LarryG on 7/9/2012 at 16:22

---

If a subset of the data does show a true correlation to gun homicides, that would mean that the criteria for selecting the subset drives the correlation, not gun ownership. Gun ownership per se is not related to gun homicides, not by itself. It is, as I think I said before, more likely driven by a) gun availability AND b) other operating factors (possibly poverty, possibly politics, possibly culture, possibly gun regulations, possibly, possibly, possibly, ...). But you cannot make the simplistic argument that more guns implies either more or less gun homicides. Neither of those relationships is supported by the data. If we had state by state data for gun homicides and gun ownership, or, better still, county by county data for the entire USA, then we could analyze that to see if there is a useful correlation for US gun ownership. But you would not necessarily be able to generalize any findings to any other country. The best we can say right now is that without gun availability there can be no gun homicides. But that's not a particularly useful observation.

No. That you don't know until you find a data set for which there is a strong correlation. All you can say is that for the data you looked at this was the result. That's all you can ever say. You can't make any statements about data you didn't look at, except statistically. If the data you looked at was a representative sample of a larger data set, then you can generalize it to the larger data set. But you would need to show that it was in fact representative first.

SO: we can be confident that the UNODC data gathered through its annual crime survey does not support the ideas that more guns = more gun homicides or that less guns = more gun homicides or that more guns = less gun homicides or less guns = less gun homicides. Instead it supports the idea that pure gun availability does not drive gun homicides and that other factors should be looked at to determine reasonable candidates for causative factors in gun homicides.

Jason Moyer on 7/9/2012 at 18:08

---

Worth reading for people who aren't from the US (I think most of what is said here is pretty obvious if you're over 30 and live here):

(http://en.wikipedia.org/wiki/Gun_violence_in_the_United_States#Homicides)

jay pettitt on 7/9/2012 at 19:28

---

That was Subjeff's argument though - at least I think it was. You can refine the query.

Also, it's very possible that A causes B, but C is also happening sometimes which skews the results. It doesn't have to be the other way around. It may be that possibly culture, possibly politics etc etc cause guns to be unusually safe in some places which masks the possible fact that guns cause mass murder sprees if those places are included in your survey.

LarryG on 7/9/2012 at 19:59

---

Sigh.

Jer. 5:21 "Hear now this, O foolish people, and without understanding; which have eyes, and see not; which have ears, and hear not."

Read: (http://stats.org/in_depth/faq/causation_correlation.htm) What is the difference between causation and correlation?. Just read the 1st paragraph:



Bluntly, correlation refers to any of a broad class of statistical relationships involving dependence. (http://en.wikipedia.org/wiki/Causality) Causation is just one of those statistical relations in that class. In particular it is the correlation relationship where the relationship is between a set of factors (causes) and a phenomenon (the effect), where the effect is understood to be a consequence of the causes. All causation relationships are correlations, but not all correlations are causation. (http://en.wikipedia.org/wiki/Correlation_does_not_imply_causation) Correlation does not imply causation. But causation does imply correlation. So lack of correlation means there cannot be causation.


Think. Understand.

LarryG on 7/9/2012 at 20:29

---

This discussion about cause and effect is in response to


The statistics do not bear that out. What they do demonstrate is that the rate of firearm possession is independent from whether people "fucking shoot each other." In looking for the causative factors for people shooting each other, you need to look elsewhere. That's the point.