3 ms·
There is a lot of history / context / baggage packed into a single question on the census, but I think our concern with race is fairly obvious when you consider
by forkandwait 16y ago
There is a lot of history / context / baggage packed into a single question on the census, but I think our concern with race is fairly obvious when you consider our history . The US has always been concerned with figuring out its racial composition, since slavery and all its multitude of consequences (Jim Crow, lots of poor blacks leaving the south for Northern cities) have made it so important. Think of all the money and power that used to go with whether you were white or black, or whether you could treat an employee (or slave) as white or black, or whether your state had a lot of black people or not who only got you 3/5 of a representative!
It is starting to seem anachronistic to worry over race (thank god!), but that history is really very recent; think about bussing riots in the 1970s, etc. Also, a huge number of government programs have a mission to try to fix racial disparities (successfully or not), and they need information on basic racial population composition.
That said, I can't wait until nobody cares about race anymore. The litmus test on this will be whether ("inter") marriage matters; in coastal cities among college educated people, it isn't really news if one's parents are asian and white, or latino and white, but it still elicits comment and introspection if they are black and white...
I think the big dividing lines today depend on education and family cultural background, but I am not sure how a decennial census might track that.
- MichaelSalib 16y agoI think the big dividing lines today depend on education and family cultural background, but I am not sure how a decennial census might track that. The Census Bureau also continually sends out the American Community Survey to a subset of the population. It asks much more detailed questions, so the data may already exist.
- forkandwait 16y agoRight. My point was that the job of decennial is basically to get a 100% count of bodies and residences, and that is hard enough without going into sociological detail.