Lots of theories are being floated as to why there was such a discrepancy between poll numbers and actual results in the New Hampshire Democratic primary. One person has been crunching numbers and thinks there may have been a problem with the Diebold machines.
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.
I'm a big proponent of technology especially if it can improve something, however there are times where there cannot be a margin of error. Electing officials is too important for mistakes to occur in the counting of votes. It may not have happened here. Chris is getting feedback from some more knowledgable statisticians. I do think that voting machines need to have paper trails for now until they have a proven track record.
No comments:
Post a Comment