDC’s place to do advocacy work? The express purpose of the existence of the CDC is that it"conducts critical science **and provides health information** that protects our nation against expensive and dangerous health threats, and responds when these arise." So what it is doing in publishing reports like this is, in fact, a fundamental part of its purpose. ---
**Author:** ![steronz](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/steronz/32/11765_2.png) [@steronz](https://boards.straightdope.com/u/steronz)\ **Post date:** [June 6, 2014, 1:34pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/44 "2014-06-06T13:34:23Z")
> [@Bone](#): > > You have an odd way of evaluating whether there is a ‘net advantage to guns’. In your example, you have 5000 armed people who defended themselves from being mugged because they shot their armed mugger. You have 5000 people who are unarmed and fell victim to their muggers. You call this no advantage to having a gun? Not getting mugged seems like a pretty big advantage. Sorry, I try to be careful with my words but I slipped up there. I meant net _safety_ advantage, and by safety I mean life or serious injury. Certainly reducing muggings is a good thing, but if it’s at the cost of human life somewhere else in the system, it’s not a good thing. > [@Bone](#): > > Not sure what you’re saying here. Are you saying that we can’t know how many or the nature of DGUs and therefore they must be treated as zero? Or are you saying that DGUs are a fantasy and that they don’t occur? Or something else? Something else. I’m saying that we can’t know how many DGUs represent a life saved or serious injury prevented. If we’re trying to figure out how many lives are saved by seatbelts every year, we don’t simply count up the number of accidents in which the driver was belted and call it good, because we don’t know if the belt played a role in all of them; certainly there are unbelted people who survive, after all. ---
**Author:** ![Hentor\_the\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/h/8edcca/32.png) [@Hentor\_the\_Barbarian](https://boards.straightdope.com/u/Hentor_the_Barbarian)\ **Post date:** [June 6, 2014, 2:15pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/45 "2014-06-06T14:15:04Z")
> [@Really\_Not\_All\_That\_Bright](#): > > And yes, there is no reason why they would distinguish between legal and illegal firearm use. In fact, and I’m repeating myself on these boards, Hemenway used survey data on DGUs that included exposition from the respondents regarding the incident. He assembled a panel of judges (in the legal sense of judges) to evaluate the descriptions. Half of the DGUs in the sample were deemed illegal by the panel. [http://injuryprevention.bmj.com/content/6/4/263.full](http://injuryprevention.bmj.com/content/6/4/263.full) For example, this was a self-reported DGU in the survey: > [@](#): > > A 58 year old male was inside his home at 2 pm. “I was watching a movie and [an acquaintance] interrupted me. I yelled that I was going to shoot him and he ran to his car”. The respondent said his acquaintance was committing a verbal assault. The respondent’s gun, a .44 Magnum, was located “in my holster on me”. > > • A 62 year old male said that at 6 pm “the police called. My alarm at my business went off so I went there to shut it off. Two men were outside my building, so from my car I shot at the ground near them”. The respondent said the men were trespassing. These were of course deemed illegal. However, if all the survey collected was a self-reported yes or no, it would be counted as an instance of DGU. An interesting aspect of Hemenway’s paper was the point that, among survey respondents, guns were used far more often (a three to one ratio) to threaten people than they were used in legal self-defense. Examples of people being threatened were: > [@](#): > > • “I’m a cattle farmer and he’s a cattle farmer. He was putting his bulls and heifers up near the fence and my bull broke out and he got mad”. > > • “I was on a date. He pulled the gun when I mentioned breaking up with him”. > > • "I was mugged in New York”. Additionally, the context of some responses calls them into question on their face. For example: > [@](#): > > Over two thirds (68%) of the 146 self defense gun use incidents from the two surveys were reported by six respondents. Three people claimed 50, 20 and 15 self defense incidents in the previous five years, but refused to describe the most recent event. In the 1999 survey, an 18 year old male reported six cases. He described the most recent incident: “I was at school and they pulled a gun during an argument. They fired and I fired”. 50 self-defensive gun uses in 5 years! Does that person live in the OK Corral? ---
**Author:** ![Really\_Not\_All\_That\_Bright](https://avatars.discourse-cdn.com/v4/letter/r/e8c25b/32.png) [@Really\_Not\_All\_That\_Bright](https://boards.straightdope.com/u/Really_Not_All_That_Bright)\ **Post date:** [June 6, 2014, 2:25pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/46 "2014-06-06T14:25:47Z")
I’m not 100% sure but it looks as though [the NCVS](http://www.bjs.gov/content/pub/pdf/ncvs2_2012.pdf) is unaffected by “multiple defenders” because it only asks for details of the most recent incident. ---
**Author:** ![Hentor\_the\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/h/8edcca/32.png) [@Hentor\_the\_Barbarian](https://boards.straightdope.com/u/Hentor_the_Barbarian)\ **Post date:** [June 6, 2014, 2:36pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/47 "2014-06-06T14:36:05Z")
> [@Really\_Not\_All\_That\_Bright](#): > > I’m not 100% sure but it looks as though [the NCVS](http://www.bjs.gov/content/pub/pdf/ncvs2_2012.pdf) is unaffected by “multiple defenders” because it only asks for details of the most recent incident. That helps a little, but that’s not quite the point. The point is more about hearing the clock strike 13 times - which strike was the false one? (To be more clear for you, the validity of each individual one of those incidents are called into question by the overall volume of incidents.) Do you really believe that someone had cause to use a firearm in self-defense 50 times in five years? I don’t. So, which of the 50 incidents was actual and which were fantasy? Now, across the remaining individual incident reports, are any motivated by a similar process - a fantasy about being threatened or using a gun to ward off harm? ---
**Author:** ![Bone](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bone/32/407_2.png) [@Bone](https://boards.straightdope.com/u/Bone)\ **Post date:** [June 6, 2014, 3:52pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/48 "2014-06-06T15:52:06Z")
> [@Really\_Not\_All\_That\_Bright](#): > > I don’t know. Hence the “I’m all ears” bit. I’m not clear, do you not accept or discount all studies that use or rely on survey data where information is self reported? > [@RickJay](#): > > The express purpose of the existence of the CDC is that it"conducts critical science **and provides health information** that protects our nation against expensive and dangerous health threats, and responds when these arise." So what it is doing in publishing reports like this is, in fact, a fundamental part of its purpose. I don’t have any issue with the report that was published and linked to in the OP as a result of the Obama executive order. I did link to previous studies that were conducted through the CDC (funding at least) that were suspect. Do you think that violent crime and firearm violence should be treated like other diseases? That’s the approach they had taken in the past. I don’t think firearm and other violent crime are like diseases at all. This isn’t to say the CDC doesn’t have a place in examining data - I think they do. I’m not sure if you agree with that but if I think back at another example relating to health care. I used to think the government shouldn’t be involved in health care, and that traditional drivers of supply and demand would control the market. When I came to the realization that health care was not like other goods and that the market for it behaved very differently was when I re-evaluated my position. > [@steronz](#): > > Something else. I’m saying that we can’t know how many DGUs represent a life saved or serious injury prevented. If we’re trying to figure out how many lives are saved by seatbelts every year, we don’t simply count up the number of accidents in which the driver was belted and call it good, because we don’t know if the belt played a role in all of them; certainly there are unbelted people who survive, after all. Is it fair to say that conversely you believe that DGUs should not be treated as zero, and that they do in fact occur? In your hypothetical, if a mugger would only take a person’s wallet and not otherwise harm them, is it a DGU if a person shoots that mugger and the mugger dies? This is assuming perfect knowledge for the sake of sussing out definitions, in reality you’d never be able to know this. > [@Hentor\_the\_Barbarian](#): > > In fact, and I’m repeating myself on these boards, Hemenway used survey data on DGUs that included exposition from the respondents regarding the incident. He assembled a panel of judges (in the legal sense of judges) to evaluate the descriptions. Half of the DGUs in the sample were **deemed illegal** by the panel. [http://injuryprevention.bmj.com/content/6/4/263.full](http://injuryprevention.bmj.com/content/6/4/263.full) (my bold) Actually, they were deemed “probably illegal”. Courts, jury, you know. Let’s look at how that was done: > [@](#): > > Self defense gun use incidents were summarized and sent to five criminal court judges (from California, Pennsylvania, and Massachusetts) who were assured anonymity). So, states that have some of the most restrictive gun laws, anonymous judges. > [@](#): > > This research was supported in part by the grants from the Centers for Disease Control and Prevention, the National Institute of Justice, the Open Society Institute, and the Joyce Foundation. From the wiki: > [@](#): > > Since 2003, the Joyce Foundation has paid grants totaling over $12 million to gun control organizations.[1][13] The largest single grantee has been the Violence Policy Center, which received $4,154,970[13] between 1996 and 2006, and calls for an outright ban on handguns, semi-automatic and other firearms, and substantial restrictions on gun owners. This is how questions of bias come about. Surveys conducted by the Census Bureau , with the lowest estimates of more than 19 studies that estimated DGU, and all of it is suspect. But research by anti gun groups using anonymous folks is treated as totally valid. If the NRA conducted a study and came up with opposite results, would you be just as accepting? Not only that, this really does speak directly to the previous assertion that without CDC funding then gun research would cease. The budget eliminated from the CDC was less than $3M at the time. Here we have one group, Joyce, granting over $12M over a period of time. I accept that not all self reported incidents of DGU were legal, and people use firearms for nefarious purposes at times. But **ALL** studies point to large numbers ( \> 100K) of DGUs - from the NCVS to Kleck, to the report linked in the OP. > [@](#): > > Additionally, the context of some responses calls them into question on their face. For example: 50 self-defensive gun uses in 5 years! Does that person live in the OK Corral? That seems high. I would also think a single store owner wouldn’t be involved in four shootouts killing 5 different people over a period of four years. [Maybe it’s an outlier.](http://articles.latimes.com/1992-02-21/local/me-2663_1_watch-shop-owner) If an unarmed person starts trying to kick in my door, and through the window they see me display a firearm so they run off, no shots fired, no police report - do you consider that a DGU? If yes, how would that be captured by any study you’d accept? If no, why not? ---
**Author:** ![Really\_Not\_All\_That\_Bright](https://avatars.discourse-cdn.com/v4/letter/r/e8c25b/32.png) [@Really\_Not\_All\_That\_Bright](https://boards.straightdope.com/u/Really_Not_All_That_Bright)\ **Post date:** [June 6, 2014, 3:54pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/49 "2014-06-06T15:54:32Z")
> [@Bone](#): > > I’m not clear, do you not accept or discount all studies that use or rely on survey data where information is self reported? I have no problem with _objective_ self-reported data. If you asked somebody how often they eat cheese, I would be happy to accept the results. Whether or not somebody used a gun “in self-defense”, however, is highly subjective. By the same token, I wouldn’t take at face value data showing that 70% of people reported that they were “good people”. ---
**Author:** ![steronz](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/steronz/32/11765_2.png) [@steronz](https://boards.straightdope.com/u/steronz)\ **Post date:** [June 6, 2014, 4:31pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/50 "2014-06-06T16:31:26Z")
> [@Bone](#): > > Is it fair to say that conversely you believe that DGUs should not be treated as zero, and that they do in fact occur? Yeah, sure. > [@Bone](#): > > In your hypothetical, if a mugger would only take a person’s wallet and not otherwise harm them, is it a DGU if a person shoots that mugger and the mugger dies? This is assuming perfect knowledge for the sake of sussing out definitions, in reality you’d never be able to know this. Sure, it’s a DGU. It’s a DGU even if the weapon isn’t fired. But if we _know_ that a life wasn’t saved, merely the contents of a wallet, then what do we do with that? How do we compare that against, say, an accidental gun death? I’m not saying that these sorts of scenarios _aren’t_ DGUs, I’m saying that DGUs don’t really mean anything. ---
**Author:** ![Bone](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bone/32/407_2.png) [@Bone](https://boards.straightdope.com/u/Bone)\ **Post date:** [June 6, 2014, 6:00pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/51 "2014-06-06T18:00:53Z")
> [@Really\_Not\_All\_That\_Bright](#): > > I have no problem with _objective_ self-reported data. If you asked somebody how often they eat cheese, I would be happy to accept the results. Whether or not somebody used a gun “in self-defense”, however, is highly subjective. Isn’t this similar to charges that the CDC wasn’t being objective? I don’t know if there is a way to satisfy this criteria. The NCVS data is available (I haven’t looked at the underlying source data myself). Data from the other studies has been available as well. While DGU is subjecvtive moreso than ‘did you eat cheese?’, the questioning wasn’t as simple as this. For the NCVS data, they only looked at people who were victims of a subset of crimes. After that, they asked if they did anything to try and prevent it. Then they ask what they did, and using a gun is a response option. Again, I’m not a statistician or a criminologist, but this is more rigorous than simply asking if someone used a gun in self defense. The population is restricted to victims of certain crimes, and there are a series of questions that must be addressed to lead down the path before it qualifies as DGU. The NCVS data while not necessarily a floor when estimating DGU, it certainly is expected to return lower estimates than other methods used. In what way would you say that the NCVS data is not objective? How would objective data be identified? ---
**Author:** ![Really\_Not\_All\_That\_Bright](https://avatars.discourse-cdn.com/v4/letter/r/e8c25b/32.png) [@Really\_Not\_All\_That\_Bright](https://boards.straightdope.com/u/Really_Not_All_That_Bright)\ **Post date:** [June 6, 2014, 7:07pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/52 "2014-06-06T19:07:45Z")
Direct observation, I suppose. As I said, I’m not sure there _is_ an accurate way to study this. ---
**Author:** ![Damuri\_Ajashi](https://avatars.discourse-cdn.com/v4/letter/d/898d66/32.png) [@Damuri\_Ajashi](https://boards.straightdope.com/u/Damuri_Ajashi)\ **Post date:** [June 7, 2014, 1:50am UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/53 "2014-06-07T01:50:17Z")
> [@Hentor\_the\_Barbarian](#): > > I can’t help but note this. I have no idea why gun ownership would lead to more non-gun homicide. That makes no sense. Hey, I just bought a gun so now I feel like strangling you? What the hell is the logic model there? > > Hypothesis A: An armed society is a polite society. > > Hypothesis B: An armed society is a society where more people get shot. > > Of course the problem with guns is gun violence. > > 😕 > [@Hentor\_the\_Barbarian](#): > > As for the idea that, if people don’t have guns they’ll just use something else… > > I just looked at Wisqars injury reports for fatal versus non fatal assaults by firearms vs stabbing. 2010 was the most recent data available. > > The ratio of fatal to non fatal firearms assault was 1:5. > The ratio of fatal to non fatal stabbing assault was 1:100. > > Proportionally more people survive stabbing attacks than firearms attacks. > > But that’s just data. Better stick with your gut feeling on this. If you cannot find a relationship between gun ownership and murders generally but you CAN find a relationship between gun ownership and GUN murders, then doesn’t it stand to reason that guns are being used for murders where guns are available and isn’t necessarily a driver of more murders being committed? ---
**Author:** ![Damuri\_Ajashi](https://avatars.discourse-cdn.com/v4/letter/d/898d66/32.png) [@Damuri\_Ajashi](https://boards.straightdope.com/u/Damuri_Ajashi)\ **Post date:** [June 7, 2014, 3:28am UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/54 "2014-06-07T03:28:29Z")
[quote=“steronz, post:40, topic:689994”] Of course. If the homicide, however legally justified, didn’t prevent an injury or death, why should we count it as such? I believe that’s called “making shit up.” How is this contentious?{/quote] OK, I see what you’re saying. I overstated things when i said that eery justifiable homicide is a life saved. Ok I agree with you, I should have said that it is evidence that guns can save lives (even though I can’t prove that any particular one would have actually saved a life). But can we also agree that guns save lives and we can’t just look at gun murders and conclude that legal gun ownership is a net negative. > [@](#): > > All I know is that, right now, studies show that they’re a net safety loss. What studies? And how can you tell without knowing how many lives guns save? > [@](#): > > DGU isn’t even gross income, though, because it doesn’t tell you how many lives were saved, from which you can then deduct lives lost. It tells you how many reported defensive gun uses from which from which you can deduct the ones that didn’t save a life. You don’t deduct lives lost, you compare the lives saved to the lives lost. > [@](#): > > I don’t know how many different ways I can explain this to you. You say that there are maybe 300,000 DGUs every year, and 30,000 gun deaths, so that’s, what, 270,000 lives saved thanks to guns? No thats not what I’m saying at all. I’m saying that there are 30,000 gun deaths of which 18,000 are suicides and don’t count (I once thought they did but upon further investigation it appears that we have absolutely average suicide rates for developed countries despite having a shitload more guns than any other developed country. So I think there may be a very strong replacement effect for suicides if guns aren’t available, unless you can explain to me why suicide rates seem so average desppite a realy really high rate of gun ownership in this country. Of the remaining 12,000, the majority are murders committed by people who aren’t allowed to own guns in the first place so making them illegal isn’t going to have much of an impact on those homicides. Of the ones that remain, we have to determine how many of those deaths would have occurred anyway by other means. THAT is the number that we compare to the lives saved. To determine the number of lives saved it might be useful to see what percentage of muggings result in the death of the victim and apply that rate to dgu to try to get some sort of estimate of how many lives it might save. > [@](#): > > I say that not only is that pure speculation, but it’s such a ridiculously high number that if were true, these studies that we’re talking about would clearly show a massive benefit to owning a gun. The actual number of lives saved by defensive gun use must be much, much lower, on the order of a few thousand. Yeah I agree with that. I don’t know what the number is but I suspect that the number of lives actually saved is much much lower than the number of DGU. > [@Really\_Not\_All\_That\_Bright](#): > > You seem to have misread the post you were responding to: “if you can show me a method other than self-reporting I’m all ears”. I don’t know how else you would do it. > [@](#): > > I am not suggesting anything about the census bureau’s surveying ability. I am pointing out an error in your post. And yes, there is no reason why they would distinguish between legal and illegal firearm use. Does it help to know how the NCVS is conducted? They only ask the victims of crimes. > [@Hentor\_the\_Barbarian](#): > > That helps a little, but that’s not quite the point. The point is more about hearing the clock strike 13 times - which strike was the false one? (To be more clear for you, the validity of each individual one of those incidents are called into question by the overall volume of incidents.) > > Do you really believe that someone had cause to use a firearm in self-defense 50 times in five years? I don’t. So, which of the 50 incidents was actual and which were fantasy? > > Now, across the remaining individual incident reports, are any motivated by a similar process - a fantasy about being threatened or using a gun to ward off harm? Is this what the NCVS is based on? I thought the NCVS was a survey of crime victims. They ask the victims of crimes if what happened and if they indicate that they used a gun to ward off the criminal, they count that as a dgu. The NCVS does not count dgu responses from people who did not report a crime, do they? You seem to be applying a critique of Kleck’s study to the NCVS. A critique that probably doesn’t apply to the NCVS. > [@Bone](#): > > Isn’t this similar to charges that the CDC wasn’t being objective? I don’t know if there is a way to satisfy this criteria. The NCVS data is available (I haven’t looked at the underlying source data myself). Data from the other studies has been available as well. While DGU is subjecvtive moreso than ‘did you eat cheese?’, the questioning wasn’t as simple as this. For the NCVS data, they only looked at people who were victims of a subset of crimes. After that, they asked if they did anything to try and prevent it. Then they ask what they did, and using a gun is a response option. Again, I’m not a statistician or a criminologist, but this is more rigorous than simply asking if someone used a gun in self defense. The population is restricted to victims of certain crimes, and there are a series of questions that must be addressed to lead down the path before it qualifies as DGU. > > The NCVS data while not necessarily a floor when estimating DGU, it certainly is expected to return lower estimates than other methods used. > > In what way would you say that the NCVS data is not objective? How would objective data be identified? To be fair, I believe the NCVS interviews people for several years after they are victimized. So it might not just be dgu that occurred during a reported crime. > [@Really\_Not\_All\_That\_Bright](#): > > Direct observation, I suppose. As I said, I’m not sure there _is_ an accurate way to study this. Can’t we get in the ballpark? ---
**Author:** ![Hentor\_the\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/h/8edcca/32.png) [@Hentor\_the\_Barbarian](https://boards.straightdope.com/u/Hentor_the_Barbarian)\ **Post date:** [June 7, 2014, 3:51am UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/55 "2014-06-07T03:51:13Z")
> [@Damuri\_Ajashi](#): > > Does it help to know how the NCVS is conducted? They only ask the victims of crimes. > > Is this what the NCVS is based on? I thought the NCVS was a survey of crime victims. They ask the victims of crimes if what happened and if they indicate that they used a gun to ward off the criminal, they count that as a dgu. The NCVS does not count dgu responses from people who did not report a crime, do they? > > You seem to be applying a critique of Kleck’s study to the NCVS. A critique that probably doesn’t apply to the NCVS. 😕 I was talking about neither the NCVS or Kleck. I said it was a study by Hemenway. You can’t just make assumptions when you don’t understand something. However, the implications apply to any self-reported defensive gun use. Unless there is some more objective measure, we can’t know what actually happened, and there’s evidence to illustrate that people claim DGU when they were the aggressor, when they erroneously perceived a threat, and when they just made shit up. ---
**Author:** ![PhillyGuy](https://avatars.discourse-cdn.com/v4/letter/p/ed655f/32.png) [@PhillyGuy](https://boards.straightdope.com/u/PhillyGuy)\ **Post date:** [June 7, 2014, 11:37pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/56 "2014-06-07T23:37:47Z")
> [@Bone](#): > > If an unarmed person starts trying to kick in my door, and through the window they see me display a firearm so they run off, no shots fired, no police report - do you consider that a DGU? If yes, how would that be captured by any study you’d accept? If no, why not? Burglaries are far more common than home invasion. Once the criminal noticed you there they would likely flee regardless. Your post illustrates why I don’t put much stock in those surveys of gun-owners where a large proportion claim to have saved themselves through defensive use of their weapon. ---
**Author:** ![PhillyGuy](https://avatars.discourse-cdn.com/v4/letter/p/ed655f/32.png) [@PhillyGuy](https://boards.straightdope.com/u/PhillyGuy)\ **Post date:** [June 8, 2014, 12:09am UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/57 "2014-06-08T00:09:40Z")
> [@Damuri\_Ajashi](#): > > Is it just me or are all three of your links talking about the correlation between guns and gun murder rather than guns and murder generally. The inference we are invited to make when someone makes this statement is that guns leads to more murder but I don’t see any of your links saying that. My earlier link came from picking the newest of its type. Here’s an older one with your stipulation: [http://jonathanstray.com/papers/FirearmAvailabilityVsHomicideRates.pdf](http://jonathanstray.com/papers/FirearmAvailabilityVsHomicideRates.pdf) As I said, the correlation isn’t as high when you pull out homicide alone, or suicide alone, rather than combining them, but it is there. Since we can’t randomly put people in gun/no gun groups, each individual study will always have weaknesses. The same thing was true of the studies that established smoking as being dangerous. While the relationship between smoking and lung cancer may seem to you obvious, there used to be plenty of objections. And of course the relationship between gun ownership and violent death looks just as obvious to me, thinking about the US, Canada, UK and Australia, as the lung cancer relationship seems to you (and me). The starting point in thinking about this is the case control studies. Here is a good one, from what is probably the best medical journal, showing the relationship between violent death in the home (all means, not just guns) and having a gun in the home: [http://www.nejm.org/doi/pdf/10.1056/NEJM199310073291506](http://www.nejm.org/doi/pdf/10.1056/NEJM199310073291506) The weakness of the above kind of study is that maybe gun owners have guns because they have far more real life enemies than non-gun-owners and/or because they always thought they might wanted to commit suicide some day and wanted to have a tool for the purpose readily at hand. I think this is pretty unlikely – most of the gun owners I know live in safe low crime areas, and do not seem to have loads of enemies. But this potential study weakness does mean that such as study, good as it is, can’t prove guns are dangerous all by itself. The weakness of the state and country correlation studies – just taken by themselves – is the ecological fallacy. But when you see both kinds of studies almost always pointing in the same direction – more guns, more death – well, that’s when you should start questioning your POV. ---
**Author:** ![Bone](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bone/32/407_2.png) [@Bone](https://boards.straightdope.com/u/Bone)\ **Post date:** [June 8, 2014, 12:12am UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/58 "2014-06-08T00:12:23Z")
> [@PhillyGuy](#): > > Burglaries are far more common than home invasion. Once the criminal noticed you there they would likely flee regardless. That’s interesting and all, but you do realize I was stating a hypothetical, right? Even still, do you recognize that home invasions do in fact occur? I’m not sure how effective it would be to tell the home invader, _“you know, burglaries are far more common than home invasions, now that you know I’m here aren’t you supposed to run away?”_ I’d rather not base my safety on the goodwill of criminals. > [@](#): > > Your post illustrates why I don’t put much stock in those surveys of gun-owners where a large proportion claim to have saved themselves through defensive use of their weapon. My post of a hypothetical leads you to discount the National Crime Victim Survey taken by the Census and supported by the DOJ? Please, elaborate on thought process you used to come to that conclusion. But first, answer the hypothetical. Do you consider that a DGU? If yes, how would that be captured by any study you’d accept? If no, why not? ---
**Author:** ![Damuri\_Ajashi](https://avatars.discourse-cdn.com/v4/letter/d/898d66/32.png) [@Damuri\_Ajashi](https://boards.straightdope.com/u/Damuri_Ajashi)\ **Post date:** [June 8, 2014, 6:49am UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/59 "2014-06-08T06:49:51Z")
> [@Hentor\_the\_Barbarian](#): > > However, the implications apply to any self-reported defensive gun use. Unless there is some more objective measure, we can’t know what actually happened, and there’s evidence to illustrate that people claim DGU when they were the aggressor, when they erroneously perceived a threat, and when they just made shit up. And i was trying to point out that the NCVS is a survey of crime victims. they ask the crime victims if they defended themselves in any way and if they say they used a gun, then that is noted. > [@PhillyGuy](#): > > Your post illustrates why I don’t put much stock in those surveys of gun-owners where a large proportion claim to have saved themselves through defensive use of their weapon. Wait. What!?!?! What proportion of gun owners do you think are claiming dgu? > [@PhillyGuy](#): > > My earlier link came from picking the newest of its type. Here’s an older one with your stipulation: > > [http://jonathanstray.com/papers/FirearmAvailabilityVsHomicideRates.pdf](http://jonathanstray.com/papers/FirearmAvailabilityVsHomicideRates.pdf) I’m perfectly willing to stipulate that America has a shitload of guns and has a shitload of murders. Most of them are committed by people who are not legally allowed to own guns. I am even willing to stipulate that there are more murders because of legally available guns There is some non-zero number of legal gun owners that commit homicide with a gun that would not have committed homicide if they did not legally own guns. But I am not at all convinced that these additional murders are large enough to erase any benefit that legally available guns guns might provide. We can certainly do more to try to keep guns out of the hands of criminals but I don’t see where legally available guns do more harm than good. > [@](#): > > As I said, the correlation isn’t as high when you pull out homicide alone, or suicide alone, rather than combining them, but it is there. Do you see any correlation if you pull out suicide alone? > [@](#): > > The same thing was true of the studies that established smoking as being dangerous. While the relationship between smoking and lung cancer may seem to you obvious, there used to be plenty of objections. none of those studies about the correlation between smoking and lung cancer was subject to the criticism that cause and effect might somewhat reversed. > [@](#): > > The starting point in thinking about this is the case control studies. Here is a good one, from what is probably the best medical journal, showing the relationship between violent death in the home (all means, not just guns) and having a gun in the home: > > [http://www.nejm.org/doi/pdf/10.1056/NEJM199310073291506](http://www.nejm.org/doi/pdf/10.1056/NEJM199310073291506) > > The weakness of the above kind of study is that maybe gun owners have guns because they have far more real life enemies than non-gun-owners and/or because they always thought they might wanted to commit suicide some day and wanted to have a tool for the purpose readily at hand. I think this is pretty unlikely – most of the gun owners I know live in safe low crime areas, and do not seem to have loads of enemies. I suspect that there are some extra deaths that occur because you have a gun in the home that would not occur if you only had knives and bats (mostly crimes of passion, finding your wife sleeping with your neighbor, etc.) but I don’t think it anywhere close to triples the average gun owner’s risk of dying a violent death. The study controls for neighborhoods doesn’t it? I am looking at page 1087 where they say that guns in the home were most strongly linked with homicide by family members or intimate acquaintances (7.8 odds ratio) and does not seem to increase chances of being killed by acquaintances unidentified intruders or strangers. In the next paragraph it goes on to say that the odds ratio of death was 20.4 if there was a history of domestic violence. Perhaps stronger constraints on gun ownership in households with histories of domestic violence might make sense. I would be curious to see how the study results would be affected if we excised homes with histories of domestic violence. > [@](#): > > But this potential study weakness does mean that such as study, good as it is, can’t prove guns are dangerous all by itself. > > The weakness of the state and country correlation studies – just taken by themselves – is the ecological fallacy. > > But when you see both kinds of studies almost always pointing in the same direction – more guns, more death – well, that’s when you should start questioning your POV. I constantly question it. I as interested in reducing the death count as anyone else. But knowing the benefits that guns can have in our society, I generally require that any proposal presented will actually reduce the death toll and not just the number of legally owned guns. ---
**Author:** ![Bone](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bone/32/407_2.png) [@Bone](https://boards.straightdope.com/u/Bone)\ **Post date:** [June 8, 2014, 3:26pm UTC](https://boards.straightdope.com/t/studying-firearm-related-violence/689994/60 "2014-06-08T15:26:28Z")
> [@Really\_Not\_All\_That\_Bright](#): > > Direct observation, I suppose. As I said, I’m not sure there _is_ an accurate way to study this. Do you really think DGU is unique in such a fashion that no acceptable evidence can be had without direct observation? That seems like an artificially high bar. Surely there would be a way to estimate or extrapolate instances of DGU occurring. We do this for nearly everything - but you think DGU is so unknowable to render it unique in this regard? At least 19 studies have been done to estimate the amount of DGU. All return figures in the multiple hundreds of thousands or more, except one. The NCVS study by Census and DOJ says just over 100K. I use that figure because it’s the lowest, and the methodology the most restrictive. It’s still a huge number. But you don’t accept it because…reasons? > [@Hentor\_the\_Barbarian](#): > > However, the implications apply to any self-reported defensive gun use. Unless there is some more objective measure, we can’t know what actually happened, and there’s evidence to illustrate that people claim DGU when they were the aggressor, when they erroneously perceived a threat, and when they just made shit up. So the data is so unreliable in your opinion, we can’t know what happened, but we do know that based on anonymous people funded by the Joyce Foundation that most instances of reported DGU from one study were probably illegal. Critical examination appears inconsistent here. In what way would you say that the NCVS data is not objective? How would objective data be identified? Consider the hypothetical I posed, _“If an unarmed person starts trying to kick in my door, and through the window they see me display a firearm so they run off, no shots fired, no police report - do you consider that a DGU? If yes, how would that be captured by any study you’d accept? If no, why not?”_ If you could address these items, it would go a long way towards understanding your position. [Previous page](https://boards.straightdope.com/t/studying-firearm-related-violence/689994.md?page=2) [Next page](https://boards.straightdope.com/t/studying-firearm-related-violence/689994.md?page=4)