November 5, 2002

3:03 PM

Voter News Service has abandoned its state and national exit polls.VNS, a group formed by the major media (ABC, CBS, NBC, CNN, Fox and Associated Press) to provide exit polls, rebuilt it's computer system after the 2000 media debacle, which VNS was blamed for. The computer system still did not work today. Rather than face embarrassment again, VNS has decided to not give any projections to the media.
posted by Steve_at_Linnwood (40 comments total)

The separate vote-counting system is unaffected... so far.
So now there is no independent check should parties unknown want to steal the Senate.
posted by Slagman at 3:11 PM on November 5, 2002

Generally, consortium members expect the first wave of exit poll data by 2 p.m. EST, giving them the first idea of how the election is going. That hour came and went with little useful information
posted by thomcatspike at 3:17 PM on November 5, 2002

I am the VNS reporter for Washington Co., Minnesota, and as far as I know, I'm still on the job. I just actually called the central office in New York to talk with my state manager. I'll ask him what's going on with the exit polls, and I'll give you the news straight from the horse's mouth when I hear about it.

I must add that I am merely reporting the tabulated totals from actual ballots counted. I am not in any way involved with exit polls, which I do not agree with on principle. I actually think they are unethical, antidemocratic, and run counter to our values as a society. So in that way, I'm glad that my employer, VNS, has dropped it's exit poll projections.

Whether the national news media will conduct their own exit polling, I don't know. I hope they don't. Just to be safe, I won't be watching the news tonight.
posted by Hammerikaner at 3:21 PM on November 5, 2002

Cool, a poller who disagrees with polls. You meet all types out here.

Still, thanks hammerikaner- updates eagerly awaited.
posted by gsteff at 3:27 PM on November 5, 2002

[So now there is no independent check should parties unknown want to steal the Senate.]

Nonsense - I hear the AP has it's own version of this, and I'm pretty sure Foxnews and NBC have a duplicate system other than that.
posted by revbrian at 3:28 PM on November 5, 2002

good riddance to them (no offense Hammerikaner), but after 2000 it should have been dismantled....and to have a Bush cousin (Ellis) on the staff should never ever been allowed!!
posted by amberglow at 3:33 PM on November 5, 2002

AP and Fox I know have their own systems, but for some reason they are saying that they are not as reliable as VNS... Though the reason is beyond me...
posted by Steve_at_Linnwood at 3:34 PM on November 5, 2002

[So now there is no independent check should parties unknown want to steal the Senate.]

I wouldn't have trusted VNS for this anyway. Whatever weaknesses we saw in election law/procedures in 2000, the weaknesses in VNS seemed much more evident.
posted by originalname37 at 3:46 PM on November 5, 2002

I just got off the phone with my State Coordinator at VNS (who also happens to be my journalism professor and newspaper advisor) and he gave me the straight word on what the people at VNS are thinking about the exit-polling.

"This has been up in the air for the past few months," he said. VNS doesn't feel confident in the exit-poll data, because this kind of data is used to report races that aren't even close. It really shouldn't have been used in Florida in 2000. Well, it should have been used, but since it was so close, VNS should not have made projections based on the data.

He said VNS is still going to project from actual tabulations, which are what I will be reporting tonight. If they can make a safe projection, they will make a projection.

My manager also told me that Minnesota is looking to be the most closely-watched elections nation-wide. I told him I wouldn't be watching television because of the way the media scrambles to get the first word of apparent results, even if they're inaccurate and misleading. He said avoiding tv would probably be a good idea.

I'm a newspaperman at heart, and I can't stand the way elections are reported on in this country. I won't watch the news, I won't try to find out who won (well, I'll know who wins in Washington Co., Minn.), and I'll sleep soundly and just pick up the newspaper in the morning.
posted by Hammerikaner at 3:49 PM on November 5, 2002

Margin of error of a poll is dependent on sample size. I would think AP and Fox would be doing less exit polling than the VNS, so less reliable.
posted by smackfu at 3:50 PM on November 5, 2002

(Sorry Hammerikaner, I should have said VNS exit polling.)
posted by originalname37 at 3:52 PM on November 5, 2002

originalname37, if you think about the way things went in Palm Beach, Florida, in the 200 election, perhaps we can see it quite differently.

Perhaps VNS actually reflected the will of the voter in exit-polls more accurately than the faulty butterfly ballots and pregnant chads.

But of course, I don't agree with exit polls... so why am I defending them?
posted by Hammerikaner at 3:52 PM on November 5, 2002

So, Hammerikaner.. how's Mondale doing in Washington Co?
posted by PrinceValium at 3:57 PM on November 5, 2002

Oh, boo-hoo. CNN is going on and on and on about this VNS "drama," as if it's the biggest thing since the Sniper and the Search for Chandra combined.
posted by raysmj at 4:05 PM on November 5, 2002

I want to know who's winning. Who wants to vote for a loser?
posted by monkeyman at 4:06 PM on November 5, 2002

Someone who's planning to stay up all night waiting for the television folk to declare a winner, please explain: why can't you just go to bed and read/hear/see it in the morning, like our man on the inside Hammerikaner plans to do?

I can understand if you're a campaign worker, or a poll worker, but if you're not, why not just get some sleep? The people who officially win tonight will still have won tomorrow.

Maybe it's obvious to everyone else and I'm not sufficiently obsessed with politics to get it.

I'm looking forward to 24 being all tiny and over in the corner tonight while results "with 4% of precincts reporting" are splayed all over the screen. [grumble]
posted by jmcmurry at 4:08 PM on November 5, 2002

PrinceValium, I haven't gone out to the county seat yet. I'm still at my university in St. Paul... so I have no idea how Mondale is doing.

I was told by the election officials at the county seat that they don't expect tallies from the various precincts untill about 10:00 or 11:00 p.m. By that time, I will be out there and most likely won't have internet access... sorry.
posted by Hammerikaner at 4:23 PM on November 5, 2002

Hammerikaner, what exactly will you be doing, besides telling VNS who will then tell the networks and cable and papers and wires- they're saying 532 for Mondale, 524 for Coleman with 5% of precincts reporting in....and then 575 for Mondale, 586 for Coleman with 6% of precincts reporting in.......? (i'm curious)
posted by amberglow at 4:27 PM on November 5, 2002

jmcmurry: at least here in finland it's pretty exciting to see the results come from all over the country.
Off-Topic went to the 24 website to check out some episode synopses and stuff, and the first page of the episode/character basically gave away everything that happens in season 1 (i've seen four episodes). Great.
posted by lazy-ville at 4:28 PM on November 5, 2002

lazy-ville: Sorry about that. :) If only I'd known...
posted by jmcmurry at 4:33 PM on November 5, 2002

Does anyone know where Matt Welch is, so we do not have to argue about it later.
posted by thirteen at 4:43 PM on November 5, 2002

amberglow, what ever crap they're saying on the television about 'precincts reporting in' is based on independently run (and horribly inaccurate) exit-polling.

the precincts are absolutely not reporting in because the polls have yet to close here in minnesota. they close at 8 p.m. once again, the networks are just trying to create news by putting their 'reporters on the ground' to get this exit-poll data.

you think they would learn that if VNS, the experts, don't trust exit-poll data as being reliable, then they wouldn't do it for themselves... but heaven forbid the networks do anything accurately, fairly, or ethically.
posted by Hammerikaner at 4:47 PM on November 5, 2002

originalname37, if you think about the way things went in Palm Beach, Florida, in the 200 election, perhaps we can see it quite differently.

You're right. I was just talking about the back-and-forth projections on election night. Also, thanks for the inside info.
posted by originalname37 at 4:53 PM on November 5, 2002

Alright, I broke down and turned on CNN. Naturally, they're scrambling to say something without having anything to say. How typical. Bob Woodward is on Larry King, agreeing with VNS's decision to not use exit-poll data to make forcasts.

Oh, and there's a helpful little clock in the lefthand corner of the screen counting down to when the next round of polls close across the country! Taking up yet more screen space on the television. Yippy!
posted by Hammerikaner at 4:54 PM on November 5, 2002

Hammerikaner: Please correct me if I'm wrong, but it isn't the exit poll data that is inaccurate, but the program that they use to analyze the data that they don't trust right now. Is that an accurate appraisal of the situation?
posted by Plunge at 5:05 PM on November 5, 2002

thirteen: rofl.

This is drama -- for the networks. Without VNS they can't do their projections and red state-blue state maps in prime time, when people are watching. It's not that important for the rest of us.

As for the "independent check" business ... we survived the first 3/4 of the republic without this questionably accurate "independent check", which itself seemed to work fine until there was a close election. Oops! The transparency and accountability in our election system is provided by, among other things, election judges from both major parties. It's not foolproof, but they have access to the actual ballots and voting machines, and the media don't.

I don't get all up in arms over exit polls -- in fact, for the first time, I actually answered one today. Perhaps this news was why the pollster looked so forlorn (or maybe it was just the rain and her wet stocking cap). But I also tended to agree with Royko and others who've urged people to lie on exit polls.

Plunge: Several of the online stories do seem to point at the software, but it's not clear whether this is a merely technical issue, or a broader one of how to use data analysis. I expect we'll know by the time it's completely moot, like tomorrow ...
posted by dhartung at 5:09 PM on November 5, 2002

Plunge, The real problem, as my state manager at VNS told me, is that exit-polling is only really effective in elections where they are not really, really close.

As I understand it, the system is flawed because in a close election, a fairly 'certain' result could flip the other way once more polling is done.

Think of it this way: If a candidate is ahead by only a hundred votes half way through one's exit-polling window (sort of like in FL in 2000), perhaps the next four hundred people polled favor the other candidate more heavily than the last batch of voters. As my VNS contact explained to me, exit-polling really only works if, say, the candidate in the lead was in the lead by so much that next batch of voters to be polled would have to be in favor of the other candidate in the range of 85-90% or some similarly enormous figure in order for the lead to flip.
posted by Hammerikaner at 5:16 PM on November 5, 2002

Alright folks... sorry to leave you, but I've got to head out for the county seat.

I hope you all have a nice evening... and may this election be fair and, well, accurate!
posted by Hammerikaner at 5:18 PM on November 5, 2002

have fun! it'll be a long night for you hammer!
posted by amberglow at 5:29 PM on November 5, 2002

Everyone uses VNS, including the Associated Press, Fox, etc. There is no separate exit polling, except on a spot basis. CNN might have some polling, for example.

The VNS employee who is posting here is participating in the actual estimated vote tally, a separate process from exit polling that is completed after polls close.

Official tallies by election boards are not available for several weeks. Unofficial tallies are what you see in your local media.

It's pretty pathetic that an operation designed to do one thing did not even test its system in advance.
posted by Slagman at 5:43 PM on November 5, 2002

Well thank goodness I am reporting from the most wealthy county in Minnesota... they actually have internet terminals set up, which reporters can use to get the updated poll results. Now I can stay connected with you all throughout the night.

If you're interested, I'll give you our numbers here from Washington county as they come in (or you can just get them for yourselves). Results from the senatorial race will not be coming in any time soon, since Senator Wellstone's death has necessitated supplemental ballots that must be counted by hand.

At the moment, I am the only person sitting here in the lobby outside Washington County's elections office in Stillwater, MN. All the governmental offices are contained within one, beautiful multi-story building on the eastern side of the county seat.

A 'stringer' for the Associated Press just showed up here to join me at the computers. She will be reporting results into the AP. Although the AP is a member of the VNS, they also send out reporters to get all the election results for every county. You see, VNS only collects the information that is relevant to the entire nation (Governor, U.S. House, U.S. Senate), and the AP collects election results from the high and mighty down to your local dog-catcher.
posted by Hammerikaner at 6:07 PM on November 5, 2002

So now there is no independent check should parties unknown want to steal the Senate.

You noticed that, eh?
posted by rushmc at 6:50 PM on November 5, 2002

Hammerikaner, you are a true MeFi hero. Give that man a pony!
posted by languagehat at 9:14 PM on November 5, 2002

hammer, is Mondale really losing? (i've been out)
posted by amberglow at 10:53 PM on November 5, 2002

Um... here are the latest results... right now it's about 50% for Coleman, 45.6% for Mondale, 2.5% for Moore, and .45% for Tricomo. Wellstone, from the write-in ballots, has 3706 votes (actually beating Tricomo).

The results for Coleman/Mondale/Moore/Tricomo are just trickling in because they are all being counted by hand. These figures reflect 65.34% of the total precincts, which is actually pretty good for this time...

I'm sad that Ray Tricomo is doing so poorly. He's a fixture at my University, and there's a "reading to ray" program which helps him connect with students and vice versa. You see, he's a blind man, which may not be common knowledge. He's described as both "sweet" and "crazy" depending on who you talk to.

I'm still in Stillwater waiting for the final results in the state's US House elections, the Governor's race, and, of course, the Senate battle.
posted by Hammerikaner at 11:22 PM on November 5, 2002

thanks hammer--are people attributing it to the memorial or just "a time for a change" ?
posted by amberglow at 11:26 PM on November 5, 2002

I don't think the support for Coleman has much to do with a negative reaction to the Wellstone memorial. I think it has much more to do with Coleman's apparent succes as mayor of St. Paul over 8 years. Time and time again, he trumpets his record of not raising taxes during his terms in office.

What you might find surprising is that Coleman was never elected to office as a Republican, but rather as a Democrat. Shortly after being elected to his second term, he did a quasi-Jim Jeffords and jumped ship from his mother party. He seems to have found his niche among the conservatives, and if he is Minnesota's next Senator, he will fit nicely into the President's pocket.

So maybe this is a "time for a change" for Minnesota. Perhaps Mondale is just too old. Despite his experience in national politics, people might favor the young, virile Coleman over the grandfatherly (or great-grandfatherly) Mondale.

In a televised debate on Monday morning, Mondale kept referring to Coleman as "Norman," as if her were a little boy. Instead of pointing out how young an inexperienced the conservative is, maybe it had the reverse effect of showing just how old Mondale is.
posted by Hammerikaner at 11:50 PM on November 5, 2002

What you might find surprising is that Coleman was never elected to office as a Republican, but rather as a Democrat.

Actually the second time he ran, he did run as a Republican and was elected as one. He switched in 1996 over the issue of Welfare reform.

From MinnesotaPolitics.net:
Mayor Coleman served three years under the Democratic-Farmer-Labor Party's banner before switching to the Republican Party in 1996. He defeated State Senator Sandy Pappas during the general election in 1997 to win reelection. In winning reelection, he became the first Republican mayor of St. Paul since George Vavoulis in 1960. During the election of 1997, he received nearly sixty percent of the vote.

posted by Steve_at_Linnwood at 11:56 PM on November 5, 2002

Hah! Then my poly-sci professor lied to us all on Monday. Boy, I'm full of half-truths tonight.
posted by Hammerikaner at 11:59 PM on November 5, 2002

I've just been informed by an official here in Washington Co. that we won't be getting in any more results this evening.

The only precinct that won't be reporting in is Hastings, which I have been told has 4 registered voters. They do things a bit different down there, I guess, and will report their number(s) in the morning.

The Minnesota election didn't go the way I wanted, but it's been an exciting, informative, and educational night, and I'm happy to have shared it with you all.

Goodnight.
posted by Hammerikaner at 12:55 AM on November 6, 2002