Gender Profiling

After hearing a job talk yesterday by Paul Heaton on racial profiling in traffic stops, I wondered about gender profiling.  Everyone I've ever talked to has the impression that women, especially pretty young women, are more likely to be let off with a warning.  If racial bias gets people so upset, why is there little concern about this gender bias?

A quick google search finds that "gender profiling" returns 50 links, while "racial profiling" returns 4000 links.  A 2004 Massachusetts study looked for gender bias in addition to racial bias, for this reason:

Similar questions about gender disparities in traffic stops have recently been given new attention.  Following a number of highly publicized incidents of officers sexually abusing women following  routine traffic stops, some have begun to question whether certain officers use their traffic enforcement powers disproportionately against female drivers

You can hear the sigh of relief when they see their results:

Overall males were more likely to be cited than their representation in either the residential or the driving population estimate.  Males were uniformly more likely to be subject to a search and to be cited than women.  These findings were consistent across virtually all communities in Massachusetts.  This report finds no indication that female drivers, in the aggregate, are more likely to be stopped, cited or subject to a search than their male counterparts.  In fact, quite the opposite appears to be the case.

The only concerns expressed about this result was that they might have missed rare cases of harassed women.  Apparently gender biases that hurt men are not a concern; this seems another example of less sympathy to male complaints.

Added: "race profiling" gets 15,000 links while "sex profiling" gives 500. 

Comments

[JewishAtheist]

I'm always amazed that car insurance companies get away with charging men more. Can you imagine if they charged women more?

[Robin Hanson]

To be fair, dry cleaners do get away with charging women more.

[Razz]

JA, isn't insurance policy pretty financially rational though?

As for the topic I'm not so sure it's a bad thing for females to get off easier. It seems they're on the whole a lot safer drivers than men. Obviously that's not why they get preferential treatment, but a reason for why it isn't particulary bad.

As for the lack of upset I think it's both due to people having a hard time to get upset at others escaping punishment for trivial crimes, and simply because almost everyone like young women. Which isn't exactly true for middle-aged guys, since almost everyone knows a few of those they really dislike.

Finally I doubt that racial profiling in fact get many people upset, unless they're actually affected by it.

[JewishAtheist]

JA, isn't insurance policy pretty financially rational though?

I'm not arguing that it's not, just that they wouldn't be able to get away with charging women more if that were the financially rational way to go.

[Robin Hanson]

To be clear, it is of course reasonable to pull over cars that seem to be speeding or driving recklessly, and if men do those things more then of course men should be pulled over more. But once someone has been pulled over, it seems unreasonable to more often release women with only a warning.

[Peter McCluskey]

There is an asymmetry in the harm people attribute to sexual abuse versus excessive traffic tickets.
Concern over profiling is partly related to concern that a group will less political power will have less ability to defend themselves against harassment. Women are perceived to have somewhat less political power than men (connected to the fraction of influential politician who are women), and racial minorities are perceived to have less power.
After taking these factors into account, how much of a mystery is left?

[Robin Hanson]

Peter, I agree concern about groups is correlated with which groups are seen to have less political power. I'm not sure that is a full explanation or justification, however.

[Razz]

To me it seems like the basic issue is that men are willing to do quite a lot of things as long as they feel there's a non-zero chance it will help them get layed. Here's an opportunity to gift a woman with around $100 with no cost to yourself. From an egoistical point of view it seems stupid not to do it.

An alternative explanation is that women simply are safer drivers, and that their excesses typically are milder.

I reasoned that the standard deviation in dangereous driving is a lot wider for men than women. Therefore the typical man exceeding a speed limit should be a lot more dangerous than a typical woman doing the same.

Of the two explanations I find the first one very much more plausible.

Concerning lack of outcry I'd say there's a huge difference in how people think about positive and negative discrimination. A group getting extra benefits simply isn't that upsetting to most people if there's no clear victim.

[Anon]

I can suggest two reasons for the greater concern about anti-female bias:

1) Chivalry. It is thought a greater crime to hurt a woman than a man.

2) Politics. There is a well organised lobby group in favour of women (feminism). They will object to anything that is thought to harm women, but will be quieter about, or actively support, anything that appears to benefit women at the expense of men. (I am sure that this group genuinely believes it is in favour of equal treatment, but this is a site about overcoming biases.)

It is therefore politically safer to argue in favour of the woman.

[TGGP]

Haven't studies shown that, contrary to stereotype, women are better drivers?

[Mark Nau]

"But once someone has been pulled over, it seems unreasonable to more often release women with only a warning."

But the probablistic mix of offenses committed by a typical man-who-is-pulled-over may well be quite different than that for women. Assume for simplicity that cars are pulled over for either reckless driving or trivial infractions. Assume further that men commit "reckless driving" at three times the rate of women. If nearly all reckless driving cases are cited, while nearly all trivial infraction cases are let go with a warning, we'll see a highly significant difference in citation rate between the genders.

[Razz]

"Haven't studies shown that, contrary to stereotype, women are better drivers?"

I think it's *safer* rather than *better*.

On average women tend to drive a fair bit less than men as well as being generally less confident and less risk taking than men. So men tend to have more driving experience but also be involved in worse accidents due to overconfidence in their driving abilities.

It also appears that women are 1/3 as likely to drive drunk and also less likely to repeat if convicted. Suggesting a somewhat different risk profile after an infraction.

[Robin Hanson]

Razz, but its nature all discrimination is both positive and negative - some people benefit relative to others.

Mark, I agree that it is logically possible that female pullover indications tend to be more of trivial infractions; I just don't find that a plausible explanation in fact.

[Razz]

"Razz, but its nature all discrimination is both positive and negative - some people benefit relative to others."

I'm not really sure I agree with that. Seems a little too zero-sum.

Certainly the damage done to others when an attractive female is let go with a warning is very indirect and individually very small. Making it damage that's very hard to emotionally relate to. Whereas the possibility of being sexually harassed should be quite easy to relate to for a lot of women.

I'm not saying it's rational reason, it's just an explanation as to why people do not care much about discrimination that benefits a easily identifiable group rather than harming them. I'd be very surprised if even an atempt at rational thinking played much of a role in determining what engages people in these kind of issues.

Mostly I imagine discrimation is only something that becomes an issue when the damage is something a bunch of people can imagine happen to them and being pretty bad. If there's only a small net negative for every individual and only indirectly I doubt many people are going to care at all.

[Zubon]

Crash data suggests that men are doing far more dangerous things on the road, in terms of both number and severity. It is unusual for men not to be at least two-thirds of any crash problem. The latest official statistics are from 2005:

In 2005, the fatal crash involvement rate per 100,000 population was almost 3 times higher for male drivers than for females.

Males accounted for 70 percent of all traffic fatalities, 70 percent of all pedestrian fatalities, and 87 percent of all pedalcyclist fatalities in 2005.

Among male drivers involved in fatal crashes in 2005, 23 percent had BAC levels of .08 g/dL or higher, compared with 13 percent of the female drivers involved in fatal crashes.

Among female drivers of passenger vehicles involved in fatal crashes in 2005, 25 percent were unrestrained at the time of the collision, compared with 38 percent of male drivers in fatal crashes.

This makes it likely that women have more trivial traffic infractions or at least ones less likely to get them killed. If there is a bias against men, it likely arises from this difference in actual behavior.

That bias could make attributions of fault or rates of ticketing/warning somewhat suspect as a reliable measure. That is, any officer on traffic enforcement knows that young men are the most likely to be doing stupid things that cause crashes. Therefore, in the event of a crash, you assume the young man was at fault. Men should receive most traffic tickets, but it will be hard to say exactly what proportion of traffic infractions men commit.

[Douglas Knight]

Zubon:
I am shocked that males account for such a small number of fatalities. Surely, males drive more than 70% of miles. That the document does not include such numbers makes it pretty worthless; it casts doubt on the good intentions and competence of the authors.

[Zubon]

Douglas Knight: according to the latest National Household Travel Survey, men drive 64% of miles (and 61% of minutes). Note that men are still disproportionately represented in crashes, but the degree of dis-proportionality decreases when your basis of comparison moves from the total population to the driving population to the miles/minutes spent driving.

[Steve Sailer]

As Lenin pointed out, what people care about is "Who? Whom?" Not justice, logic, or anything objective. They want to see their favorites win and the people they don't like lose, and they don't really care why.