« Analysis of the Big Four | Main | Tracking the Tracking Polls: The Sequel »

RVs or LVs?

I've made no secret of my preference for RV over LV data during the course of this campaign, for reasons that have been outlined in detail in many posts.

But what about now? If LV data make any sense at all, it's now, on the eve of the election, when polls are no longer, in effect, screening voters for participation in some mythical "snap" election substantially preceding the real election, but are getting close to screening voters for the real election itself. This was, after all, the original purpose of LV screens and models and perhaps these screens and models still deserve our trust this close to the election.

I'm not so sure. Just because these models can now be employed in a way closer to their original purpose doesn't mean they are now particularly trustworthy. The same problems that have vexed LV samples throughout this election seem likely to persist, at least at some level, right up to election day. I have, of course, reviewed these problems and the spotty track record of LV predictions in detail, and readers are urged to go back and consult those posts if they wish. And for extra credit, I strongly urge you to consult Mark Blumenthal's (of Mystery Pollster) epic seven part series on likely voter methods (soon to be major motion picture, I understand). You really should read the whole thing, but here's the Classic Comics version:

1. Likely voter methodology is extremely complicated.

2. Everybody does it differently.

3. They don't really know for sure their methods work--especially during the bulk of the campaign season--and there are all kinds of ways in which likely voter methods can produce biased samples and wild swings in results that are unrelated to voter sentiment.

4. It's not clear that the likely voter models have superior accuracy, even at election time, in today's political environment.

Which leads me to the following query: why bother? If these likely voter methods don't work that well, are prone to fairly serious bias and volatility and everybody has a different approach so you can't easily compare different polls, why go to all the trouble of drawing these samples anyway? Maybe good old-fashioned, low-hassle RV samples are just as good--in fact, better.

That said, we're kind of stuck with LV samples in most polls today, so in many cases it's either LV data or no data. As a mitigating factor, many of these polls that just report LVs have comparatively simple and modest screens--a question or two--that produce samples not too far off from a straight RV sample. But for the complicated LV approaches that use an elaborate series of questions and scores to ferret out the allegedly likely voters (think Gallup, but others like WP/ABC use similar systems), I'd still be inclined to take their RV data (assuming they're still reporting it) a bit more seriously than their LV data. Or, if you want a compromise, average the two.

OK. Class dismissed. Back to obsessively following the latest polls...

Comments

Without their own proprietary likely voter model, how would a polling company distinguish itself from the competition?

I would still like you to take on the task of talking about the "unlikely voter."
http://www.corante.com/mooreslore/archives/026771.html

IMHO the election won't be decided by undecideds "breaking" one way or another. I don't think they do break one way or another. I think most stay home.

I believe elections are won or lost by whether you can get a large group of "unlikely" voters to the polls, folks who don't usually go.

No pollster I've heard of has done any work to my knowledge in trying to figure out who these "unlikely" voters are, what they think, and what is the likelihood they will actually vote.

You could do it like this:

50% unlikely (don't vote usually but say they will this time, or have) B -- % K - -- %
25% likely (don't vote usually but say they might)
25% likely (usually vote but say they might not)
10% likely ( usually don't and say they probably won't, but might)

I was wondering if anyone had any thoughts or information about the prospect of Bush winning the popular vote because he is slightly ahead in the tracking polls and losing the electoral college because his national numbers are a reflection of increased support in places like Mississippi, Texas Alabama, etc. (the bigoted states of America) but hopefully decreased support in the battleground states.

Sooooo. An RV poll is an opinon poll of registered voters and a LV poll is an opinion about an opinion poll of registered voters. Can I just sleep till November 2?

An analysis of the battleground polls from 2000, shows that none of the top 5 predicted battlegrounds finished in the top 5 closest contested states.

http://www.campaigndesk.org/archives/001058.asp

Another point to consider when doing today's battleground analysis.

In future postings, especially as the election approaches, can you include ALWAYS the previous poll the corresponds with the "current" finding for say "Florida LVs" so we'll know whether the current results represent a gain or a loss?

This Osama tape and its timing are very interesting, and could determine the outcome of the election. Obviously, the very fact that they are watching the elections and have something to say on the eve of the elections will likely (whether it "SHOULD" or not is a different question) help Bush. People thinking about Bin Laden and conditioned to the reflexive response of saying that Bush is good in the 'war on terror' will tend to be more likely to vote Bush. This is the first video that Bin Laden has released in 2 years, and it comes less than one week before the election and focuses on Bush. Even though there was apparently a hoax report of threats of retaliatory attacks if Bush wins, ANY tape from OBL focusing on Bush predictably helps Bush. Bin Laden obviously DOES have a clue -- the 9-11 attacks reflected a sophisticated understanding of the details of the US system. He could not but know that his tape may (decisively) help Bush. Which should make voters think -- WHY does Bin Laden do something likely to help, possibly decisively, Bush's election at the polls. The unsavoriness of the implications go beyond what most "reasonable" people are willling to consider -- but it DOESN'T require any kind of elaborate conspiracy thinking either. Al Qaeda clearly sees the predictable results of their tape as being to their advantage. Focusing on Iraq rather than Al Qaeda has been a boon to Al Qaeda in TWO ways. Now, Al Qaeda is acting to (in all likelihood) insure a continuation of this status quo, with the common pattern of reactionaries and terrorists benefitting from one another (as in Israel). In Israel, it was pointed out to terrorists that their attacks during an election season would tend to elect Netanyahu and they said they were AOK with that. But here OBL has GONE OUT OF HIS WAY to issue a statement that could only help Bush. Then there's Krauthammer claiming that Bin Laden wants a Kerry victory, as the earlier Drudge report suggested. This is VERY BIG in the election folks. The milking of 9-11 goes on. It is "Christmas for Tories" and Santa has returned with more goodies.