Two very good pollster posts, one by Mark Blumenthal (pollster.com) on the likely voter and one by Frank Newport (Gallup) on the early voter, help bring some needed clarity but don't answer all questions.
By now, most of you are used to seeing the (RV) registered voter designation, a smaller universe than "all adults". "All adults" is useful for looking at policy questions, but if you are not registered, you can't vote, and if you can't vote, you are not relevant for election questions.
Of course, just because you can vote doesn't mean you will vote (we don't all live in Minnesota.) So an even smaller number is the (LV) or likely voter. These are people who are likely to show up at the polls, early or late... but if they have already voted, they aren't just a "likely" voter, they're an "early voter". And early voters tend to skew older and retired, and tend to exist more in the west than the east.
Early voters are partisans, and do not vote the same as the general population. How do they vote? Usually we know their party affiliation, but not who they voted for. That's important in NV and AK, where voters dislike candidates, and may not vote their parties.
Throughout the campaign season, then, you'll see (RV) or (LV), sometimes without explanation, next to poll posts.
Mark Blumenthal notes that the screening questions to identify LVs vary between pollsters, but one big discrepancy is the difference between a quesion about intent (smaller GOP advantage) and interest (larger GOP advantage). Using Pew, compare 2002, 2006 and 2010, with the blue line being "attentive":
corrected graph as per mark blumenthal
In all three elections, the "definitely vote" (green line) favored the GOP by 5 points as it does this year, but the "always vote" (red line) folks are more GOP this year. And the "paying attention" blue line is almost off the charts GOP this year.
Which is the "real indicator"? They all are, but the pollster that weight the blue data more heavily will look more GOP. Still, all the data suggest a GOP year in some proportion. Blumenthal:
We will know in a few days whether the narrowing that was evident in 2006 in the Gallup, Pew Research and other surveys repeats this year. It may not. And again, even if it does, Democrats are in for a drubbing that is likely to equal or exceed their loses in 1994. But at some point, pollsters need to address why their results sometimes diverge widely a few weeks or months before the election, only to converge at the end.
Gallup is starting to report their own data from surveys for actual voters and concludes:
Few Political Clues
While interested observers have been poring over reports of early voting in an attempt to get a handle on the direction of the election, Gallup's current data do not show much of a difference in early voting by party affiliation. Thirteen percent of self-identified Republican registered voters say they have already voted, compared with 9% of independents and Democrats. The percentages of those in each party group who say they will vote between now and Election Day are roughly equal.
However, without knowing how they vote, the data so far skews older.
The finding that older voters have a higher propensity to vote early is not a new one, but confirms that many senior citizens, like residents in the West, are by this point in the election cycle essentially "out of the game" as far as the campaigning is concerned. A disproportionately high number of younger registered voters volunteer that they will not end up voting this year, also confirming what is well-known in American politics -- that young voters are as a rule not highly involved in the election process.
Although that could change by election day, betting on the "unlikely voter" generally loses.
We did a media round-up of headlines on early voting yesterday, and the bottom line is still "we don't know". But Jon Ralston is the best tea leaf reader in NV; go here for the latest there (Republicans had a good day Monday.)
And remember that partisans tend to vote early, but independents tend to decide elections.
So, GOTV, and especially do so back east, where we sleep in on elections more often than not, and we vote late if we vote at all.