Acura TSX Forum banner

1 - 3 of 3 Posts

, Moderator Emeritus
1,003 Posts
Discussion Starter #1
Independent of all else, the color of your car might change your chances of being in a car accident, and silver looks to be best. At least, that's what New Zealand epidemiologists are presenting as the result of a two-year study of accident data in their country.

The study results, published in an article in the December 20-27, 2003, issue of the British Medical Journal, used data from 1159 New Zealand drivers (about half who had been in an accident and half not) and concluded that occupants of silver cars, compared to those in cars of other colors, are the least likely to be in an injurious accident - even when adjusted for a host of factors that could be blamed for the difference. Those factors include any significant differences in the driver's age, sex, educational level, recent alcohol consumption, use of recreational drugs, seatbelt use, driver's license status, and average time spent driving per week; plus the vehicle speed, vehicle age, engine size, registration factors, vehicle condition, type of insurance, road type, weather, and light conditions. The color white was the control color to which others were compared. Then that ratio was adjusted (with results from other studies) for each of those other driver and vehicle factors and averaged, to produce a figure that should rule out many obvious (or not-so-obvious) reasons for the differences.

Unadjusted for any of those other factors, silver cars were shown to have a 50 percent less likely chance of serious injury to occupants than white cars, and even lower-60 percent-when adjusted. Grey also fared well. On the other hand, dark and earthy colors didn't fare well at all. There was an especially increased likelihood of injury in brown, black, and green vehicles, which averaged 110, 100, and 80 percent, respectively, more than white when adjusted for those other factors (40, 20, and 10 unadjusted).

Yellow was an especially interesting case. Unadjusted, it shows a 200 percent greater chance of injury. But adjusted for all relevant factors (perhaps for typically faster or younger drivers who choose yellow cars), the chance of injury is actually 20 percent lower than white. Red was a similar though less extreme case.

But here's where it gets puzzling. Adjusted for all known factors, common sense might say that bright yellow or red cars would be best seen and avoided (compared to grey or silver), but the results of the study don't conclude that. The study does admit that "the possibility of residual confounding remains," meaning in plain-speak that there might be other variables that the study neglected to include or variables that are nearly impossible to measure, like the psychological effect a car's exterior color has on the driver's choices.

Even if you don't like silver, the moral is: get a light-colored car and you'll probably cut your chances of being in an accident. Silver is currently the most popular new-car color, and with the results of this study there's an even better reason to buy a silver car, even if we can't entirely explain why. At the very least, we're sure insurance companies around the world are asking just that.

-Bengt Halvorson

0 Posts
I saw no mention of an adjustment for how many of each color car are sold. Maybe more white cars are on the road so they get in more accidents. Also, in a different environment (read as somewhere other than New Zealand) results will most likely be different based on the environment of the surroundings, weather conditions and the like.

So what I am getting at is that study is really a bunch of BS unless it is conducted in multiple areas, etc,

HOWEVER, it is a very interesting read and it would be great if someone did a follow-up/ expansion on it. I'm definetly not trying to flame the study or article, just criticizing the fact a bit more work is needed.

1 - 3 of 3 Posts