“North Dakota tops list in study for percentage of drivers with a DUI history.”

That recent headline was over an article describing research done by an insurance quotes website. Which is often the case with these sort of articles.

Companies hawking things like insurance or credit cards produce these “studies” ranking states on one metric or another. They announce them in press releases hoping reporters, always thirsty for easy copy, will write stories about them thus boosting web traffic and name recognition for the companies.

They’re an exercise in marketing, not data science, and that would be fine except these news articles often end up getting used in real world public policy debates.

Which is problematic because, believe it or not, DUI arrests are not really the goal of DUI policy.

The reason why inebriated driving is illegal is because it’s unsafe.

The goal of DUI enforcement is safe roads. It is not seeing how many people we can arrest.

The arrests are a means to an end, not the end itself.

Which is not the impression you get from articles comparing states based on DUI arrest statistics. Statistics which can be influenced by variables which don’t really have a lot to do with how safe our roads are.

Allow me to illustrate.

The government often incentivizes DUI enforcement with grants for extra patrols or other types of enforcement.

A state which provides more incentives for DUI enforcement will inevitably see more arrests which are then measured in shallow marketing “studies” which, in turn, make headlines.

Leaving, paradoxically, the impression that states committing more resources to enforcing drunk driving laws is less safe than a state doing less because they have the most arrests.

See the problem?

North Dakota does a lot of DUI enforcement. We’re lucky that we’re generally a very low crime state, and our law enforcement officers have the resources to commit to it because they aren’t dealing with the sort of workloads their colleagues in other parts of the country contend with.

That reality also drives up DUI arrest rates.

If reducing DUI arrests is the goal, we could do so by reducing DUI enforcement.

Absurd, I know, but so is measuring the problem in terms of arrests.

What we should care about is data relating to alcohol-related traffic incidents. How much death, injury, and property damage is caused by drunk drivers? Are those numbers going up or down?

That’s what we should care about, because those metrics measure safety, and safety is the goal.

Sadly, the arrest numbers get the attention, and what we end up with is public policy responding to arrest metrics and not safety.

Rob Port, founder of SayAnythingBlog.com, a North Dakota political blog, is a Forum Communications commentator. Listen to his Talk Podcast and follow him on Twitter at @RobPort.