« Ohhhh, yeah baby | Main | Remember how concerned everyone was »
October 05, 2003
Data be Damned. Hand over your gun.
Hoooo, baby. This must have been a tough report for the CDC to release - and for CNN to report. I mean, we've been told for years that "gun control laws work" and that they "save lives" and "prevent violent crime" - despite evidence to the contrary in the UK,
So, what does an objective evaluation of the data suggest?
A sweeping federal review of the nation's gun control laws -- including mandatory waiting periods and bans on certain weapons -- found no proof such measures reduce firearm violence.
Ahhhh, but these things are always open to interpretation, you see:
The CDC said the report suggests more study is needed, not that gun laws don't work. But the agency said it has no plans to spend more money on firearms study.
Yep, they didn't get the results they wanted - so it means more study is needed. BUT, they aren't going to spend any more money doing so. Is this because it's no longer considered to be an important question, or because they know that spending more money will yield the same results. The CDC has been VERY active in the gun control arena: you can bet that the study was designed to find a positive effect for gun laws, and that the data were massaged every way possible to see if there was an effect. The fact that they were unable to find any benefit, coupled with their decision to not pursue the matter further, speaks volumes about what the CDC did (and did not) find when they tabulated their data.
And, of course, the Brady Center takes the opportunity yet again to tell us that gun control laws work - even if their benefit can't be discerned by reviewing the literature on gun control laws as was done by the CDC:
A spokesman for the Brady Campaign to Prevent Gun Violence said the laws work, but it is nearly impossible to prove it [how convenient - ed] because people can buy guns in one state and carry them into one of the handful of states with strong antigun measures.
"It's hard to study whether gun control laws work in this country because we have so few of them," said Peter Hamm. "Talking about studying gun control in this country is like talking about studying democracy in Iraq."
So few gun laws? I thought there were something like 28,000 gun control laws already on the books across the US (I have no idea what the real number is, just that it is huge. If anyone can provide the real number, let me know). The Brady Center response is, however, typical of any ideologically driven argument. There is no evidence that gun control laws work. The CDC was unable to find any evidence that they work. But the Brady Center knows they work, and knows what is best for you. So they want the federal government to keep funding studies until they get the results they want. Then, once they get the results they want (from the taxpayer-funded studies that will be performed until the "appropriate" conclusion is reached, of course), they will try and get the same federal government to undo the second amendment. It doesn't matter what the data says - it only matters what THEY say. After all, THEY are the Brady Center to Prevent Gun Violence, and have some kind of omnipotent knowledge when it comes to the effectiveness of gun control in the US. Study results be damned - THEY KNOW GUN CONTROL WORKS, DAMNIT!
But, of course, it is not that gun control laws don't work, it's because the studies are flawed:
"When we say we don't know the effect of a law, we don't mean it has no effect. We mean we don't know," said Dr. Jonathan Fielding, chairman of the CDC task force. "We are calling for additional high-quality studies."
This is classic researcher mental masturbation. We all go into a research project (including this kind of meta-analysis reviewing all of the literature on a topic) with pre-conceived ideas. This so-called "researcher bias" is very difficult to control for, and why, in the biological sciences, things are typically done in a "blinded" fashion. For example, if we are testing to see if a new cancer drug increases survival, then there will be a study done that compares the effects of the drug against a placebo - and the researcher won't know which compound an individual patient is receiving. This is harder to do for social sciences research, and allows investigator bias to come through, sometimes even in the face of perceived objectivity. So, if these results had come out of a meta-analysis performed by the NRA, you would be right to be skeptical and wonder where the bias lay. However, these results came from the pro-gun legislation CDC - which is telling indeed.
Of course, the CDC spokesman tells us their negative results aren't really negative, and offers all kinds of reasons why this is so - most of which blame the quality of the studies evaluated. This is also a classical researcher two step. If the results of the study support your pre-conceived ideas - then you are perfectly willing to believe that the study is well designed and appropriately controlled. If the results are NOT what you wanted, the tendency is to look for the problems with the study - after all, it couldn't possibly be your preconceived ideas that are in error, right?
I don't doubt the CDC spokesman's claims that the studies were less than ideal - retrospective meta-analyses never are. There are a huge amount of confounding factors involved. But I would be willing to bet a small fortune that, if there had been even the slightest positive benefit shown, these results would be being presented to us as definitive proof that gun control works, and there would be nary a word about how "inadequate" the studies reviewed were.
First Kyoto gets dissed by Russia, now this. It�s been a bad week for the tree-hugging "we know what's best for you" crowd.
Posted by Nukevet at October 5, 2003 09:48 AM
Trackback Pings
TrackBack URL for this entry:
https://www.nukevet.com/rnsarchive/mt-tb.cgi/75
Listed below are links to weblogs that reference Data be Damned. Hand over your gun.:
» Linky Love from Ravenwood's Universe
Phil at The Speculist takes a look at what life will be like in the future. Phil at Delusional Duck notes that Missouri doesn't want certain people to participate in their highway cleanup program. Robert Prather of Insults Unpunished says [Read More]
Tracked on October 6, 2003 04:50 AM
Comments
What bothers me most about this, though, is that they are likely to say now that since it appears that gun control laws don't work, we need to do away with private ownership/sales of guns altogether. That's where they (Brady, CDC and their ilk) have been heading all along. It's never been about "gun control". It's about abolishment of gun ownership by private citizens.
Posted by: Cait at October 5, 2003 05:12 PM
What I find interesting is that the CDC report is essentially a modern rehash of the one done at the behest of the Carter administration that resulted in the book "Under the Gun: Crime and Violence in America" by Wright, Rossi, and Daley (https://search.barnesandnoble.com/booksearch/isbnInquiry.asp?userid=2VLJRQ2UCR&isbn=0202303055&itm=1) - which said for all intents and purposes exactly the same thing - the reports are all too narrow, too poorly done, or done with such apparent bias as to be unuseable. What they DID find was that the bias was almost entirely ANTI-gun.
The interesting thing about Under the Gun was the fact that the three researchers began their study in favor of "gun control," but by the end of their study had concluded that it was, in fact, pretty useless as a crime control measure and seemed only good for disarming the general citizenry - which they thought was a poor idea.
Posted by: Kevin Baker at October 6, 2003 11:33 AM