Nonprofit, nonpartisan journalism. Supported by readers.


How you’ve been misled on teen driving fatalities

An utterly misleading statistic has been spreading virally in the Minnesota news media and on the floor of the Legislature.

The correct statistic, which is wonderful news for Minnesota teens and their parents, is this:
Teens in Minnesota are substantially less likely to die in car crashes than are teens nationally.

If you haven’t been following the news, this may not be a big surprise. Minnesotans expect to rank well among the states in such categories. But if you have been reading the news, my good news fact may take you by surprise. You have been bombarded with prominent stories from both metro dailies (I haven’t checked MPR or the TV news, but I suspect it’s there too) asserting as an unchallenged and established fact that Minnesota has the highest teen fatality rate from motor vehicle crashes in the nation.

It doesn’t. It has the 37th highest rate (or, if you like, the 13th lowest rate).

Stats in the papers
But in the past week, flat statements to the effect that Minnesota has the highest teen fatality rate have appeared at least six times (seven times if you count twice in one story) in the two metro dailies.

On Thursday, front page stories in the Strib and the Pi-Press stated flatly that Minnesota has the highest teen driving death rate in the nation. They appeared in stories about debate in the Minnesota House over a proposal to put new restrictions on the hours and passenger loads of teen drivers.

(The bill passed the House Thursday and, by the way, nothing in this piece should be taken as an argument against the bill. I have a daughter, the last best hope for the future of humanity, who is currently on her learner’s permit. She is a good driver and getting better. But I would gladly support a law requiring all other Minnesota drivers to get off the roads whenever she is on them.)

After the Strib said it (“highest teen driving death rate in the country”) in the first sentence of its story as a flat statement without attribution, state Rep. Kim Norton of Rochester, the House sponsor of the proposed new teen driving restrictions, said it (“number one in teen deaths”) in the second paragraph. Strib editor Nancy Barnes cited it (“Minnesota had the highest teen death rate in the country related to automobile accidents”) again in a Sunday Op-Ex piece defending the paper against complaints that the front page isn’t newsy enough anymore. (And by the way, Nancy, I’ve noticed some newsier front pages lately. Please keep it up.)

The Strib editorial page weighed in on Monday (“The state has one of the nation’s highest percentages of fatal crashes involving teen drivers”). I note that the editorial writer has demoted Minnesota to only “one … of the highest percentages,” but this is also incorrect, except in a very technical sense. See below. 

State Sen. Steve Murphy said it (“We have the highest teen death rate of anyplace in the United States”) on the Senate floor on Monday and in Pioneer Press and the Strib on Tuesday. (Murphy, chairman of the Transportation Committee, is the Senate sponsor of the new restrictions.)

The meaningful stat
OK, are you ready for actual numbers on how Minnesota teen traffic deaths compare with other states and the national average?

In 2006, the last year for which figures are available from the National Highway Traffic Safety Administration, 6,037 kids (ages 15-20) tragically died in crashes in the United States out of 25,517,042 teens in the population. That’s a rate of 23.66 teen traffic fatalities for every 100,000 U.S. teens.

The lowest (best) state was Rhode Island, with 10.03 teen deaths per 100,000 teens. Minnesota, with a 2006 rate of 19.05 deaths per 100,000 teens, ranked 13th lowest among the states.

Wyoming (which, you will note, is not Minnesota) had the highest and worst rate, with 53.72. In all, 18 states — including two of our neighbors, South (39.89) and North (33.82) Dakota — had rates over 30.

So, how could Minnesota have the highest teen driving death rate in the country? Well, not in any meaningful sense. But there is a way of calculating a number that makes it appear so. I warn you, the next few paragraphs may be tough sledding.

Back in February, the Strib ran a story, and a front page chart titled “A dismal ranking.” I can’t prove it, but I believe the publication of this table was the moment when the misleading “fact” was injected into Minnesota discourse. (Who says newspapers don’t have any influence anymore?)

According to “A dismal ranking,” “Minnesota led the nation in the percentage of fatal crashes involving teen drivers from 2004-2006.”

This, apparently, is technically correct, if you take Minnesota fatal crashes involving teens as a percentage of all fatal crashes in Minnesota. But why would you do that, unless you were trying to make Minnesota teens appear to be in greater danger than they are?

Minnesota’s roads and highways are, blessedly and thankfully but not too surprisingly, among the safest in the nation. If you use the proper way of measuring highway death rates (number of fatalities as a percent of population), Minnesota (9.56 deaths per 100,000 of population) is the sixth safest of states. This is not the teen rate, but the overall rate.

In Minnesota (and, I feel confident, in most if not all states) teens are involved in more traffic fatalities than other age groups. But because Minnesota has one of the lowest overall rates of traffic fatalities, Minnesota’s teen deaths loom large only if you compare them to the low death rate among Minnesota’s non-teens. (I warned you about these next few paragraphs.)

Does this help?
Try this analogy. Suppose I told you that Twins slugger Justin Morneau had a higher rate of home-runs over the past two seasons than Alex Rodriguez of the accursed Yankees.

We look it up and find that A-Rod hit 89 homers over the last two years, compared with 65 for Morneau. Maybe so, sez I, but Morneau has the higher rate.

Oh, sez you, so you mean like number of homers per time at bat? Yes that might actually be a better way to rank home run hitters, since it takes into account how many opportunities each hitter had to go yard (that’s baseballese for hitting a home run). But it turns out that in 2006-07, A-Rod hit a homer, on average, every 13 at bats, compared with one every 18 at bats for Morneau. So if it’s rate of home runs you want, Rodriguez wins by even more.

Maybe so, sez I, but Morneau hit 25 percent of all the Twins home runs in those years. A-Rod hit only 22 percent of Yankee home runs. So Morneau had a higher rate.

But wait, sez you, that’s just because the rest of the Twins don’t hit as many home runs as the rest of the Yankees. That has nothing to do with which hitter is more likely, based on past performance, to jack one (also baseballese) each time he comes to the plate.


Morneau has a higher home run rate than Rodriguez in exactly the same way Minnesota has a higher rate of teen highway deaths than Wyoming.  

Teen traffic fatalities: state by state

NCSA’s Information Services Team

Comments (7)

  1. Submitted by Tony Wagner on 05/02/2008 - 01:55 pm.

    Excellent work, Mr. Black.

    It’s amazing how one twisted little nugget of misinformation can lead to so much wasted time, money, and energy, both in the media AND in the legislature. Shame on both of them!

  2. Submitted by Michael Ernst on 05/02/2008 - 02:08 pm.


    It’s the National Highway Traffic Safety Administration, not the National Highway Safety Traffic Administration.

    Nice article!

  3. Submitted by Bob Collins on 05/02/2008 - 02:11 pm.

    There’s a fundamental flaw in your research.

    You attempted to debunk Kim Norton’s assertion (which, by the way, was that we had a high rate of teen *crashes* and fatalities involving teen drivers) by pointing out the number of TEENS killed.

    However, according to research, the majority of people killed in “fatal accidents involving teen drivers,” are not the teenagers.

    According to AAA Minnesota, “between 1995-2004, crashes involving 15-, 16- and 17-year-old drivers claimed the lives of 567 people in Minnesota, of which 212 (37.4 percent) were the teen drivers themselves. The remaining 355 (62.6 percent) included 171 passengers of the 15- to 17-year-old drivers, 155 occupants of other vehicles, and 29 non-motorists.”

  4. Submitted by John Olson on 05/02/2008 - 03:32 pm.

    I think the general point here Bob is that the case that is being made has been built on (perhaps) information that has been manipulated.

    Where does the “Nanny State” stuff end here?

  5. Submitted by Bob Collins on 05/02/2008 - 05:44 pm.

    I certainly understand the point, but in order to make it you have to first misstate the original assertion, which — as I indicated — was that the rate of crashes and fatalities in accidents involving teen drivers — not that the number of teens killed in teen drivers led the nation. Norton didn’t say what the Strib said she said.

    Here’s Norton’s ACTUAL assertion:

    “Minnesota has the most teenagers in the nation behind the wheel in deadly crashes.”

    Focusing on the number of dead teenagers doesn’t disprove that. You’d have to total up the number of deadly crashes involving teenagers — not the number of dead teenagers.

  6. Submitted by Eric Black on 05/03/2008 - 01:06 pm.

    Hey Bob,
    There are several statistical ways to measure what we’re talking about here.
    Measuring the percentage of all Minnesota accidents that involve teen drivers is one of the least helpful and most misleading, since it is clearly being used to suggest that Minnesota is the worst in the nation in some category of traffic safety involving teens.
    It isn’t. As the data in my post suggests, Minnesota is among the safest states, for teens or non-teens.
    The statistic on which I concentrated is the number of teen fatalities in traffic accidents as a percentage of all teens in the population. This, by the way, picks up teens who died in these accidents who weren’t driving, which would include many of the passenger deaths to which you refer. I think it’s the best measure and most honest way to measure a teen traffic fatality rate for comparison with other states.
    Minnesota’s overall (not just teen) traffic fatality rate is even better, compared with other states.
    I don’t take any of these as valid arguments against Rep. Norton’s bill, which would almost certainly save some lives.
    But I do take the constant citation of the worst in the nation stat as misleading.
    I also don’t know who, among the many who cited the misleading stat, did so with knowledge or intent to mislead. But it is misleading, at least to the extent that it suggests that Minn is the most dangerous of states for teen drivers — and even for non-teens who share the roads with them. It is not. cheers,

  7. Submitted by John E Iacono on 05/04/2008 - 03:52 pm.

    While I can see the difference Bob points out, I note that he does not assert we would be the worst state if his measure were used.

    And I am impressed with Eric’s research.

    As a father who steered two daughters through the shoals of learning to drive, I can appreciate the concerns of other parents as fledglings learn to fly, and legislators’ efforts to help.

    It seems to me, however, that these efforts are somewhat off the mark.

    More likely to save teen lives, it seems to me, would be a law making tailgating an offense with a large fine — this seems to me to be the most dangerous and common offense I notice with teen drivers. They can’t seem to grasp that they might not be able to stop.

    And targeting cell phones and texting similarly would certainly affect teens, but also a number of others who frighten me when I see them thus distracted in my rear view mirror.

    Of course this would be less popular, as a number of adults commit the same offense, but teens seem quite unaware of the risks they are taking — until they have an accident.

Leave a Reply