# So you killed your kid while cleaning your gun. **URL:** \ **Category:** The BBQ Pit\ **Created:** [March 26, 2013, 5:03pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020 "2013-03-26T17:03:44Z")\ **Posts on this page:** 20\ **Page:** 15
**Author:** ![filmore](https://avatars.discourse-cdn.com/v4/letter/f/7993a0/32.png) [@filmore](https://boards.straightdope.com/u/filmore)\ **Post date:** [April 4, 2013, 5:01pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/281 "2013-04-04T17:01:51Z")
Yes, thanks **Hentor** for explaining the stats. Stats are one place where you have to really understand the analysis to really understand what it means. My mom said using common sense to interpret the data is wrong 95% of the time 🙂 ---
**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:** [April 4, 2013, 7:08pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/282 "2013-04-04T19:08:58Z")
> [@Hentor\_the\_Barbarian](#): > > Well, again, the example you’re using is a bit like saying “I’ve done a lot of research, and I can predict with 95% confidence that an average American’s annual salary falls somewhere between $0 and $1 million.” Really? Boy, that’s some valuable research there, pal! If 1 is the zero effect then wouldn’t it be more like saying that I can say with 95% certainty that dropping out of high school will result in an income of between 1% and 101% of the median salary. You seem to be saying that the zero effect is the equivalent of a $1 million salary. > [@](#): > > But that’s just due to your example being so bad. If I understand the point behind it, you are asking something more like “If the 95% confidence interval just barely includes the zero effect at one end, can’t we kind of think that our estimate is probably different from zero.” And in that case, you kind of can, in the same way that if you were using p values, and came up with a p-value of .07, you might be tempted to say “Yeah, I know it doesn’t meet the established criteria, but maybe perhaps it means something.” So you cant really assume anything approaching a normal distribution? You can’t really infer anything from the fact that while the confidence interval includes the zero effect it doesn’t center anywhere near the zero effect? I’m not saying that a .3 to 1.1 means that the confidence interval just barely includes the zero effect but it does seem to be off to the side. > [@](#): > > It’s just a little more difficult to do that with the 95% CI, because it’s easier to eyeball the CI to see that it does or does not include the zero effect than it is to more precisely see how far from .05 the p-value would be. > > It’s also not relevant to the case of the Kellermann paper, since the 95% CI for rifles and shotguns also includes values indicating 10% or 30% _increases_ in the risk of homicide when they are in the home. They’re not just barely non-significant, in other words. > > Bottom line, if the CI does include the zero effect, you cannot say it is significant. > > Furthermore, those effects that are just barely significant at the univariate level are almost always explained away entirely when other significant variables are also included in multivariate testing. If the confidence interval doesn’t correct for things like demographics and locality I can see why the number is basically useless. ---
**Author:** ![Chimera](https://avatars.discourse-cdn.com/v4/letter/c/8e8cbc/32.png) [@Chimera](https://boards.straightdope.com/u/Chimera)\ **Post date:** [April 4, 2013, 7:24pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/283 "2013-04-04T19:24:16Z")
Haven’t read the thread, won’t read the thread. Keep seeing the thread title and thinking; “Didn’t want that kid anyway” ---
**Author:** ![Euphonious\_Polemic](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/euphonious_polemic/32/3002_2.png) [@Euphonious\_Polemic](https://boards.straightdope.com/u/Euphonious_Polemic)\ **Post date:** [April 4, 2013, 7:31pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/284 "2013-04-04T19:31:00Z")
> [@Damuri\_Ajashi](#): > > I’m not saying that a .3 to 1.1 means that the confidence interval just barely includes the zero effect but it does seem to be off to the side. I’m not a statistics whiz, but I can recall that in my stats class, if I gave an answer that was along the lines of “it is just a little off to the side”, I would have failed. That’s not how mathematics or statistics work. ---
**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:** [April 5, 2013, 1:20am UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/285 "2013-04-05T01:20:27Z")
> [@Damuri\_Ajashi](#): > > If 1 is the zero effect then wouldn’t it be more like saying that I can say with 95% certainty that dropping out of high school will result in an income of between 1% and 101% of the median salary. > > You seem to be saying that the zero effect is the equivalent of a $1 million salary. Look, my only point is that the range you picked basically covered most of the possible values, which suggests your prediction isn’t really very useful. > [@](#): > > So you cant really assume anything approaching a normal distribution? Well, the estimate and confidence interval are already based on the normal distribution. > [@](#): > > You can’t really infer anything from the fact that while the confidence interval includes the zero effect it doesn’t center anywhere near the zero effect? You’re trying to parse for the purpose of serving your agenda. The standard is 5%, or p \<.05. You do the test and accept the results. > [@](#): > > I’m not saying that a .3 to 1.1 means that the confidence interval just barely includes the zero effect but it does seem to be off to the side. Again, you don’t do science by squinting to see what you want to see. You work with accepted rules. If you cannot be sure your results are significantly different from zero, that’s what the results are. > [@](#): > > If the confidence interval doesn’t correct for things like demographics and locality I can see why the number is basically useless. I think I’ve explained univariate associations versus multivariate modeling like six or seven times. ---
**Author:** ![tim314](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/tim314/32/3468_2.png) [@tim314](https://boards.straightdope.com/u/tim314)\ **Post date:** [April 5, 2013, 4:50am UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/286 "2013-04-05T04:50:25Z")
> [@Lemur866](#): > > “Deserves got nuthin to do with it.” > > If you point a loaded shotgun at a kid’s head and pull the trigger, the kid’s death is a foreseeable circumstance. Well, yeah. I think I acknowledged the dad’s an idiot. But I’m just saying I feel bad for the guy. The OP said “I hope you suffer a lifetime of ostracism and horrible guilty feelings as a result of your felonious stupidity”. I guess I’m saying, even though he brought it on himself with his own stupid actions, I wouldn’t wish that kind of pain on anyone. ---
**Author:** ![Cartooniverse](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/cartooniverse/32/3084_2.png) [@Cartooniverse](https://boards.straightdope.com/u/Cartooniverse)\ **Post date:** [April 5, 2013, 4:28pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/287 "2013-04-05T16:28:31Z")
> [@tim314](#): > > Well, yeah. I think I acknowledged the dad’s an idiot. But I’m just saying I feel bad for the guy. > > The OP said “I hope you suffer a lifetime of ostracism and horrible guilty feelings as a result of your felonious stupidity”. I guess I’m saying, even though he brought it on himself with his own stupid actions, I wouldn’t wish that kind of pain on anyone. I have what some here call insanely strong reactions to gun violence threads. Have to say, while I agree with the OP’s statement, I am aware that time dulls the hone on that blade. Guilt diminishes and what is left is a painful stone in the heart. A corner of what otherwise might have been a bright happy life is dark. Unless the shooter is a mindless sociopath, that dark heavy spot should exist until they die. Is life of no value that such a burden should not exist? I think not. The shooter won’t be ostracized. He doesn’t have to tell anyone in his life that doesn’t already know. It shouldn’t be the front and center element of his every waking day. Instead, the knowledge that he loved his fucking gun more than his own child should be a knowledge that seeps in during otherwise happy times. Another child is wed, and the child he killed is not in the room. Joyous family vacations are sullied inside his heart because his dead son is not there. That is the caul that is pulled over the heart of any parent who kills their child. ---
**Author:** ![Trinopus](https://avatars.discourse-cdn.com/v4/letter/t/2bfe46/32.png) [@Trinopus](https://boards.straightdope.com/u/Trinopus)\ **Post date:** [April 5, 2013, 11:45pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/288 "2013-04-05T23:45:29Z")
> [@Cartooniverse](#): > > . . . Joyous family vacations are sullied inside his heart because his dead son is not there. That is the caul that is pulled over the heart of any parent who kills their child. There seems to be a pattern in the annual tragedy of people who leave children in cars in parking lots. The first case of the summer usually is not punished too very harshly; judges tend to say, rightly, that the parent is punished enough by the sorrow that you describe. Then the next dumb shit leaves his kid in a car…and the judges come down hard, with severe penalties. Ideally, society learns from tragedies. In practice…not so much. ---
**Author:** ![Cartooniverse](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/cartooniverse/32/3084_2.png) [@Cartooniverse](https://boards.straightdope.com/u/Cartooniverse)\ **Post date:** [April 5, 2013, 11:49pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/289 "2013-04-05T23:49:43Z")
Agreed. Sad thing is, summertime heat comes…every summertime. When people clean guns…they know they can go off if loaded. Sad. ---
**Author:** ![bobot](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bobot/32/21622_2.png) [@bobot](https://boards.straightdope.com/u/bobot)\ **Post date:** [April 5, 2013, 11:50pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/290 "2013-04-05T23:50:14Z")
> [@Chimera](#): > > Haven’t read the thread, won’t read the thread. > > Keep seeing the thread title and thinking; > > “Didn’t want that kid anyway” You should read it head to toe. There’s all kinds of enlightning bickering over statistical bullshit (redundant, I know…) going on here. It’s a real page turner! ---
**Author:** ![elucidator](https://avatars.discourse-cdn.com/v4/letter/e/8edcca/32.png) [@elucidator](https://boards.straightdope.com/u/elucidator)\ **Post date:** [April 6, 2013, 2:05am UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/291 "2013-04-06T02:05:34Z")
Well, this is very educational! Before this, I had no idea about univariate associations versus multivariate modeling. Now, I know that they exist. ---
**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:** [April 7, 2013, 2:06pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/292 "2013-04-07T14:06:37Z")
> [@Hentor\_the\_Barbarian](#): > > Look, my only point is that the range you picked basically covered most of the possible values, which suggests your prediction isn’t really very useful.Well, the estimate and confidence interval are already based on the normal distribution.You’re trying to parse for the purpose of serving your agenda. The standard is 5%, or p \<.05. You do the test and accept the results.Again, you don’t do science by squinting to see what you want to see. You work with accepted rules. If you cannot be sure your results are significantly different from zero, that’s what the results are.I think I’ve explained univariate associations versus multivariate modeling like six or seven times. Putting the multivariate stuff aside, I’m not trying to squint but you seem to be trying to say that an interval of .3 to 1.1 doesn’t tell you anything more than an interval of .6 to 1.4 or .201 to 1.001. I’m not parsing, I’m asking what this tells us other than the fact that it is possible that there is no association and you answer seems to be “absolutely nothing” > [@Cartooniverse](#): > > Instead, the knowledge that he loved his fucking gun more than his own child The guy was an idiot but I see no evidence of this. Now why would you infer this sort of thing? Would you infer this sort of thing about anyone that owns a gun and cleans it while their child is in the room? > [@elucidator](#): > > Well, this is very educational! Before this, I had no idea about univariate associations versus multivariate modeling. Now, I know that they exist. You and me both. ---
**Author:** ![Kable](https://avatars.discourse-cdn.com/v4/letter/k/ed655f/32.png) [@Kable](https://boards.straightdope.com/u/Kable)\ **Post date:** [April 7, 2013, 2:34pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/293 "2013-04-07T14:34:33Z")
> [@Damuri\_Ajashi](#): > > Putting the multivariate stuff aside, I’m not trying to squint but you seem to be trying to say that an interval of .3 to 1.1 doesn’t tell you anything more than an interval of .6 to 1.4 or .201 to 1.001. I’m not parsing, I’m asking what this tells us other than the fact that it is possible that there is no association and you answer seems to be “absolutely nothing” There are lies, damn lies, and Hentors interpretation of statistics. ---
**Author:** ![ladyfoxfyre](https://avatars.discourse-cdn.com/v4/letter/l/ad7895/32.png) [@ladyfoxfyre](https://boards.straightdope.com/u/ladyfoxfyre)\ **Post date:** [April 7, 2013, 5:36pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/294 "2013-04-07T17:36:03Z")
That’s pretty ironic coming from your disingenuous ass. ---
**Author:** ![Kable](https://avatars.discourse-cdn.com/v4/letter/k/ed655f/32.png) [@Kable](https://boards.straightdope.com/u/Kable)\ **Post date:** [April 7, 2013, 5:51pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/295 "2013-04-07T17:51:28Z")
> [@ladyfoxfyre](#): > > That’s pretty ironic coming from your disingenuous ass. How am I disingenuous? ---
**Author:** ![ladyfoxfyre](https://avatars.discourse-cdn.com/v4/letter/l/ad7895/32.png) [@ladyfoxfyre](https://boards.straightdope.com/u/ladyfoxfyre)\ **Post date:** [April 7, 2013, 5:55pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/296 "2013-04-07T17:55:57Z")
I think **Hentor** has done a fine job explaining how your conclusions were based on a complete lack of understanding of statistical analysis. You’ve basically fallen back on “statistics are always manipulated! But only when the other guy is doing it!” ---
**Author:** ![Kable](https://avatars.discourse-cdn.com/v4/letter/k/ed655f/32.png) [@Kable](https://boards.straightdope.com/u/Kable)\ **Post date:** [April 7, 2013, 6:01pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/297 "2013-04-07T18:01:27Z")
> [@ladyfoxfyre](#): > > I think **Hentor** has done a fine job explaining how your conclusions were based on a complete lack of understanding of statistical analysis. You’ve basically fallen back on “statistics are always manipulated! But only when the other guy is doing it!” Actually I thought I was pretty clear with what I claimed and that Kellerman only found an association of lesser homicides (rather than statistical significance), and I then called Kellerman dishonest when he claimed that guns in general (rather than just handguns) were associated with increased homicides. I don’t see anything at all that’s disingenuous about that, but I see Hentor was able to baffle you with big words and bullshit pretty well. You should learn to think more critically. ---
**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:** [April 7, 2013, 8:09pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/298 "2013-04-07T20:09:23Z")
> [@Damuri\_Ajashi](#): > > Putting the multivariate stuff aside, I’m not trying to squint but you seem to be trying to say that an interval of .3 to 1.1 doesn’t tell you anything more than an interval of .6 to 1.4 or .201 to 1.001. I’m not parsing, I’m asking what this tells us other than the fact that it is possible that there is no association and you answer seems to be “absolutely nothing” No, not absolutely nothing. But it does tell you that you cannot acceptably rule out the zero effect, and thus cannot conclude that there is any meaningful association between the variables. What do you want it to tell you? ---
**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:** [April 7, 2013, 8:11pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/299 "2013-04-07T20:11:35Z")
> [@Kable](#): > > Actually I thought I was pretty clear with what I claimed and that Kellerman only found an association of lesser homicides (rather than statistical significance), and I then called Kellerman dishonest when he claimed that guns in general (rather than just handguns) were associated with increased homicides. He didn’t find an association with lesser homicides. You cannot stop yourself from trying to interpret a non-significant association. > [@](#): > > I don’t see anything at all that’s disingenuous about that, but I see Hentor was able to baffle you with big words and bullshit pretty well. You should learn to think more critically. I think at this point it is only **Damuri Ajashi** who does not think you are one dumb, deceitful fuck. Congratulations! That’s an impressive accomplishment. ---
**Author:** ![elucidator](https://avatars.discourse-cdn.com/v4/letter/e/8edcca/32.png) [@elucidator](https://boards.straightdope.com/u/elucidator)\ **Post date:** [April 7, 2013, 10:36pm UTC](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020/300 "2013-04-07T22:36:53Z")
> [@Kable](#): > > …but I see Hentor was able to baffle you with big words and bullshit pretty well. You should learn to think more critically. Its not that. Smart as I am, and that is damned smart, there is a smattering of human knowledge outside my range. On the rare occassion that I run into those limitations, like any math more complex than long division, I rely on others. Because I can, because this is Da Boards, chock a block with elitists, specialists, and snotty pedants. Hallelujah! If **Hentor** was blowing smoke, he’d have been busted by now. \* Really\* busted. Not the crippled effort you keep making. But… [QUOTE=Sneering Math Geek] … **Hentor** , your analysis fails due to the bichromatic declension of the univariate perversion, which you have not factored into the boogaloo… [/QUOTE] I don’t trust a lot, but here I don’t have to trust, I can depend that there are enough experts on any given topic that any scrabbling rodent of ignorance will be stomped out toot damn sweet! [Previous page](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020.md?page=14) [Next page](https://boards.straightdope.com/t/so-you-killed-your-kid-while-cleaning-your-gun/654020.md?page=16)