A new study grabbed headlines this week and had editorial pages buzzing with anxiety about drunken drivers ("Minnesota ranks among worst in DWIs, study shows," April 23).

However, a closer look at the survey reveals that this so-called "study" doesn't actually tell us anything about how many people are driving drunk.

The survey asked respondents: "During the past 12 months, have you driven a vehicle while you were under the influence of alcohol?"

For a "scientific study," that question is awfully subjective. It makes no distinction between whether the driver was impaired or simply drove after a having a sensible amount of alcohol. Respondents may have driven responsibly (and legally) after a glass of wine with dinner, or a beer at a ballgame.

But researchers wrongly assumed that respondents meant they were driving drunk when they answered "yes."

This study's downfall is that it fails to differentiate between moderate, legal drinking and driving and drunken driving. Two very different situations that we should be careful not to confuse.