I have not come across any studies on the accuracy of past elections. Unfortunately, I think it might be difficult to do so. For example, past elections which used mechanical lever machines for a major chunk of the population would be impossible to go back and recount. I think Douglas’s point was that each of these methods had inherent error rates, which probably made them about equally accurate under real conditions. Obviously, ballot box stuffing and that kind of election fraud was probably a lot easier to get away with back in the day of paper, write-in ballots.

I spoke with a former professor who is an octogenarian, and his conclusion was that elections are a lot cleaner today than they were in the past. For example, if you go back enough, you have Boss Tweed. But he was citing examples like Richard Daley Sr. in Chicago and that whole political machine. The problem is that when the elections are really close, there is naturally more suspicion about the accuracy and reliability of the outcome. And, many of the closest contests in our country’s history have been recent, including the Presidential elections in 2000 and 2004, the Washington gubernatorial election in 2004, the U.S. Senate election in Montana in 2006, and the Minnesota gubernatorial election in 2006.

In any case, looking forward to the next post. Looking at the interactions and processes might spark some ideas on how we can lift Douglas’ point to a greater audience.