Pookie
Chop!! Chop!!
- Joined
- Aug 25, 2002
- Posts
- 58,778
I'm glad you finally seem to realize I was right about the USC/LA Times poll being a likely voter poll, and also that it's impossible, at this point, to evaluate it for accuracy, as I previously explained.
I never said it wasn't. Here is my earlier post (emphasis added) ...
I'm not so sure that's accurate:
http://www.latimes.com/politics/la-na-pol-usc-daybreak-poll-methodology-20160714-snap-story.html
Each day's poll respondents are a subset of the UAS election panel, roughly 3000 U.S. citizens who were randomly recruited from among all households in the United States. Respondents are asked three predictive questions: What is the percent chance that... (1) you will vote in the presidential election? (2) you will vote for Clinton, Trump, or someone else? and (3) Clinton, Trump or someone else will win?
It appears they try to discern who likely voters are from a group of "3000 U.S. citizens" by asking the percent chance they will vote from the subset of 400 they call each day.
Asking the percent chance someone will vote is a rather novel approach for determining likely voters. Typically, polls who's methodology has been validated ask a series of nominal and/or interval questions to develop a profile in order to make that "likely voter" determination instead. The latter method would tend to weed out someone giving a false percentage to avoid telling the surveyor something that might seem embarrassing. That plays into the whole validity issue of this poll. I suspect it's part of the reason Nate Silver's adjustments results in such a large change in the Clinton lead, and more closely mirrors other polls during roughly the same time period.