Wintemute/UCDavis DUI v. Later Crime Study: Methodology

You know the one; the anti-rights mainstream media is hyping it as demonstrating a strong link between prior DUI convictions and later serious violent crime offenses by law-abiding firearms owners.

I’ve already addressed the horrifically Orwellian privacy violations* by Wintemute and his junior police-statists. That distracted me from my original intent to look at his methodology. Looking at Wintemute’s methodologies is always an exercise in morbid fascination.

As usual, he abused the process to generate a foregone anti-rights conclusion.

What he did

Working from DROSs, the study found 160,619 handgun purchasers. They preselected 79,678 of them by excluding various groups, notably anyone 50yo or over because “violent crime rates among persons 50 years or older are relatively low.”

The study preselected for those most likely to reoffend.

That left 79,678 (less than half the total N of HG buyers; 49.7%). They managed to find 1,495 with prior DUI convictions. That’s 1.9% of the selected study population, which you must recall was already pared down to the most likely reoffenders.

Of the 1,495, just 131 were later arrested for “Crime Index-listed violent crimes.” They call that “8.8%,” but properly it should referenced as 0.16% of the study population. Their “confidence intervals look suspiciously large,too, but I’ll leave that question to trained statisticians.

It appears that they just proved that, even in the worst age groups, a tiny fraction of one percent will reoffend. That should be reassuring, proving how honest lawful firearms owners tend to be.

But wait. There’s more. The 131 is people charged or arrested. Not convicted. They specifically rejected checking for actual convictions, so we have no way of knowing how many charges were reduced, dropped, or exonerated. By looking only at charge/arrest, they lassoed in more more to raise the percentage.

Please consider that this is California. I guarantee that in the majority of counties, even a justified self defense shooting will result in an initial arrest pending investigation.

Wintemute’s bogus study is virtually certain to include valid self defense incidents as part of the miniscule 0.16% “offenders.”

No wonder the charlatan effectively covered his eyes and chanted, La la la. I can’t see convictions or exonerations.

What he should have done

If Wintemute really wanted to examine any “Association of Prior Convictions for Driving Under the Influence With Risk of Subsequent Arrest for Violent Crimes Among Handgun Purchasers,” he had an opportunity to be honest.

He should have looked for all DUI convictions for a given, pre-study period; say, 1995-2000. Then he could have — privacy issues aside for the moment — picked out the 160,619 handgun purchasers in 2001.

Then, over the study period of 2001-2013, he could have compared two groups: non-HG buyers with DUIs and HG buyers with DUIs. Which group had the higher rate of later violent crime convictions? How does each group compare to the general population?

My prediction, based on other studies: He would have found that lawful gun owners are far less likely to commit violent crimes than the general population.

Of course he didn’t want to look at that. Or publish it.

* Someone more familiar with California privacy laws may know for sure, but I suspect that Wintemute’s official status as a state-funded director of “gun violence research” covers him legally on obtaining the personal data. It’s still morally repugnant.

The fact that he tracked personally identified individuals in a study without obtaining their consent would appear to be — at least — a violation of scientific research ethics protocols. And my opinion of that is that it is [expurgated even for this blog].

If you found this post useful, please consider dropping something in my tip jar. I could really use the money, what with ISP and web host bills, and general life expenses. And the rabbits need feed. Click here to donate via PayPal.
(More Tip Jar Options)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s