So the basic stats. City of Pittsburgh 2010 population: 26.1% African American. That does not really tell you much about voting impact. For several reasons the population under age
18, and thus not yet voting eligible, is proportionally more African American than the older population. The impact of college students and migration factor in to that, but a bigger bit of analysis. Nonetheless the result is that the
population age 18 and over is just 22.1% African American resident the city. Still far from a final answer. People tend to overlook the obvious that this
is the primary coming up. So for a primary election, only those registered in a
major party matter. How big a difference
does that make? In general, a lot even
before you talk race.
So for example. 4
years ago in the 2009 Democratic Party primary, the entire 7th ward (2010 population
13,907) had a total 1,399 votes cast. Yet
in the West End the 20th Ward (which had a very similar population of 13,261) had 2,067 votes cast in the Democratic Party
primary. Quick, without looking at a map, can you name the main neighborhoods of the 20th Ward?? The difference a result of the transient student population in
Shadyside, but also the proportion of Republican voters there as well. Voting registration patterns imply that
the preponderance all African Americans resident in the city are registered as
Democrats, so their impact is higher in the Democratic primary.
At this point folks usually point out to me that turnout among African Americans voters is lower than for others. It is a very resilient belief out there. The thing is that
even if that may be true broadly across the country, I have looked pretty closely at local voting patterns by race in Allegheny County and just don't see it at all. If anything, in some
circumstances voter participation among the African American population in the
city of Pittsburgh is higher than other groups. Especially among 'younger' voters which in Pittsburgh is everything under age 50 or so. Given the extremely low proportion of votes from those who are college students, who are less likely to be African American, the result is that AA voters are a big chunk of younger voters in the city.
Anyway, my estimate is that just about 30% of voters who show
up to cast ballots in the city's Democratic Party primary in May will be African American. Still does not get you to a final answer as
to what impact African American voters will or will not have. It depends a lot on the
pattern on the candidates of course. 4-6 strong candidates means only a shallow
plurality is seeded to win. Starting with 30% becomes
a virtual lock. Whether or not African American voters support one candidate
strongly is another question altogether.
Again, from my older work, the conclusion was that African
American voters vote lots of different ways. Sometimes they vote homgeneously
if there is a candidate that has broad appeal.
Sometimes they split the vote nearly down the middle. My calculations had the 2001 Murphy-OConnor
primary in 1999 split the African American vote as close to 50-50 as is statistically
calcuable. Literally 50-50. In the Wecht-Dawida primary for County Executive just 2 years earlier in 1999, the split
was 89% for Wecht.
The lesson, or at
least my lesson, is to not jump to conclusions that the bloc of African American voters goes any
one way all the time. Sometimes they support one candidate strongly, sometimes they don't. Sometimes they go with the ACDC endorsed candidate, and sometimes they do not. It just isn't true
that they always vote one way or the other and I presume it is a bit insulting to imply they do; even though folks who don't get AA
support always tend to think that way. The data says otherwise.
* Apologies to Professor West.