General Discussion

Recursion

(56,582 posts)

19. I do love that chart

Mon Dec 31, 2012, 10:43 AM

Dec 2012

The US is an outlier on both tails of that regression, you'll notice (re-do it without the US there and the best-fit line is actually well below Switzerland's data point -- it's important to remember with regressions that the line is created by the data points, not the other way around).

Interesting questions that come out of that are:
1. Why do we have more guns per 100 people than any other country, by far?
2. Why do we have so much more violence than our level of gun ownership suggests we should (eyeballing it, we and Canada are well over one sigma; only Argentina has a bigger variance)?
3. What policies can lower our rate of gun ownership? (before you suggest a prohibition regime, note that we have a higher rate of marijuana use than the Netherlands)
4. What policies can bring our gun violence rate down closer to the line suggested by the linear regression with other countries?

Edit history

Please sign in to view edit histories.

Recommendations

0 members have recommended this reply (displayed in chronological order):

69 replies

= new reply since forum marked as read

Highlight:

On guns, and the way the discussions are conducted. [View all] redgreenandblue Dec 2012 OP

Thanks for the inane advice billh58 Dec 2012 #1

The statistics twisting isn't isolated pipoman Dec 2012 #2

Still trying to ascribe that chart to Brady, huh? Scuba Jan 2013 #46

I can ascribe it to anyone I wish. pipoman Jan 2013 #61

Yes, proper use of statistical data is "inane." Lizzie Poppet Dec 2012 #28

Your post is a long-winded exercise in willful ignorance. Robb Dec 2012 #3

Have you comprehended my post? redgreenandblue Dec 2012 #4

plot this? farminator3000 Jan 2013 #57

Actually, 10,000 gun deaths is 10,000 more than nothing. Iggo Dec 2012 #5

Good point, one that the gun culture misses. Hoyt Dec 2012 #11

Sometimes absolute numbers are interesting and revealing in and of themselves Fumesucker Dec 2012 #6

When something scales up with system size, you need to divide by system size. redgreenandblue Dec 2012 #8

You said the numbers always had to divide by population Fumesucker Dec 2012 #12

"Argumentum ad Logicam" i.e. "Fallacist's Fallacy" redgreenandblue Dec 2012 #16

So you assumed that a group of mostly laymen would have your grasp of statistics immediately Fumesucker Dec 2012 #17

Exactly like the NRA assumes billh58 Dec 2012 #20

There may be a relationship, and I am not in fact claiming that America doesn't have a gun problem. redgreenandblue Dec 2012 #21

No... LP2K12 Dec 2012 #7

Gun huggers hate absolute numbers. 99Forever Dec 2012 #9

Here's an easy enough rule...IF there were no guns, there could be no gun violence. jmg257 Dec 2012 #10

The graph you are looking for is this one: redgreenandblue Dec 2012 #13

Thanks! nt jmg257 Dec 2012 #14

I do love that chart Recursion Dec 2012 #19

Very interesting. redgreenandblue Jan 2013 #55

Linear Corr is too subjective AldoLeopold Dec 2012 #29

It can be revealing despite its limitations Recursion Jan 2013 #58

Excellent post.... ileus Dec 2012 #15

What else do guns do to people? nt Robyn66 Dec 2012 #43

Guns don't do anything to people? ileus Dec 2012 #44

So the gun has nothing to do with people being shot? Robyn66 Jan 2013 #47

OK. DanTex Dec 2012 #18

+1000 billh58 Dec 2012 #22

Thanks. redgreenandblue Dec 2012 #23

I think the comparison to Vietnam is not entirely valid. redgreenandblue Dec 2012 #24

I was there, and you're wrong. billh58 Dec 2012 #30

I think you read my post the wrong way. redgreenandblue Dec 2012 #38

Some fair points, but... DanTex Dec 2012 #32

Great post. Thank you. nt Chorophyll Dec 2012 #34

that's more than OK! farminator3000 Jan 2013 #50

I have a problem with this whole discussion Robyn66 Dec 2012 #25

Statistics are cold, but they are what matters when thinking about policy. redgreenandblue Dec 2012 #26

Then why do you need to use statistics Robyn66 Dec 2012 #33

Because manipulated billh58 Dec 2012 #35

It is often in a person's best interest to be treated as a statistic. redgreenandblue Dec 2012 #39

But you ARE arguing in opposition to billh58 Dec 2012 #42

Can you argue a point without resorting to ad hominem attacks? redgreenandblue Jan 2013 #52

Of course your billh58 Jan 2013 #56

Because "common sense" can lead to wrong conclusions. redgreenandblue Jan 2013 #54

That's great when you arent talking about lives being lost. (nt) Robyn66 Jan 2013 #60

You are kidding, right? redgreenandblue Jan 2013 #64

Ok I give up Robyn66 Jan 2013 #68

I consider any life lost a tragedy. redgreenandblue Jan 2013 #69

Why is per capita so critical, when the U.S. is 10th highest in the world? DirkGently Dec 2012 #27

There you go. billh58 Dec 2012 #31

Among "developed" nations, we are the bottom of the barrel in more ways than one. Chorophyll Dec 2012 #36

Unfortunately, too true. billh58 Dec 2012 #40

I don't think the per capita helps you here dsc Dec 2012 #37

I fear the other driver more than any terrorist or gun, plane crash, robber, home invaders, police, L0oniX Dec 2012 #41

"The fact that over ten thousand people die from guns each year contains zero usable information." Scuba Jan 2013 #45

There is a disturbing trend on this site Robyn66 Jan 2013 #49

Actually, the 2nd Amendment "has nothing to do with hunting"..n/t pipoman Jan 2013 #62

Worst offender: Michael Moore. BULLSHIT spanone Jan 2013 #48

Only those supporting control of firearms are under this kind of scrutiny? Buzz Clik Jan 2013 #51

No. The arguments in favor of gun posession should face the same scrutiny. redgreenandblue Jan 2013 #53

Good! I would love to see.... Buzz Clik Jan 2013 #59

Or the number of billh58 Jan 2013 #63

0.0000000008.25 % of US population killed in Sandy Hook Elementary School Massacre. ehrenfeucht games Jan 2013 #65

No, because this is again an incorrect way to argue with statistics. redgreenandblue Jan 2013 #66

We could talk in absolute numbers. 20 children. 6 women. ehrenfeucht games Jan 2013 #67