Summations Only

But I didn't commit the ecological fallacy by making inferences about individuals. The argument I made, and that you quoted, is that personal experience and feelings are a poor way to understand the larger world outside our personal lives. Statistics are much better for that purpose. You feel safer with a gun, then you extrapolate that to the population at large and conclude that the more people with guns the safer everyone will be, but when we measure this statistically we find this conclusion is wrong. Statistics tell us that most people who believe themselves safer after purchasing a gun are wrong. I am not committing the ecological fallacy by claiming I know which individuals that would be.And you don't know, either. Unless you have some special circumstances, you have no idea whether you're safer or less safe with a gun in the house. But if you believe you are safer, then statistics say you're more likely wrong than right. Poor impulse control? Well of course it doesn't make everyone more safe. We're talking statistically (or at least I am), and statistically this is so obvious as to go without saying. Would I be correct in guessing that the statistical significance of the dangers of not owning a gun is something you *feel* is true, rather than something you have evidence for?From your next Message 968: One could spend all one's time defending oneself against the stuff you make up, but why bother.--Percy

Here's a chart where I eyeballed the points from Dr A's chart to a website that does best fit of scatter plots (Scatterplot - NLVM): The line is very similar to Dr A's, r=0.724 shows a fairly strong linear relationship.The problem with Coyote's chart was that it included dramatically different countries possessing poor baselines for comparison, and it jammed all the similar countries up against one axis. Dr A's chart attempts to compare countries that are roughly similar, namely close approximations to western style societies.Since zero guns must by mathematical necessity correspond to zero gun deaths, and since the line can only rise from the 0/0 origin, and since the data supports a linear relationship, you're missing both data and a mechanism behind your position that increasing the number of guns would cause gun deaths to decrease.--Percy

Well, Crash, I can see you'll just go on and on about ecological fallacies and the greater value of individual judgment over statistical analysis, so I'll just leave you to it. The fact remains that the best way to reduce gun deaths would be to reduce gun ownership.--Percy

Dr Adequate already addressed the rest of your post, but about this: I'm again sensing that there's something you're simply missing about statistics. Can you describe why you expect a best fit line on a scatter plot of data from 37 different countries to go through 0/0?The point I actually made was that we know mathematically that 0 guns must correspond to 0 gun deaths and that the line must rise from there, and the data we have is consistent with this. You have no data and no mechanism for the line going in any other direction.--Percy

You're talking about the Monster Hunter Nation website, right? There are plenty of articulate pro-gun advocates out there you can quote, but I'd prefer that people condense the information they gather into their own words, because otherwise I'm responding to something someone else said who's never going to answer, and I can't even be sure that my interpretation of someone else's words is the same as the person quoting them.But anyway, he mentions the Brady Center, so here's a link to a relevant fact sheet from the Brady Center: Comparing the Incidence of Self-Defense Gun Use and Criminal Gun Use. It includes this table and explains that the error people like Mr. Correia are making is comparing results from two different approaches to studying the problem: But no one is arguing that guns are never successfully used for self-defense. The point is that the number of gun deaths is proportional to gun prevalence. What you're offering as a counterargument isn't a counterargument at all, it's just additional relevant information. Even if we assume your counterargument that taking away our guns will leave us helpless before criminals is totally true, it doesn't change the fact that gun deaths are proportional to gun prevalence. Your solution to the crime problem, increasing gun prevalence, will only increase gun deaths. For the sake of making a point let me grant Crash's position that fewer guns means higher crime rates. To most of the rest of the civilized world where life is more precious than money trading higher crime rates for lower death rates would be an excellent tradeoff.But the real reason arming the citizenry isn't an effective crime deterrent is simply because of the nature of crime. The criminal is ready, you're not (the impersonal you). In home defense the criminal has his gun in his pocket or even in his hand, and you have yours locked in a gun cabinet (the safest way to go but totally useless for home defense) or is in a drawer somewhere (potentially much more useful, but if it's loaded then, well, you've got a loaded gun in the house that anyone can find and use).For personal defense when you're out and about you'll be carrying the gun on your person in a holster somewhere, so you now need to get it out without the criminal noticing. And what are the odds that you're actually carrying the gun? After a few years of carrying the gun around with nothing ever happening, how many people actually keep carrying it? Do you take the gun on vacation? Not if you travel by air you don't. Do you take your gun to the beach? Not likely. Now you've got a gun sitting forgotten at home.It is also important to note that guns used to thwart crimes are not often used in the defense of one's life. It is rare when the criminal's intent is murder. The goal is usually robbery, rape, etc. This is why I keep referring to the irony of the gun-lobby argument. They want to arm the citizenry and put everyone's life at greater risk in order to thwart crime whose predominate goal is coercion and theft, not murder. Anyone who believes life is most precious above all else but supports the gun lobby has some serious inner contradictions to work out.--PercyEdited by Percy, : Wordsmithing.Edited by Percy, : Tiny bit more wordsmithing.

Hi Crash,I appreciate that you have a minor in mathematics, but there does seem to be something you're missing about statistics. You also don't seem to grasp the difference between real world data and ideal mathematical relationships. Instead of actually discussing anything you're maintaining a fusillade of false accusations about ecological fallacies and failure to understand derivatives and so on (and by the way I didn't define the origin as -1,-1, the software did that, I assume it rounded to the nearest integer, and if you really knew anything about statistics you wouldn't expect real world data to go through 0,0). As I said earlier, if that's to be your approach then I'll just leave you to it.To state the obvious even more clearly for your sake, mathematically in an ideal isolated region far from any crazed Canadians shooting Americans from across the border, 0 guns must correspond to 0 gun deaths. The line can only rise from there. You have no mechanism or data for the line reversing direction.--Percy

Hi Crash,I'm not trying to create "teachable moments" for you, I'm just here trying to discuss gun control. I haven't found much you've said in your messages that is correct, whether it be about graph origins or first derivatives or ecological fallacies or even lightning strikes, but you seem determined to carry on in the typical way that you do when you're on the wrong side of an argument, so I'll just leave you to it. One could waste a lot of time responding to the stuff you make up, I don't see any point to it.The data we have indicates that gun death rates are proportional to gun ownership rates, and this is precisely what one would expect. More cars means more vehicle accidents, more food consumption means more fat, more guns means more gun deaths. There's no data or mechanism indicating any other relationship. To reduce gun deaths we must reduce gun ownership.--Percy

Really, yet another person is now willfully misrepresenting you? Wow! About the rest, I don't know how you came up with the misimpressions you did. I've seen the posts from others trying to correct you about a couple of them, but it didn't help and I don't think I could do any better. Sometimes it seems like you just get yourself into a state and, throwing logic and rationality to the wind, issue barrages of senseless accusations in the hope that something sticks while in the meantime distracting all attention from the topic. They're all just things you've made up (I'm sure they appear very real to you), and it doesn't seem like a worthwhile investment of time trying to engage with you about them.Meanwhile, back at the topic, if there's a mechanism you've mentioned whereby the positively sloped line could reverse direction then it's really only necessary to describe it one more time.--Percy

You're in essence in a room full of people looking at you funny while you're screaming at them that they're all crazy. The trendline doesn't begin at -1,-1. This has been explained several times already, but it's hopeless explaining things to you once you're all in a dither, I don't see the point in trying one more time. Look, if I'm as obviously wrong as you seem to think then everyone can see it and there's no sense in going on and on about it. As I persist in my senseless errors more and more people will start chiming in about how wrong I am. If you're right you won't be a lone ranger. This has been responded to multiple times. Don't you need to follow that with, "Now I know that people have offered several counterarguments, such as , but ."

​

--Percy

​

Edited by Percy, : Forgot close quote.

This has been explained before, and it's blindingly obvious. Here's the image again: The x/y minimums and maximums are labeled "Scale" and are for the graph, not the line. I can see where it might be a bit confusing for you because the lower left of the line almost hits -1,-1, but if you look at the upper right of the line you'll see that it's nowhere near the x/y maximum for the graph of 10,89. That's because 10,89 corresponds to the top right of the graph, not the line.The software chose the values for the x/y minimums and maximums that you see, and while the user can change them if he wishes, I didn't do that. And no matter what values you choose, when you click on Apply it only changes how the graph is displayed. It doesn't affect the line one bit. The line can only be influenced by the x,y values in the table.Here's the link to that website: Play with it a bit. You'll see I'm not wrong, I'm not making it up, and I'm not misrepresenting you.--Percy

It does go very close to -1,-1. The actual point is -1,-1.2. If you click on Ctrl-+ a few times you'll see that the line does actually pass just a bit below -1 of the y-axis.But that's a meaningless extrapolation of the line - there's no such thing as negative guns or gun deaths. Crash somehow concluded from it that I was manipulating the drawing of the line on the graph. But I provided the link to the website, he has Dr Adequate's graph just like I did, he can eyeball values onto that website just like I did, and hopefully he'll do a better job than I did, Xongsmith says I left out some points.--PercyEdited by Percy, : Wordsmithing.

Hi Crash,I never described the best-fit line as rising from 0,0. The y-intercept is 5.6, it couldn't possibly rise from 0,0.What I did say is that mathematically we know that 0 guns must correspond to 0 gun deaths, and that the line can only rise from there.--PercyEdited by Percy, : Correct the y-intercept value.

You're making the very good point that correlation doesn't imply causation, but in this case we already know several things. We know that guns cause gun deaths. We also know mathematically that 0 guns must correspond to 0 gun deaths and that you cannot have any gun deaths until the number of guns rises above 0. Guns are a prerequisite for gun deaths, and therefore we know that increasing the number of guns increases the number of gun deaths.I think that at heart your argument is actually the same as Crash's, namely that with increasing gun ownership at some point gun deaths begin decreasing because of the deterrent effect. The available data we have doesn't support this, the anecdotal stories aren't worth much, and just a few moments thought about the likelihood of your gun being available and loaded at the moment you need it illustrates the inherent weakness of the deterrence argument.Eternal vigilance is required for a gun to be an effective deterrent, but a gun once owned is a threat that never ends. Normal everyday human qualities like absentmindedness and lack of interest and fatigue and so on just compound the problem. The topic of many casual conversations is someone doing something stupid. We all do stupid things. Even the smartest of us does stupid things, even vice presidents. The gun lobby advocates arming even more people, and that's just a recipe for more gun deaths.--Percy

Whoops, I was playing with the best-fit website in another window and glanced at the wrong image. The correct figure for the y-intercept is obviously 5.6. The point is that obviously the line couldn't go through 0,0 with a non-zero y-intercept. I was explaining why the best-fit line makes sense by describing what we know must be mathematically true, that 0 guns must correspond to 0 gun deaths, and that the line can only rise from there. That's precisely what the best-fit line of real-world data shows, though of course it can't be expected to go through 0,0 - it would be amazing if it did.--PercyEdited by Percy, : Typo.

Sorry, I was playing with the best-fit software in another window to verify my explanation to Crash of how it worked, and I glanced at the wrong image when I needed the y-intercept value. Obviously the y-intercept is 5.6.--Percy