Friday, May 9, 2014

statistical discrimination is convenient

in this incident, at least.

My car broke down in the middle of the road in downtown Oakland at rush hour, which while unfortunate for everyone around me, was the most convenient situation for car trouble from my perspective. Immediately two guys pulled over and jumped out to help push me to the side of the road, and then set about discussing possible diagnoses and establishing the best course of action. Jump starting didn't work, so we they decided to push me into the gas station lot across the street that luckily had a mechanic/garage that could deal with it. Going uphill is harder, so they recruited five more guys to push and a woman to stop traffic. In less than ten minutes I was calling a cab home.

For me, my gender is the least salient part of my identity, and all my interests/hobbies are male-dominated, so every time someone treats me in an obviously female-specific way it really surprises me. I've never encountered a situation in which statistical discrimination actually held me back in any way after the first impression was wiped out with more pertinent information, so I'm not offended by it*, but it's always surprising and/or amusing. In this case, as I jumped out of my car with my jumper cables and set about connecting things for the millionth jump start I've done in my life, and Guy #1 took them from me and said "you just take a seat", turning to Guy #2 for assistance instead, I did a confused double take before realizing what their first impression and statistical inference of me must be. But then I figured, you know, I really don't mind letting some guys play hero, especially if it's going to get me out of this mess in particularly short order. So sure, I'll stand here while you round up a crowd of more hero-playing men to push me across the street, no problem. The least I can do for your generous offers of help is to let you feel good and manly about it.

Life isn't fair, so might as well enjoy the times when the unfairness is a win-win.


**Don't blame Bayesian reasoning for a lack of good information!

Friday, May 2, 2014


In light of re-SMBC, I feel obligated to highlight this economics strip. Well done Mr. Wiener :)