Driving Statistics with Gender Comparison
13 Comments Published August 28th, 2006 in Automobiles, CulturalFor those who don’t know, I am often fascinated by statistics. It’s interesting the things you can prove, and the things people think they prove by misinterpretting statistics. Here’s an example:
For young males, driver fatalities rose by 1 percent [from 1994 to 2004], compared with a 15-percent increase for young females
It’s easy to say that the younger population of females consists of worse drivers than a decade prior, but it’s more likely the case that females are driving more, which puts them into more “possible accident” situations. Statistics continually claim that young male drivers are worse, but they never compare all of the relevant factors. How much do the two genders drive? Which gender is at fault more frequently? What vehicles do they drive?
I commonly hear people saying things like “SUV’s are safer than cars.” They point to statistics that show SUV occupants are less likely to be killed in an accident, ignoring statistics from sources such as NHTSA that suggest an SUV hitting a car is six times more likely to kill the car’s occupants than if it were a car-to-car collision and four times more likely to roll. Maybe we should all drive tanks around, since they are probably even safer than SUV’s! I think the rough translation is “I should be able to kill you regardless of which of us made the mistake, because I bought an SUV.”
People often blame speeding directly for accidents. Speeding doesn’t cause accidents, it lowers the amount of time available to react to a potential accident. According to the NHTSA, speeding contributed for 30 percent of fatal crashes. The other 70 percent were caused by slow drivers in the left lane (okay, so I made the 70 percent part up, but it might have some truth to it).
Here are some more statistics that might make you think twice about pre-conceptions:
[For accidents involving trucks greater than 10,000lbs, a] total of 5,190 people died (12% of all the traffic fatalities reported in 2004) and an additional 116,000 were injured in those crashes. … In 2003, large trucks accounted for 3 percent of all registered vehicles and 7 percent of total vehicle miles traveled
For every age group, the fatality rate per 100,000 population was lower for females than for males. The injury rate based on population was higher for females than for males in every age group, except for people over 74 years old.
After an exhaustive amount of research, I finally came across a useful study that took into account gender, accidents, and miles driven! At last, researchers that think! The John Hopkins report (from the June 1998 issue of Epidemiology) says this:
Overall, men were involved in 5.1 crashes per million miles driven compared to 5.7 crashes for women, despite the fact that on average they drove 74 percent more miles per year than did women.
The investigators determined that about half of the 3.1-fold difference between the sexes’ fatal crash involvement rates was due to the fact that males’ crashes were more severe. Another 40 percent was due to the fact that men, who on average drove many more miles than women, thus had a greater opportunity of being in a crash; and 8 percent because of gender differences in “crash incidence density,” the number of crashes per million person-miles.
Regardless of what the statistics say, the majority of bad drivers I personally see are females; by “bad drivers” I mean things like changing into an occupied lane, often without ever realizing what they did. With that said, the majority of stupidly aggressive drivers are young males, probably below drinking age. They’re the kids that think giant primer-gray wings on a FWD car are cool. Regardless of where you are in the statistics, in the end, insurance rewards the terrible drivers and robs the good ones.
Amazing! The most thorough and well thought out article I have ever read. Also it was not biased. Well done sir!
Yes, I find your research enlightening! I have been driving since I was 17 and I have seen A LOT of bad drivers. I got into my first accident when I was 20 and was hit by a man. Is was only a “fender-bender” and no one was hurt. I got out of my car to see if the other driver was okay and he was still talking on his phone and then told the person on the phone that he would call them back later. There are bad drivers everywhere regardless of age, gender, race, etc. But yes, I do see more female drivers driving poorly than men but I see a lot of male road rage too. I think it’s 50/50. I have been in two accidents involving another driver. Neither one were my fault or severe but due to someone else not paying attention. I have been driving for 13 years and for the last 8 years I have been driving nearly 100 miles a day!
I think that the cost of insurance should only be based on the individual with out gender discrimination. Of course, men pay more, and they shouldn’t. My dad is 56 years old and has NEVER been in an accident!
Thank you for the thoughtful comment, Sarah. I think we’re in agreement: it should really be based on the individual. I’m 25 and still paying a fair bit for insurance (being a male and owning a sports car doesn’t help). Yet, I only drove 3000 miles in 10 months, because I walk and take the bus so frequently (public transportation here is excellent), but I bet I was (and am) paying more than someone with a “regular” car who drove 10 times as much.
Personally I have no idea if women or men are better or worse drivers, but I do know that from a child I heard the ‘women drivers’ stereotype. It’s a chicken or egg thing.
I think people shouldn’t have preconceptions. More often than not, your brain ceases on ideas that validate your preconceived notions of the world. In this case, we believe women drivers to be worse and so it registers more in our minds when a woman drives poorly because it is reaffirming a pre-existing belief.
I think it would be edifiying if you made a record of bad drivers in your car. A simple pencil attached via tape and string to a piece of paper divided into gender would give you more factual results. Try it out for a week or two and see who the real bad drivers are.
I think it would be extremely challenging to really prove one way or another. We could drive around writing down every incident of bad driving that we see, but it would be easy to subconsciously consider lesser infractions as “bad” by the group we think is the worst. Plus, in most places, males drive roughly 2/3 to 3/4 of all miles driven, so you’re statistically more likely to see bad male drivers if males and females are equal. Of course, I could easily take that bit of knowledge and really note the significant number of bad female drivers here, without recognizing that there is a slightly larger female population (probably due to the college). You would have to have very specific rules on what “bad driving” is, and you’d need to note every driver you go by, male or female (assuming those are the groups you break them into), in order to get reasonably fair percentages. Not a light undertaking by any means…
Really, I don’t care which group is the “worst.” I just don’t like hearing misrepresentative statistics, and I really hate that insurance companies base their policies on that sort of thing. The group responsible for the most accidents is the group that considers driving “down time,” i.e., time to call someone, space out, text, etc. It doesn’t matter whether they’re young or old, male or female.
Hi,
I’m interested in the statistical data you have found.
Unfortunately, the link to John Hopkins report is down.
Can you tell me where I can find this report or any other usefull report ?
best regards
Hi Ben,
It looks like they’ve changed file extensions on that server, so you can try the updated link. I’ve also updated the link in the main post. Thanks for bringing it to my attention.
Because women drive less, they have less experience and practice in that, less experience leads to more accidents.
or it’s because they are women
The Johns Hopkins study may be outdated. According to the IIHS, men have a fatality rate of 2.5 per 100 millions mile driven (2001-2002) compared to 1.7.
So, while you are more likely to get in a crash if your driver is female, you are more likely to die if that driver is male
Sorry to spam… but the link is http://www.iihs.org/research/fatality_facts_2007/gender.html
A few questions:
Did the research state how the data was retrieved/collected? Was it based on national figures or statewide figures? What was the population size, meaning did the population consist of X number of women and X number of men? This base-rate information is highly important to truly determine who would be the worse driver, men or women. It is also important to define what a ‘bad’ drive is as well. But, I think knowing who causes more accidents is also important to settling the stereotype of women being worse drivers than men. IMHO
Surprisingly it’s very hard to come by statistics that are interpreted as carefully as you do, Gordaen. So thanks for doing that and being a real statistician with imagination and insight.
As always this sort of thing turns political in the blink of an eye and becomes a question of which gender is the better driver. This of course has most value for those into gender/identity politics.
The question, however, will be one of apples and oranges to a large extent due to some other supporting statistics.
One is probably pretty well-known: that men are bigger risk takers than women. Clearly that should be expected to influence outcomes, especially in a negative way for men as they are more likely to be involved in worse accidents (as that’s where the risks are). The question would be how much such negatives are countered by other well-known superior skills of males such as reaction times, motor skills, ability to function under duress, etc.
Another lesser known fact is that men and women are hugely different genders in the following way: if you look at pretty much any natural propensity studies have shown that the male genome expresses itself more extremely than the female genome. (It’s a long story, basically, but it’s a biological-evolutionary mechanism by which weaknesses and strengths of the genome are expressed more readily in the XY chromosome build of men, whereas the extremes are dulled in the female XX chromosome build. Because women make up the reproductive bottleneck nature has been “unwilling” to experiment much with women, and thus it falls to men to behave like both heroes and villians and let women decide who to procreate with – thus keeping the genome as healthy as can be. Refer to Steve Moxon’s book “The Woman Racket” to get an introduction to this.)
In statistics this mean that the curve for females will be more bell-shaped and looking much more like a normal distribution. As such women are more average or mediocre (than men), i.e. when you look at the extremes of the female curve there are very few women at either end. This means there are few very stupid or very smart women. The spread/variance is smaller for women compared to men.
Men’s curve, on the other hand, is much more like a road bump, i.e. with fewer present in the middle (where most women are) and with a whole lot more at the really good and really bad ends. That’s probably why we find most men at both the pinnacle of society and just as many at the rock bottom of it. For every CEO of a large international corporation there’s a homeless, drug abusing, suicide prone, criminal male. So for the male population the spread is much larger and it’s not uncommon (depending on the skill or whatever is being looked at) for women at the extremes to be around 5-6 times the number of either succesful or loser males.
It’s also why pure averages are misleading when it comes to comparing men and women.
So in conclusion, if one has to determine who’s the “better driver” it becomes a difficult question to answer as we’re really comparing apples and oranges. Because one would indeed expect to find more men right at the top with at best a few women exceptions that prove the rule. At the same time some of the worst drivers are bound to be male due to some extreme expression of their genome (try combining major aggression, low intelligence and in in some of the worst motor skills that men have ever produced and you know who I’m talking about).