Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

Callisto32

(2,997 posts)
26. Dude:
Fri Jun 15, 2012, 10:22 AM
Jun 2012

If you have a sample of 4 seconds of data from the sun, and you have 3.99999 seconds of it being X degrees, and suddenly there is a spike of X+150 Degrees, you may think something odd happened. So you look at the data you have and control things out, and find that immediately before the spike, Y happened. Based on your data, you may conclude that Y has something to do with the spike.

Then you go back and look at a year's worth of data from the sun and see that these spikes happen a couple of times, and in none of the other situations did Y occur. You must therefore conclude that Y was not necessarily the cause of the X+150 spike. It may be, but it may not be, you have a metric shit-tonne of more studying to do before your conclusions mean anything. That's all I'm saying.

This study, by not looking at the decades of available crime data is basically employing post hoc ergo propter hoc.

Gosh, it's so horrible that people are allowed to defend themselves. TheWraith Jun 2012 #1
Amen, Brother Wraith! Hells Liberal Jun 2012 #3
Another study... rrneck Jun 2012 #2
Rude practitioners of a hateful science... ileus Jun 2012 #4
Interesting. The usual suspects will arrive shortly with their head-in-the-sand denialism... DanTex Jun 2012 #5
That horse you're beating.... PavePusher Jun 2012 #7
Hmmm. Callisto32 Jun 2012 #10
Actually, they used the "differences in differences" method. DanTex Jun 2012 #11
Indeed. Callisto32 Jun 2012 #13
Well, apparently it was enough time, because at that time scale they found a statistically... DanTex Jun 2012 #15
Go look at the raw crime rate data for these states. Callisto32 Jun 2012 #16
That's not how statistical significance works. DanTex Jun 2012 #18
Dude: Callisto32 Jun 2012 #26
Dude! DanTex Jun 2012 #28
I bet you would... TPaine7 Jun 2012 #20
The pathetic thing is that no one would mind if police killed more armed people who were pointing TPaine7 Jun 2012 #22
So, it seems that some homocides that were being wrongly treated as murders... PavePusher Jun 2012 #6
Wrong. DanTex Jun 2012 #8
They aren't necessarily due to the SYG laws either. Callisto32 Jun 2012 #12
Actually, yes, according to the study, they are due to SYG. DanTex Jun 2012 #14
the very first sentence is demonstrably false gejohnston Jun 2012 #9
So more criminals are getting killed? OK nt hack89 Jun 2012 #17
this is an honest question gejohnston Jun 2012 #19
I wonder how they account for economic recession? OneTenthofOnePercent Jun 2012 #21
How to they quantify the number of predatory criminals not killed by their potential victims? SkatmanRoth Jun 2012 #23
Or...is SYG about protecting life ileus Jun 2012 #24
Interesting. Per wikipedia, the only published, refereed academic study shows 9% murder decline TPaine7 Jun 2012 #25
Details ebb and flow, but at the end of the day the probability of death remains exactly 1 slackmaster Jun 2012 #27
Latest Discussions»Issue Forums»Gun Control & RKBA»....the primary impact of...»Reply #26