3/26/2010

Bogus MSNBC article relating Concealed handgun laws to murder rates



Please note that the years at the bottom of these two graphs may start out the same, but they don't match up after that. About the same time that the top graph reaches 2009, the bottom graph reaches 2006. It they matched up the drop in murders would more closely line up with the increase in the number of states with right-to-carry laws. Here is what the graphs look like if they are actually lined up so that the years match.

Labels:

4 Comments:

Blogger TheBronze said...

John,

I can't see the revised chart.

I'm not sure what your point is with the disparity between the original two charts that MessNBC posted, as the number of states allowing CCW's did not change between 2006 and 2008.

Help me out with that. What Am I missing?

On the whole, I thought it was a "fairly" balanced article, although the "reporter" didn't seem to challenge the (obviously) speculative and anecdotal statements that the anti's spouted.

3/26/2010 5:02 AM  
Blogger Kevin Crystal said...

While there was a lot of bias and misinformation in the article, I was surprised that they actually gave a voice to some pro-gun positions. I almost puked however when they said there are no credible studies showing guns reduce crime. I have the second edition of MGLC on my nightstand.

3/26/2010 10:08 AM  
Blogger Al B. said...

When I read this article on msnbc.com yesterday, I was struck by the fact that the author failed to mention “More Guns, Less Crime” which is certainly the most comprehensive and persuasive treatment of the gun control issue that I’ve ever heard of. I’ll take a multiple regression analysis of a large database over an emotional argument based on anecdotes any day.

When I read the NAC report a couple of years ago that the author refers to, I was struck by the fact that they didn’t actually say that the Lott study was wrong. If I understand the report correctly, they rejected the findings of the Lott study because they thought that its statistical model was too “delicate,” i.e., when they started changing parameters, they got different results. While this may be a legitimate test of a model’s robustness in some cases, it reminds me of the tobacco companies rejecting detailed statistical analyses of over 50 years of data because it didn’t reveal a specific cause.

Unfortunately, when I read this article, I failed to note the different time scales of the two graphs, which tend to obscure the apparent correlation. While the author of the article may have made the same mistake, it’s hard to imagine that this is anything other than a deliberate deception on someone’s part.

3/26/2010 10:55 AM  
Blogger John Lott said...

Dear TheBronze:

Thanks for the post. The point is that the bottom axes don't line up. Pick any point along the bottom of the axes and you will see that the years don't line up. I lined things up properly in the second set of graphs that I provide.

I guess that I didn't see the discussion as very balanced in that the only academic that they had claimed no benefit from these laws and the didn't accurately describe the entire NRC report.

3/27/2010 10:30 AM  

Post a Comment

<< Home