Paul Andrews

Ride and Prejudice, Part II: When an error goes viral

In Bicycle advocacy, Obama Bikes on August 31, 2009 at 12:25 am

The links are flying on a (supposedly) new study that (supposedly) found drivers are to blame for 9 out of 10 bike–car accidents. This situation cries out for journalistic followup, but for now let’s see if we can clear the air a bit.

A Toronto study we reported on (along with Freakonomics, which was then picked up by The New York Times and went viral from there) had some interesting findings regarding car-bike accidents. Unfortunately it got conflated with a different, and relatively dated, New York report that found drivers are at fault for as much as 90 percent of bike/pedestrian-car collisions.

First, the Toronto study, which looked at bike-car collisions in Toronto police reports.

“The most common type of crash in this study involved a motorist entering an intersection and either failing to stop properly or proceeding before it was safe to do so. The second most common crash type involved a motorist overtaking unsafely. The third involved a motorist opening a door onto an oncoming cyclist.”

Fair enough. Sounds reasonable to any experienced cyclist. No one seems to question the Toronto findings. But there’s no “90 percent” figure in them, either.

Instead, the 90 percent comes from a New York study referred to in passing by the Toronto report. And the New York study has some shall we say issues.

First off, it was based on 1997 statistics (and published in March, 1999). Not to diminish its findings from the get-go, that’s quite a while back. Hopefully progress has been made — but let’s continue.

The study, called “Killed By Automobile” and written by Charles Komanoff and Right of Way, “a group of New Yorkers who are organizing for safe streets and for the rights of walkers and bike-riders,” analyzed the year’s fatalities for “culpability” (based “largely” on New York State traffic law). The data showed that in 22 percent of the cases, culpability could not be determined. Setting aside those cases, the study found drivers to be “strictly or largely culpable” in 74 percent of accidents, “partly culpable” in 16 percent, and not culpable in just 10 percent.

Two things should be noted about the data, however. First, it incorporated pedestrian accidents as well as cycling. Typically more pedestrians — quite a few more — are killed by cars than cyclists. Although today the assumption is that texting and iPods and whatnot distracting pedestrians are to blame, statistically the difference has been fairly consistent historically going well back.

Another factor is that there are simply more pedestrians than cyclists. Whatever the reason, to lump walkers in with riders kind of jimmies the New York study from the get-go.

The New York data involve 223 pedestrian fatalities and 19 cyclist deaths. If you break out only the cyclist deaths, drivers were culpable 70 percent of the time (not 90 percent).

Second, as Richard Masoner pointed out on Cyclelicio.us, the study acknowledges it “frames crash culpability primarily in terms of driver action” rather than that of the pedestrian or cyclist. This is a bit of a head-scratcher and again begs followup. Presumably culpability should be neutral rather than “framed,” and should emerge organically from analysis of the accident itself (as it does time and again in the case studies cited in the report). The New York study, however, says it places the onus for being safe on the driver, because drivers can kill pedestrians and cyclists, but pedestrians and cyclists cannot kill drivers.

True. But to incorporate a bias into a statistical analysis based on that reasoning to my mind muddies the waters. And needlessly so. The data seem to impugn driver behavior clearly enough without any “framing.”

It’s worth reading the full New York study to get a feel for what pedestrians and cyclists are up against (veteran cyclists will not be surprised). Discounting its emotionally charged prose, the study makes a compelling case for better police analysis and judicial-system followup of walking and cycling accidents involving cars.

And that, bottom line, is why the 10 percent figure got such wide play. Skewed as it might be, it reflects the reality of experience by avid and longtime cyclists.

Of the handful of fatal cycling accidents in Seattle so far this year (which has been a bad year for cyclists v. cars), only one involved a clear-cut mistake, or at-fault action, by the cyclist. In the other three, cyclists by all accounts had the right of way. They were simply mowed down due to “driver error.”

Granted, that’s only 75 percent. But yeah, we’re in the same territory as the New York study of yore.

Morbidly enough, there are several recent “life-threatening” injury accidents — that may turn into fatalities — where the cyclist apparently was not at fault (they are still under investigation). We may make 90 percent yet.

The point — and the reason I originally linked to the Toronto account (incorporating the reference to the New York study) — is this: Society carries a prejudice, regrettably amplified by police investigations, that automatically blames a cyclist in a car-bike collision. There are many reasons for this, but it’s simply wrong. It’s not statistically valid. It marginalizes cycling as a legitimate form of transportation and is inimical to progress for cycling as a healthy and environmentally beneficial alternative to the automobile. It vitiates legislative remedies to protect cyclists. It essentially excuses and perpetuates actions which, had they involved a second driver instead of a cyclist, would result in manslaughter or homicide charges being filed.

The Toronto and New York studies may not be perfect, but until we get something better, they’re a useful step forward in recasting public perception of car-bike accidents. Hopefully studies are under way now that will further put the issue in statistically and scientifically sound perspective. The growth of cycling advocacy — both on the road and off — attests to the exploding momentum for changing society’s tired old prejudices toward cycling.

Links: The Toronto report. The New York study (click on “Research” and “Killed By Automobile”). My initial mention. The Freakonomics reference picked up by The New York Times. Cyclelicio.us gut-check. Dangerous times in Seattle for bikes.

About these ads

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: