Who Drives Better, Men or Women?

We’ve established that men are more likely to take the wheel when a couple rides together, but should we care? I say we should. Aside from the cultural, sociological and psychological implications, the gender driving disparity might be costing us lives and treasure. If women are more skilled drivers than men, perhaps we’d all be better off if they were behind the wheel and men were in the passenger seat knitting. What do the data say?

Despite comprising half the population, women drivers are unquestionably involved in many fewer accidents than men are. In fact, it isn’t even close. Insurers are well-aware of this fact (see this from the Australian insurance company AAMI), and hence they often charge women lower premiums.

So there it is: women have fewer accidents than men. Time to move the driver’s seat up?

Not quite-we haven’t yet heard the whole story (and we all know that once that seat is moved? you can never again get it as comfortable as you once had it).

The problem is that if we’re trying to determine whether men or women should take the wheel when the couple is in the car, aggregate accident totals are of limited usefulness.

Why? Because men drive lots more than women do. The American Time Use Survey shows that adult men average about 60 minutes a day behind the wheel, while women average around 40. The National Household Transportation Survey reports that in 2001 men averaged 16,749 miles driven per year, and women 10,174. Given that they drive so much less, it’s not particularly surprising that women are involved in fewer accidents and pay lower premiums.

To really answer our question we need to look at data adjusted for “exposure,” that is, the amount of time people spend behind the wheel. In our case, what we’re looking for is accidents per mile driven.

Research (gated) by Guohua Li, Susan P. Baker, Jean A. Langlois and Gabor D. Kelen showed that, as of the mid-1990s, women drivers were involved in about 5.7 accidents per million miles driven. Men, on the other hand, were involved in about 5.1. Women were thus 12 percent more likely to be in crashes per mile driven. This is confirmed by another paper (gated), by Dawn L. Massie, Kenneth L. Campbell, and Allan F. Williams, which found women were involved in 16 percent more accidents than men on a per mile driven basis.

We might want to think not just about accidents, but about the severity of those accidents. Here again, it doesn’t look good for the ladies. Massie et al. found that for each mile driven women were 26 percent more likely than men to be in crashes involving injuries.

So there we have it; with men having fewer accidents and fewer injury accidents per mile, all those women driver jokes from my grandparents’ generation can be dusted off and put to good use.

Or can they be?

If we move to the next level of severity, fatal accidents, we find quite a different picture. Li et al. found that male drivers are 80 percent more likely to be in a fatal accident than women on a per mile basis. Massie et al. confirmed this, finding men are 55 percent more likely to be in fatal crashes. More recent data,?collected by the Insurance Institute for Highway Safety, paints an almost identical picture.

Thus women are somewhat more likely to be in crashes, but men are lots more likely to be in very severe crashes. According to AAMI:

Men are more likely than women to be involved in serious accidents – that is, men experience more head-on collisions, roll-overs, loss-of-control crashes and collisions involving pedestrians, cyclists or animals whereas women are more likely than men to collide with stationary objects or reverse into other cars.

So if saving lives is our goal, perhaps it’s time the women take the wheel, and we allocate transportation stimulus funds to a team of comics to craft an all-new set of male driver jokes to be used by future generations.

Well, maybe. Even given these data, a number of factors might be clouding the picture. For one thing, I don’t have data on who was at fault in accidents (one study had this but didn’t do per mile exposure). I suppose it’s possible that one sex or the other has a greater propensity to be the innocent victim in accidents, but it strikes me as unlikely.

Another complicating factor could be the driving environment. For example, Jeff Wise, author of Extreme Fear: The Science of your Mind in Danger, has blogged in this space about the fact that freeways are considerably safer than surface roads. If men do a higher share of their driving on the freeways (and I suspect that this is the case due to differing commuting patterns), this would make them appear to be less dangerous drivers than they really are. Also, it’s possible that since men do most of couples’ driving they are more likely to be behind the wheel at night when visibility is poor.

In this vein, another possibility is that the tradition of the man driving is keeping the truly terrible female drivers off the road, and thus our population of women drivers exhibits selection bias. In that case, handing women who don’t currently drive the keys might have exactly the opposite of the intended effect.

In any event, to truly understand this issue we need to dig a bit deeper to find not just the numbers, but the reasons for what’s going on. More on this coming up.

J - man driver

I wonder if the stats might also be skewed by young male drivers, especially with regard to fatal accidents. If we looked at only married men and women (presumably over age 21 or so) would the fatality difference still exist? I know that I certainly drive different now than when I was 16, especially with my female counterpart in the car.


Very interesting and a good example of why many people become so frustrated with overly simplified analyses of complex issues, especially where passions are aroused. And of course that includes any discussion of gender roles, bias, abilities, etc. Some folks are just so convinced that their gender (or other identity group) is discriminated against under all circumstances that they can take any fact or condition and use it as evidence to support their claim, or in this case they can (and will!) sieze on any one of the arguments in the article to "prove" that members of their gender are better drivers.

The opposite problem occurs as well, where someone takes a relatively straightforward fact and through convoluted arguments claims to prove that it REALLY means just the opposite to what you would first suppose. A classic example of this contorted thinking occurred in a recent NYT column where the author (apparently no longer with the Times by the way) claimed that the current recession, which has disproportionatly affected men, was REALLY harder on women. I can't even remember the "logic" used to arrive at this conclusion. Apparently, she just started with the mind-set that every measurable difference in the status of men and women must somehow disadvantage women. Men and other groups can play this game as well. I guess the lesson is to be careful when someone starts explaining what facts REALLY mean. Sometimes they just mean what they say.



Isn't it also possible that women have higher (non-fatal) accident rates because cultural biases prevent them from practicing enough? I wonder what the accident rates are for men and women who have similar driving experience, as calculated by actual miles driven. If women improve their driving skills with more practice, then perhaps it is time to hand over the keys.


The driving environment could also be responsible for why men have more fatal accidents then women. Presumably freeway driving is less likely to lead to accidents, but due to higher speeds the accidents that do occur are more likely to involve fatalities. So it men do a higher share of their driving on freeways than women, it makes sense that they have a lower accident rate per mile driven, and also that they have a higher rate of fatal accidents.


Another hypothesis; women are relatively more likely to do their driving "miles" in urban/suburban environments (commuting to jobs, carting kids around, running household errands) whereas men are relatively more likely to do their driving "miles" on highways (man = main driver on family trips, longer commuting distances).

High-speed highway crashes are much more likely to be fatal than low-speed urban sub-urban driving.

Hence, men have an apparent higher fatality rate per crash.

Do we have data that would let us test this?


If you are trying to decide who should take the wheel when both are in the car, then the relevant data would be "accident rates when the couple are both in the car".

The accident experience when only one member of the couple is in the car might be quite different for many reasons (such as, but not limited to, backseat driving).

It is only when both are in the car that there is a choice about who drives.


Two additional factors worth considering:
1. I'd bet that men are more likely to drive drunk. I'm thinking when discussing which gender is safer behind the wheel, we probably are discussing sober drivers, in which case the men are being punished by their drunken brethren.
2. Since women tend to outlive men, I wonder if the data is taking age into account. If not, the women are likely getting a bad rap by the data.

cristin ross

My husband rides a motorcycle; I drive a car. When we travel together I drive because it's my car. We each can't stand the way the other drives, but since he decided it was stupid to take two vehicles to the same destination and he doesn't drive my car, he just has to deal with the way I drive. Or walk.


I just had another thought when reading the title of this post. When trying to determine who is the "better" driver. One must first accurately describe the definition of Better. Otherwise the exercise is as fruitless as to determine who is "Christian", without defining a specific definition of "Christian".


I would also be interested to know the percentage of male drivers that end up in accidents where alcohol is related as apposed to female drivers. My guess would be that male drivers are more likely to drive under the influence of alcohol especially when with a female. Because of the prejudice against female drivers, males would be even more likely to insist on driving when with a female under these circumstances, and skew the numbers even further.


Looking at the incongruity between the time and miles spent driving per day, it looks like the men drivers have a higher speed. This could indicate that the women do more "city" driving than men. If men do more highway driving at higher speeds, then it is not surprising that they are involved in more fatalities. There could also be a professional driver component which should be muddled out of the data as well. Professional long haul drivers must comply with restrictions intended to give them adequate oppotunity to rest and may be more alert - thus skewing everything in an artifical manner.
Even if i am a worse driver than my wife, i would certainly enjoy the analysis that resolves the aggragate behavior!


my personal stereotype is that women are more cautious drivers than men, and that men are much more likely to speed, drive recklessly, or drive under the influence.

i also think that men have better control over their vehicles, are more adept at making the car do what they want, and have a better sense of timing, quicker reaction time etc. i mean race car drivers are all overwhelmingly male.


Women might engage in less reckless behavior while driving (drinking, smoking pot, falling asleep), but I refuse to believe they're statistically better than men at driving.

If a car is going 20 mph below the speed limit, I'll pass them and notice that a woman is behind the wheel. If a car is tearing through a parking lot at 40 mph, there's a woman behind the wheel. If a car's front bumper stops 1/2 inch away from your rear bumper, look in your rearview mirror and you'll see a woman behind the wheel.

I'm a 29 year old man who's never been in an accident--how many women my age could say the same thing?


Another factor to take into consideration is distractions while driving. If men do more of the driving when both sexes are in the car (as well as when any family members are in the car) it suggests that women do more of their driving while alone. Thus men are more likely to be in a distracting environment while driving - in conversation with the accompanying woman, kids, etc.


Why better, why not relatively different? I am the sole driver by necessity in my family. I know that when it comes to certain things, I find my husband to be a big help as my intellectual father and real father once were (and still are in a certain sense). With other things, however, there is no question in my mind that I am better off driving calmly and reasonably (limiting myself to the facts and only the facts) all by myself. I definately learned this art of remaining completely focused from my mother.


I agree with a lot of the other posts. Can you keep going and break it down into age groups? It would also be nice to see if the "expert" effect comes into play based on how many miles a person drives on average a month or has driven in their lifetime. Perhaps I'm an expert because I've put in my 10,000 hours of driving already.


I'm curious about the reasons women have more crashes than men (5.7 to 5.1). I'm guessin that women spend more time than men playing chauffeur for kids. Is there a difference in crash stasticis between women (or men) who drive with small children in the car and women who do not drive children?


Bingo. My girlfriend will not drive on the highway - never has. Nor will she drive on difficult roads, or roads with a lot of high-speed traffic. We live near Boston, so when we need to take Storrow Drive or Rt 93 or Rt 2, I will drive. When we need to go to the local YMCA, either of us is equally likely to drive.

It's all about weighting for degree of difficulty :-)


> I'm a 29 year old man who's never been in an accident
> - how many women my age could say the same thing?

I think we should look into the definition of 'accident'.

An accident is when something freaky happens (and it could happen to anyone, regardless of sex) - a tire blowout, a deer running into the road giving no time to react etc.
If males spend more time at the wheel, *true* accidents would happen more, and if most of that time is spent at higher speed, it would be understandable that more of those crashes would involve fatalities.

An accident is *not* something that occurs through negligence or inability - succumbing to distraction (using mobile phones / applying makeup), reversing into a post through incompetence etc.

Mike B

Why don't people take into consideration some measure of getting to a destination in a shorter period of time. Last I looked we don't live in a nanny state so why should we only evaluate utility on the basis of safety. In fact using trip performance factors in safety because an accident will incur a significant large lateness penalty and a fatal accident will incur a penalty that lasts the rest of the driver's estimated life.

Someone should definitely work out an experiment to study driver effectiveness based on travel time.