Wednesday, May 25, 2005

More Nonsense on the Lampley "Hypothesis"

LOL. Someone claiming to be a 'stat geek' attempts to defend Lampley's Moon-landing faked delusions regarding the 2004 election as a 'sensible' take here.

Needless to say, color me unimpressed. But let's go over it one more time!

Jim Lampley’s concern over the gap between the exit polls and the final results in the 2004 election is sensible. Exit polls are, after all, used by independent election observers to monitor elections. When large gaps between the exit polls and the final results emerge, then there is a chance that the election was stolen. While we live in a mature democracy, the lust for power could conceivably lead criminals to try to steal an election.

Sure, there is a chance that the election was stolen. There is a chance that Lee Harvey Oswald was involved in a multi-faceted conspiracy involving LBJ, Nixon, the Mob, the Defense Dept, and Kevin Bacon to assasinate Mrs. Kennedy (Oswald was a bad shot, afterall). Saying that there is a 'chance' of something isn't saying anything at all. Rational, reasonable individuals must sort out possibilities that are remotely rational, reasonable, and likely from the host (near infinite) possibilites that don't meet that criteria. We are talking about a multi-state conspiracy involving multiple levels of government in which actual vote tabulations were falsified, any trace of this fraud having been covered up, and not a single leak from anyone on down the chain of involved personel. We aren't talking about 'finding' a couple of 'votes' after the election. We aren't talking about low-level shananigans (like the Dems who saw fit to slice a bunch of tires on Republican vehicles)...we are talking about a highly coordinated and directed conspiracy to steal an election. The odds of success are remote. The odds of getting away with it even more so. Note, this is before we even *get* to a question about the statistics involved. This conspiracy they're droning on about is a remote possibility even before we get to the evidence.

Since the election outcome and the exit poll are two possible views of the “correct” result, one can not prove that one is right and the other is not. One can only construct hypotheses other than fraud that might explain the differences, and confirm or reject these with the data. If no reasonable alternative hypotheses can be supported by the data, then the suspicion of fraud intensifies.

I don't know what credentials this guy has where he feels comfortable claiming the mantle of 'stat geek'...but who ever accredited him should strip him of his pocket protector and masking tape-wrapped bifocals. The election outcome and the exit poll are not two equally valid 'views' where it is impossible to prove one correct and the other not. As I mentioned in my Fisking of Lampley, the exit polls are a survey of voters attempting to *estimate* the results of the elections.

The exit poll is a preview of the movie that is the election. The election isn't a view. It is a census of the population. There is no 'margin of error' associated with it in a statistical sense. The exit poll is a survey of voters as they walk out of the booth. Hence the exit poll involves answers to questions, whereas the election results are *votes* for candidates. These aren't two views. These are two very different things. And the relationship between the two is that exit polls are educated guesses (i.e. estimates) of what that vote will look like. A preview may give you a good feel for what a movie is going to be like. A judgement based ona preview will very often be the correct one. But if you see the preview and think the movie will suck, then go see the movie and note that it is don't then turn around and think, gee, I guess there are two possible 'views' of what the movie is like...and I really can't tell whether the movie was good or not.

As to constructing alternative hypotheses, we've already seen a credible and reasonable one offered by Mitofsky and company. There are others as well. And our stat geek here eventually gets to the right place. But more on that later...

For the 2004 election, there are two main hypotheses consistent with the disagreement between the two numbers. The first is that Bush stole the election. The second is that Republican voters were unusually unwilling to chat with exit pollsters, and the exit polls were wrong. While fraud can take many different forms, making the first hypothesis quite diffuse in its implications, the latter hypothesis suggests a number of possible patterns in the exit poll data that we can look for.

Ummm, no. While the Mitofsky hypothesis is reasonable (that's the one about reluctant Bush voter respondents and eager Kerry voter respondes), it certainly isn't the only one. Dick Morris and Michael Barone argued that the Dems may have been leaked poll locations and attempted to create an artifical surge for their candidate. This small conspiracy is by orders of magnitute more reasonable than the massive vote stealing conspiracy envisioned by Lampley. There are also statistical explanations as well. A problem in the sampling frame. A failure resulting from the new methodology being employed this election (note that a previous exit polling debacle caused VNS to be dissolved and this new system put in place). The juxtaposition of the two possibilities mentioned above is nothing but a false dichotomy. Whatsmore, it provides a misleading parallelism as well. The two possibilities mentioned are not equally likely. There is no parity. One is ridiculously unlikely while the other is a reasonable possibility.

While investigation of fraud is not the main objective of the Edison/Mitofsky study, it does allow one to collect evidence that sheds light on these issues. My read of the study is that it is quite supportive of the second hypothesis.

Exactly. As I mentioned in my Fisking. This 'stat geek' should be pummeling Lampley with rights and lefts on his misleading cite to Mitofsky as supportive of his insane fraud theory. Instead he's trying to give him a pass. Go figure.

While there are many patterns in the data that suggest something other than fraud, for me, the most interesting finding is that exit poll completion rates were strikingly low for polls taken by college and graduate students. When college kids did the interviewing, a higher percentage of interviewees refused to cooperate. Since conservative speakers have a hard time speaking at a college or university these days without getting a pie in the face, it is plausible that conversations with these interviewers were not attractive to many Republican voters. (And note that this observation does not depend on assuming that the actual vote was correct).

I don't put any credence in the notion that conservatives avoided college kids because of the 'pie in the face' incidents. While conservatives had recived pies in the face in the past, the rash of such events didn't occur until *after* the election. Hence it is temporally impossible for the one to have caused the other. However, the general point here (that Bush voters may have shied away from certain demographics of poll takers) is solid and just one of the potential sources of bias (psychologists call this the "Hawthorne Effect").

Since the underlying patterns in the data support the alternative hypothesis, one need not appeal to fraud to understand the difference between the exit polls and the actual outcome. I side with Byron. Jim can keep digging if he wants to, but my guess is that no evidence of election-changing fraud will ever be discovered because there was no fraud.

He went about it in a roundabout way, but eventually he got around to the correct point: Lampley is off his rocker. Even if the guy doesn't have the stones to *say* that Lampley is off his rocker. Heh. D.GOOCH


Post a Comment

<< Home