|
Register | Sign In |
|
QuickSearch
EvC Forum active members: 64 (9163 total) |
| |
ChatGPT | |
Total: 916,417 Year: 3,674/9,624 Month: 545/974 Week: 158/276 Day: 32/23 Hour: 2/3 |
Thread ▼ Details |
|
|
Author | Topic: Gun Control Again | |||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
Percy posts:
The line is very similar to Dr A's, r=0.724 shows a fairly strong linear relationship. First off, any statistician worth his or her salt would notice that there is a huge OUTLIER in the data set. This OUTLIER at the top right should be removed from the correlation math. Secondly, the r-squared value is .524, including the hugely effective outlier, or - as they say in stat parlance - this only seems to explain 52.4% of what's going on. Edited by xongsmith, : on Dasher, on Vixen- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
Tangle speculates humorously:
Now wouldn't it be funny if that outlier had to be on the graph because it was actually the USA? Oops. *wink*- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
Well, another factor here is that the numbers come from many different years. 1994 to 2009 or so. So basically, we need more data from a consistent sample.
See List of countries by firearm-related death rate - Wikipedia and Estimated number of civilian guns per capita by country - Wikipedia for the originals. But back to my point of removing obvious outliers - if statistical practice is to be followed, then the USA point must be thrown out of the analysis. But if the USA point is thrown out, then how can the result be used to analyze the USA situation?- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
Percy writes:
by the way I didn't define the origin as -1,-1, the software did that, I assume it rounded to the nearest integer The software did that to make the data easy to see. Our dear esteemed crashfrog will certainly notice that there zero points to left or below the 0,0 origin. It's not just the nearest integer, it's nearest integer that covers the data span in a manner that makes it easy to see. Perhaps if we throw out the bad data point, the software would rescale everything so that the points close to 0,0 would be far away enough from 0,0 for the software to choose the 0 axis in each direction. Also, this is a 1st-order analysis getting a straight line through the data. As RAZD has pointed out in Message 1023, assuming a linear relationship is going to be conceding to an approximation of fit. For example, one might easily imagine that the 1st gun in a country could kill more than 1 person, but that the 1st gun death in a country would be unlikely to have more than 1 gun responsible (don't most firing squads have only 1 gun loaded?).- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
crashfrog replies to Percy:
by the way I didn't define the origin as -1,-1, the software did that No, you did it, perhaps by accident, and I know that because you posted a screencap of the software you used. Xmin and Ymin are user-editable fields in your screencap, and they're both set to -1 - thus, defining the origin at -1,-1. Whether or not you did that intentionally or unintentionally, you didn't fix it and therefore it's your mistake, and a discrepancy between your model and your description of your model. You either didn't notice, or assumed we wouldn't. But the thing is - I know more about statistics and model-fitting than you do. So I did notice because I went looking for the kind of common errors people make in stats work when they don't know what they're doing. Don't be stupid. See Message 1038. There is nothing wrong here. What is wrong is your notion of what Xmin and Ymin must be. The Singapore and Japan points are easier to see this way.- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
Panda writes:
Why are you wanting to remove the USA outlier? I didn't think outliers had to be removed - I thought they only needed to be checked that they were accurate and legitimate. Ok. But notice if you remove the USA point that what you have at the lower left is a cloud of data with a bad r-squared. How can that cloud have much of any significance to say about the USA point?
If my statement above is correct, then the answer is: don't throw out the USA point. You can do a point-by-point analysis to see how much each point contributes to the r-squared value. Maybe Argentina should also be cast out, given that they are currently going through a violent phase compared to their normal numbers?- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5
|
crashfrog writes:
Obviously not, but the trend line rises from -1, -1 - not from 0,0 as Percy has repeatedly stated. It doesn't even pass through 0,0. And again - not that I expect it to, I expect Percy'd description of what he's showing to be accurate. I don't see how this is difficult to grasp. If you tell me you're showing me a trend line that begins at 0,0 and rises from there, then you should not show me a line that begins at -1, -1 and doesn't even pass through 0,0. Saying one thing and showing another is a misrepresentation. It's the definition of misrepresentation. The equation was y = 6.8x + 5.6, so the line rises from the point {0.0,5.6}, not your {-1, -1}. If we plug in your x=-1, then the y value is -1.2 not -1. See how easy it is to make noise over nothing? It is a linear approximation of what might certainly be a very non-linear relationship. As such, the straight line derived from the data is not required to pass through the entire real world truth of {0.0,0.0}. Canadians are in the real world. If there were no guns in the entire world, then there would be, by definition, no gun deaths in the entire world. This is Percy's point. I got it. Most of us here got it. Why are you making such a big deal about this? I think you should be making more noise about the lower left cloud having any statistical relevancy to the USA point.- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
Dr. A posted these way back in the beginning:
from Message 663 All the countries with a Human Development Index over 73% ("very high" according to UNDP) are represented. Figures are taken from the WP articles: * List of countries by Human Development Index - Wikipedia* List of countries by firearm-related death rate - Wikipedia * Estimated number of civilian guns per capita by country - Wikipedia Following the links will show where they got their figures from. and from Message 698:
First thoughts: Looks like someone hit it with a shotgun. I think the trend is fairly clear.
I see there are 47 countries in the Very High category. That's how many points the guy said were on the graph. I'll see if I can find out what happened to the rest. ETA: It must be because WP doesn't provide figures for some countries, e.g. Lichtenstein is in the Very High category of the HDI, but WP has no figures for guns per capita.
And then if I look down to where it goes below 73%, i find that to be country number 77. What's up with that? It must be that .73 was a typo for .793.
Who are "they"? Wikipedia.
Where'd you get the image from? Some guy. A biologist, I think. It hardly matters if he's using figures from Wikipedia. Some guy....... ok, but what does the red line mean? is it a regression line? Do you know? Or does this "some guy" know? BTW, crashfrog, the red line here is most definitely NOT Y=X. And it most definitely does not begin at {-1, -1}. Here is Percy's plot (eyeballed from the above plot) again, from Message 986:
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. My first question is: What happened to Argentina? And 9 other data points? My second is more of a courtesy: Percy, the image cuts off the first data pairs due to the scroll bar. Can you just post the plain data pairs? Or maybe Dr. A should post his 47 pairs.- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
crashfrog writes:
When you show me a trendline that begins at -1, -1 and call it a trendline that begins at 0,0, there's absolutely no reason I should believe you over my eyes. Why on Earth would I do that? Firstly, the only one in this discussion that LIES that the trendline begins at {-1, -1} is YOU. Stop that shit. Secondly, Percy did not use the term "trendline", he used the term "line". The line could be curved, curved sharply downward into {0,0}.- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
crashfrog writes:
I think you've misunderstood what line I've been talking about. I've been talking about the trendline on Percy's plot, not the one on Dr. A's. Percy's line is BLUE not RED. Perhaps you failed to notice that. And Percy threw out 10 data pairs for some unknown reason, including Argentina, which really hurts the correlation coefficient. And the blue line also does NOT go through the point {-1, -1}. You be busted now.- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
crsashfrog continues to be stupid:
But the equation Y = 6.8X + 5.6 doesn't intersect -1, -1 (6.8(-1) + 5.6 doesn't equal -1) so the line on the graph can't be the line described by that equation. Get a magnifying glass. The trendline crosses the x axis at x=-.8235 and it crosses the y axis at -1.2.- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
Percy writes:
The y-intercept is 2.47 Huh???
y = 6.8x + 5.6 <==== looks like 5.6 to me. What did I miss?- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
You... do understand that using a magnifying glass on my screen is just going to show me bigger pixels, right? Did you try Ctrl-Shft-+ a few times? But here is your magnifying glass:
y = 6.8x + 5.6 solve for y when x=0 and for x when y=0. I admit that the line looks black now, but my lcd screen tilted a bit shows a dark blue tint.- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
crashfrog says:
What did I miss? That you've been defending someone who doesn't know anything about statistics. You have it wrong here - what I wanted you to do was to get off the -1,-1 versus 0,0 shit and look at the non-USA data. CS said it looked like buckshot. One of the purposes of finding a trendline is to project it out. So find the non-USA trendline (including Argentina and the other 9 countries Dr.A had!) and project it out to the USA gun ownership value for X and see what kind of Y it comes up with. What kind of confidence do we have in the non-USA line? What if it matched anyway - would we still be right to make the claim that they are proportional? Argue from there. HTH- xongsmith, 5.7d
|
|||||||||||||||||||||||||||||||||||||||
xongsmith Member Posts: 2587 From: massachusetts US Joined: Member Rating: 6.5 |
crashfrog writes:
You can't use a taser or pepper spray if you don't have a taser or pepper spray. You can't use your fists if they're not strong enough to stop the other guy, or you don't know how to throw an effective punch. It's a firearm or nothing, sometimes, and in that kind of situation people have a moral right to use the lethal force required to defend themselves. Geez. The simple answer here is GET A TASER & PEPPER SPRAY. Forget about getting a gun. You can't use a gun if you don't have a gun, either. Go non-Lethal when you are choosing what you will be using. Also get your house compartmentalized with electric locks to trap the invader, then call 911. Or sleeping gas. Bright stroboscopic lighting with loud Barry Manilow music. I don't know. Anything but lethal force. There should be ways to defend that are far more effective at stopping the intruder than a gun. There should be a huge industry with economy of scale on their side furiously working night & day to provide our citizens with non-lethal ways to stop an intruder - even an intruder who is armed with a fully blown Thompson machine gun. Now stopping an intruder with a tank crashing through your walls is probably beyond what a normal citizen could be expected to resist. Or a missile attack. But this still doesn't address the elephant in the room - the inner city shootings that are not home invasions, that are not mental wingnuts going crazy, that are not the government ATF/FBI taking issue with you....- xongsmith, 5.7d
|
|
|
Do Nothing Button
Copyright 2001-2023 by EvC Forum, All Rights Reserved
Version 4.2
Innovative software from Qwixotic © 2024