Hacker News Reader

Comment by opportune | Hacker News Reader

I would like to know the contexts behind these defensive gun usages before I come to a conclusion. It could be that a significant portion are "empty" uses (e.g. brandishing a gun when scared by a weird noise at night when nobody was there), not actually defensive / used in a situation where the gun greatly escalated the situation (e.g. someone cuts you off in traffic and you "defend" your space), etc. - even though these wouldn't technically count according to the wording of the survey, we are still relying on respondents' interpretations of the questions. The author mentions that if anything the rate was underreported, but knowing gun users personally, I find it just as likely that they were willing to overstate their gun use or the utility of their gun as a type of bragadocious machoism.

It's true the CDC was obviously quite biased, but the study could have still been flawed

View on HN

The CDC study being discussed in this paper didn't followup but according to OP Kleck's other studies did - in Kleck's studies when people said they had a DGU they were asked to elaborate on the circumstances so this could be evaluated. By his analysis about 9% of reported DGUs didn't qualify as defensive, so he applied that as a modifier when comparing the new results to his.

On the flip side there also exist reasons to think the CDC undercounts DGUs, most notably that they only asked people who had a gun in the home if they had used a gun in self-defense. If respondents only had a gun at work or used someone else's gun or used to have a gun at home but no longer did, they weren't asked the DGU question. Also, most DGUs are arguably illegal so reporting one is reporting having engaged in illegal activity.

If you want to make the data more accurate you don't get to just apply reasons to think it's an overcount, you also need to apply reasons to think it's an undercount. (On the whole, Kleck's estimates seem pretty conservative.)

> reasons to think the CDC undercounts DGUs, most notably that they only asked people who had a gun in the home if they had used a gun in self-defense.

Can you point me to something to back-up this claim? My gun-enthusiast friends often accuse the NCVS of under-counting, usually having to do with survey methodology flaws. But I've looked at the survey[1], and I can't find anything so substantiate their (or your) claim.

But, even still, the CDC numbers are _orders of magnitude_ lower than Kleck et al.--like 10x lower (afaik-not an expert here). So it's a tough pill to swallow that minor (or even substantial) survey errors would lead to that sort of discrepency.

https://www.bjs.gov/content/pub/pdf/ncvs104.pdf

Here's the exact wording of the question:

> “During the last 12 months, have you confronted another person with a firearm, even if you did not fire it, to protect yourself, your property, or someone else?”

I don't think we should seriously modify those answers with anecdotes of "bragadocious machoism". The article says, "in sum, the surveys used an excellent, carefully worded DGU question, in contrast to the wordings used in so many other surveys".

I would like to see a more nuanced analysis of this. How many cases were there where the gun made a positive difference? The answers to the asked question don't really tell us much.

I have had once a guy draw a gun on me while I was walking my dog and yelled at him because he almost ran me over. He would probably answer this question with "Yes" but this was certainly not defensive use and the gun made a normal dispute into a potentially life or death situation.

I know, I read the article. I'm saying that we're relying on respondents' interpretations of "confront another person" and "defend." Given that a small absolute percentage of people replied in the affirmative, I think it's reasonable to assume that some portion of those may have been confused by the question or over interpreted it.

I'm not disagreeing with the author, I just would want to see more data. The reason this doesn't 100% pass my sanity check is that the study's figures suggest that ~2m people use gun's defensively every year, which is a very high figure given that there are ~680k violent crimes every year per the study. So I would want to know the distribution of the cases when guns are used defensively, to understand this discrepancy

>I'm saying that we're relying on respondents' interpretations of "confront another person" and "defend."

If confronting someone with lethal force prevented an attack from occurring, do you count that as defense? It sounds like you want some kind of guarantee that the attack was going to happen if the person was not confronted, which is impossible for you to know for sure.

Personally, I like erring on the side of preventing an attack on myself/loved ones then the side of "let's make sure they're actually going to hurt me/my family before I do anything."

I get what you're saying, and that it's hard to come up with simple, straightforward questions that address all the likely misunderstandings. But, in this case, I was observing that in response to your comment,

> even though these wouldn't technically count according to the wording of the survey, we are still relying on respondents' interpretations of the questions

it seemed that in this case, the question was worded rather well. Particularly since your response was based in personal anecdote.

Perhaps I'm not communicating myself very well. To be blunt, the question could be worded as well as possible, that doesn't preclude the subject material from introducing a bias anyway. And given that the subject is relatively controversial and not 100% well-defined (we have no way of determining whether the use was "justified" or even actually defensive), that could indeed introduce a significant bias.

Well, I see what you mean, but it sounds like _no_ servey could be acceptable by that criteria.

Keep in mind, brandishing your weapon is illegal in most of the states (in fact I am unaware of any state where it's legal to brandish your weapon).

Regarding cases of road rage, there are many incidences where one side was clearly aggravating the situation (quite possibly by getting out of the car or pulling out a baseball bat) when the situation was deescalated because of a gun involved or many times gun was the only way out [1].

1. http://www.skyvalleychronicle.com/FEATURE-NEWS/Road-Rage-A-G...

The article mentions this as a potential bias

Reply on news.ycombinator.com