Maintaining Optimism in the Face of Reality. Occasional observations on the state of the world, society, business and politics. Usually anchored by facts, always augmented by opinion.

Since You're Not There, Can We Ask You Some Questions?  | e-mail post

My new must-read blogger, and fellow obsessive-compulsive (which I mean as a sincere compliment), Jay Cost offers news and insight on a few items, starting with the media's new approach to handling exit polls this year, in light of the Voter News Service debacles with the '00 and '02 elections. For one thing, it might be a consortium, but the networks deferred to the AP's experience and put them in charge. [AP via Yahoo!] [AP Q&A from Mercury News] From the AP:
This time, the news organizations contracted with two veteran polling companies — Mitofsky International and Edison Media Research — to conduct exit polls. They agreed that the AP — which has been tallying votes in elections since 1848 — would be their sole source for vote counts, and the news cooperative has significantly beefed up its system in response.
One flaw exposed in 2000 — the failure of VNS to account for the increased use of absentee ballots — has been corrected, said Linda Mason of CBS News, an NEP [the news org consortium] spokeswoman. Telephone surveys of people voting by absentee ballots will be conducted in 13 states this year, instead of just three.
(As an aside, although this fits more with my eyetoIT writings, you can read about why VNS failed. Quick take: a tech vendor over its head and too many cooks spoil the broth.)

Jay is openly skeptical of polls so it's no surprise that Jay expresses some healthy skepticism about who these polling firms are that are to be the great white hope for exit polls, Mitofsky and Edison are both quite legit. We're not talking Zogby.

Mitofsky's the guy who invented the exit poll and founded the CBS/NYT poll back in the late 60's. He also invented random digit dialing. He's on the board of Edison as well, which was founded by a couple of guys who went to Princeton and the Penn. My recollection from my formal and mathematical logic classes is that they must put something in the water in Princeton that enhances math skills. In all seriousness, though, I don't think you could possibly get two firms that have the combined credibility and pedigree in the space. I would say that it's like appointing Paul Volker to head an independent commission, but that has a more open interpretation for some people today than it has in the past.

While Jay doesn't bring it up, I am extremely skeptical about their solution to addressing absentee ballot issues. Just as a reference point, 6.7% of Minnesota voted absentee in 2000 [Pioneer-Press], Gore won by 2.41%.

It would seem that the very idea of conducting telephone surveys, for any subset of states, for absentee voters sounds like a joke. Who are the absentee voters? OK, some older people who, likely due to health reasons, find it problematic to visit the polls do vote absentee, even though they will be in their election precint on Nov 2. I don't think you need a crystal ball to guess these people will break hard for Kerry.

Now when you consider snowbirds, deployed military personnel and people who have extended work travel assignments (e.g. employees of large consulting and accounting firms), these people aren't around to answer the phone. At first blush, it would appear that there would be so many obstacles to obtaining a reasonable and defensible sample that I think it would be preferable to not even include such samples and simply report the omission of absentee results in the numbers (and an estimate or actual count of absentee ballots) until polling sites can count and report actuals.

e-mail post | Link Cosmos | [Permalink]  |  | Wednesday, October 13, 2004
Comments: Post a Comment

This page is powered by Blogger. Isn't yours?