This will be a short diary, because there’s not much to say. As some of you already know, I’ve been compiling 2008 Obama/McCain precinct-level data for Illinois and translating it into the format on Dave’s App. The Illinois Project is essentially done and the data will hopefully be available on Dave’s App soon.
So, my question then is: Which state should I do next? I was planning on California, but I’m not set on that. My preliminary survey tells me that the six big states below are viable candidates (though I’m not entirely certain on Florida). Since this can be very time-consuming (though I doubt any of the states below will be as challenging as Illinois), I figured I’d run this poll to decide which state I should work on next. Happy voting!
PS. If anyone knows where I can get precinct-level results for New York then please let me know! (besides contacting county clerks, of course)
Loading ...
My vote is for Florida for the following reasons:
California – has a statewide precinct file at UC Berkeley’s redistricting website.
Michigan – has a statewide precinct file at the Secretary of State’s website.
Ohio – has a statewide precinct file at the SoS’s website.
Pennsylvania – has a statewide precinct file via CD purchased from the Secretary of the Commonwealth.
Texas – has a statewide precinct database at the Texas Legislative Council’s website.
I have the statewide Michigan, Pennsylvania, and Texas datafiles.
New York I have to at least the township level.
Florida would be my choice.
Thanks for your work on Illinois!
😉
No really, I voted California. It will be nice to see the updated precinct data.
But hopefully, pretty soon the rest of us (Nevada, Arizona, Colorado, etc.) out West will get some love.
That’s one I really would like to see.
I thought Texas, California, and Pennsylvania were already there.
there is TONS of data in the SSP document collection, including entire states like MI, NV, AL, etc.
i’m assuming you already know this, but just in case you don’t, it’s all there.
anyone else reading this, maybe someone like atdleft wants to do NV.
http://www.scribd.com/collecti…
Especially to see what the Republicans can get away with if they at least make a minimum gesture at following the new guidelines (frankly I don’t think they much have to).
Barring a shocking California surge, it seems Florida has this one locked down.
I’ve already tackled the first 7 of 67 counties and have precinct results on my spreadsheet for 5. That means 2 of 7 counties that I need to contact directly. So, this looks quite promising and doable.
As for translating the data to Dave’s App, I’ve only glanced over the relevant counties but so far it doesn’t strike me as too problematic. Certainly not so much of a challenge as Illinois.
My biggest concern, obviously, is the heavily populated counties, and the one that I’ve already looked at (Broward County) seems easy enough.
So, if all goes according to plan, I should have a list of counties to contact by Monday at the latest, and then if I get the requested data swiftly perhaps I can get Florida done by the end of next week.
Since I’m now also set up to receive faxes again, the less that I need to have data sent by mail. When I started the Illinois Project I received two faxes before the machine quit working. >sigh< That didn’t help matters..
I’m not sure if you’re aware of this, or if it’s even useful, but SSP crowdsourced election results for most states in 2008. Florida is here.
You can find links to everything else by scrolling to the bottom of this link.
Send me and Jeff an email. I’d hate to see you duplicate our efforts from 2008/2009. We collected precinct data for every split county that we needed in order to do compile pres-by-CD for all 50 states. We have a LOT of data. Many people have already pointed you in the right direction, but I think I have some stuff that’s not uploaded as well.
The shape file that the statewide database has on their website is pretty poor. The tabular database its self is pretty good. I’ve compared SWDB’s precinct file to GIS data I’ve gotten directly county GIS dept’s and the geographies frequently don’t match. Its crap for anything but macro statewide analysis. large numbers of polygons are unattributed and/or don’t join. I don’t really blame UC Berkeley for this, they’re dealing with data that comes from 58 different counties that is filtered through Sacramento. Many of counties don’t even have GIS systems of their own. so, CAVEAT EMPTOR.
For anyone keeping track here, I’ve now sent Illinois, Nevada, and Arizona to Dave for upload to the Redistricting App.
Arizona was by far the easiest of the three.
Over the next week or so I hope to do Florida, Minnesota, Wisconsin, the Carolinas, and Georgia.
That should keep me busy!
I’ve been sorting through Florida data in preparation for the transfer to the csv file from Dave’s App. Here’s where things stand:
#1) I will not be including an “Other” tally for Florida. In short, much of the data I have combines undervotes, overvotes, and/or blank ballots into the “Other” column. This obviously won’t work for my purposes. In order to fix that, I would need to get canvass reports for a number of counties and enter the figures manually, precinct-by-precinct for 11 minor candidates. In my view, this isn’t worth the effort or the weeks of delay. The minor candidates added up to 0.8% of the 2008 presidential vote in Florida, so hopefully no one will be too tore up that I exclude those votes.
#2) The “Official Results” reported on the Florida Secretary of State website differ from the figures on the final canvass by precinct for a number of counties. I don’t know why this is (it is Florida after all) but the short of it is that obviously I’ll be using the county precinct canvasses.
This includes the following counties: Brevard, Broward, Columbia, Franklin, Hernando, Lafayette, Manatee, Marion, Monroe, Palm Beach, Pinellas, St Johns, Volusia, and Wakulla.
The differences are not significant – Palm Beach has the greatest differential with 65 Obama votes and 37 McCain votes – but I’m just noting this for the record.
#3) That leaves the following counties with issues that I cannot resolve online: Baker, Clay, Gadsden, Glades, Hamilton, Seminole, Union, and Washington. I will contact these counties on Monday to clarify results or request their election data. Again, many of these involve less than two dozen votes, with Glades being the most noteworthy exception (several hundred votes).
So, in the meantime, I will finally begin transferring the Florida data that I can work with. If all goes well, I hope to have Florida done by the end of the week.