Nate Silver examined the turnout at various Occupy Wall Street rallies during this past Saturday. He notes a couple of reasons why more people might turn out in one city than in another, including racial politics, organizational technology, and city political structures. Here's another idea: temperature. Does cold weather keep people indoors?
To answer this question, I used Silver's rally turnout estimates (Excel file) for the 36 largest rallies and divided them by each city's 2010 population to get a per capita estimate. Then I collected high temperature data for each city on that date. Here's the result:
Also, there's a big heteroskedasticity problem in the data, and there just aren't many cases above 85 degrees. But dammit, I collected that data, and I was going to put the scatterplot up no matter what it said.
Seth, how did you measure the size of the city? Do you think the numbers would be different if you measured the size of the downtown?
Did you find the same results if you used temperature lows instead of highs?
I used 2010 census numbers for city populations. That's not perfect, of course -- metropolitan areas might be more appropriate. I'm not sure how to go about measuring just downtown populations; often not a lot of people live downtown.
Good idea about the low temperatures, but I haven't investigated that myself.
Post a Comment