Link from MoveOn.com explaining difference between RV/LV

[yellerpup]

I thought this was a comprehensive explanation. Sorry if it is a dupe.

<http://www.americanprogress.org/site/pp.asp?c=biJRJ8OVF&b=183679>

[tritsofme]

1. Here's more context to the Gallup memo he refers to throughout the article

One reason not to use likely voters early in the campaign is cost. Registered voters represent about 85% of a national sample of adults, while likely voters represent only about 50% to 60% of a national sample. To identify the subset of likely voters, pollsters need to include extra screening questions and conduct many more interviews to find enough people who are deemed likely to vote.

Apart from cost, some pollsters (and presumably those whom Cook referred to) object to using an LV model early on in the campaign because they think the more stringent subset of respondents may not include all voters. One measure that many polling organizations use to identify likely voters is a respondent's level of interest in the election. The problem is that for many people, interest may not be stimulated until late in the campaign, and in the meantime, their low interest could cause them to be classified as non-voters. Some observers argue that excluding large numbers of people early in the year because of lack of interest may give a false reading of voter preferences.

Gallup tries to avoid the lack-of-interest problem by specifying that 55% of its sample (which corresponds with a 55% projected turnout in the election) will be considered likely voters. Gallup asks each respondent seven LV screening questions, and gives each person an LV score of 0 to 7. The top 55% are classified as likely voters. This classification works even if interest is low early in the year, because low-interest respondents are considered likely voters as long as they are in the top 55% of the sample on the overall LV scale.

Bottom Line

While some critics may not like the likely voter model, there is no empirical basis for rejecting it in favor of all registered voters. The only empirical test of the voter models comes at the end of the campaign, when the last pre-election poll is compared with the actual vote. At that time, most -- if not all -- of the major media pollsters will opt for likely voters, because of their desire to project as accurate a forecast of the election results as possible. We simply do not know, nor can we know, which model is better during the campaign itself.

Personally, I favor the LV model, based on the logic that if it is the most accurate model just before the election, it is probably the most accurate during the campaign as well. But for the skeptics, Gallup presents the results based on registered voters as well. As the saying goes, "You pays your money and you takes your chances."

[ret5hd]

2. i really have a feeling that...

this particular election will blow the LV model out of the water. Not because it is historically faulty, but because of the circumstances of this election.