(Not my blog, found on the web)
In contrast to exit polls, the final vote tally from the NH democratic primary shows a surprise victory for Hillary Clinton. People quickly noticed an anomaly in the voting tallies which seemed to show an advantage to Hillary conferred by the use of Diebold machines.
http://scienceblogs.com/developingintelligence/2008/01/the_diebold_effect_hillarys_vo.phpHowever, there was an easy explanation: towns with Diebold machines are more urban on average, and Hillary was always thought to have more support in urban areas. So, like many others, I was supremely irritated by the lack of analyses which statistically controlled for this obvious factor.
So I got a copy of the vote counts, and thanks to Brian London at BlackBoxVoting, the demographic information from each town (most notably, the % holding bachelor's degrees, the median household income, and the total town population). Now, Mike LaBonte at BlackBoxVoting has provided estimates of the mileage for each district, allowing for the calculation of population density.
To my complete (and continuing) amazement, the "diebold effect" on Hillary's votes remains after controlling for any and all of those demographic variables, with a p-value of <.001: that is, there are less than 1:1000 odds for this difference occurring through chance alone, and that's after adjusting for variability in Hillary's votes due to education, income, total population, and population density.
While this "diebold effect" varies in magnitude depending on the exact covariates used, it seems to center around an additional 5.2% of votes going for Clinton from Diebold machines. The same analysis shows a Diebold disadvantage for Obama of about -4.2%, significant with a p<.001, using the same covariates.......
NONETHELESS ... the general conclusion is buttressed by the following analyses, all of which have come to similar conclusions:
- Elecion Archive's analysis
http://electionarchive.org/ucvData/NH/DemPrimary2008-PairedPrecinctStudy.pdf- This one by an econ professor at Dartmouth.
http://robertghansen.blogspot.com/2008/01/new-hampshire-machine-count-bias.html- The european tribune reviews the case, with a variety of analyses
http://www.eurotrib.com/?op=displaystory;sid=2008/1/12/191247/981- An analysis using R
http://call-with-current-continuation.blogspot.com/2008/01/diebold-effect-sticks-around-need.html- BrFox's analysis
http://electionstats.wordpress.com/As you can see, something appears to be highly amiss. There may be an unmeasured third variable (it's probably not urban vs rural) or there may be something more nefarious.
Draw your own conclusions. Here are all the data files:
- The correct list of NH precincts using Diebold machines
http://www.sos.nh.gov/voting%20machines2006.htm- Mark Shauer's List of Votes in NH precincts
http://www.bbvforums.org/forums/messages/1954/NH_Primary_Data-71320.xls , Brian Fox's data of the same,
http://abaababa.ouvaton.org/fm.dat and Semmelweiss's data of the same
http://otto.ouvaton.org/new-hampshire-pack-v2.tar.gz- NH town square mileage, for calculating population density
http://www.bbvforums.org/forums/messages/1954/NH_muni_area-71354.unk- My "mega file" with all demographic information, squaremileage, and voting information.
http://scienceblogs.com/developingintelligence/ChathamNHDemPrimary.csv