Friday, November 07, 2008


Not quite sure what to make of this, if anything. Here is a quick pass at a version of a turnout map.. This ought to show the percentage change in the total votes cast for president within Allegheny County. There is a little bit of overall population loss, and certainly population change within the region that could be reflected in the patterns on the margin. But here is what I get for the change in turnout this time around compared to the recent past. Later on I will add a link embed in the graphic to pull up a larger PDF file.

correction: ok... I corrected the earlier map which didnt take into account redistricting in a couple places which caused a few red municipalites that shouldn't have been there.

% Change in Total Votes Cast for President 2004-2008

update: The WSJ numbers guy points out that the early look at the national turnout looks to not be a record, at least once you realize the national population has been going up. It's even far more complicated than that right? When you consider the baby boom generation is aging, and elderly vote more than other groups no matter what Tuesday's data shows... it really should be the case that voter participation is going to shoot up for the next couple decades. And none of that really begins to address different turnout rates across the country or the impact of ever-expanding early voting processes. Basically, the political scientists will be parsing all this for years to come.


Anonymous Jerry said...

I see you have the "raw number of voters" layers for '04 and '08 available on your web map application -- very cool.

Is it convenient for you to post a layer showing registration rates, too? It might help in explaining how much of the disparity in raw numbers of voters can be attributed to the size of the district and how much can be attributed to, well, registration rates.

To be more concrete, I see a lot of those suburban districts had 750 or more voters (with 71% or greater turnout), while a lot of city precincts fell in the 251-500 range (with 60% or lower turnout). Is that because there are three times as many people in the suburban precincts? Or do they register at higher rates than city people?

It seems like there might be a reason to keep these precincts roughly the same size in terms of total population, although I can't really explain why, since the by-precinct count doesn't count for anything, electorally speaking. (Right?)

Friday, November 07, 2008 3:25:00 PM  
Blogger C. Briem said...

I do agree with most of that... some of which we will try to address. I honestly think we have too much stuff to put up. There is not great underlying population data this many years from the census to really get at some of that. I think we will get some more registration data up.

Is there a reason to keep precincts the same size? Depends. I will throw this out that I suspect will tweak somebody out there. the local parties elect their committees by voting district.. one male and one female per district. Is the vast disparity in the size of voting districts an equal representation issue?

Friday, November 07, 2008 3:31:00 PM  

Post a Comment

Links to this post:

Create a Link

<< Home