http://www.geocities.com/electionmodel/FalseRecallRebut... So which is it? Slow-drifting fog? Mere forgetfulness? Retrospective bandwagon? The answer is:
NONE OF THE ABOVE! THE NES RESPONDERS TOLD THE TRUTH ABOUT THEIR VOTE.
False Recall is based on the NES 600-sample retrospective vote survey. It was the final naysayer “Hail Mary” pass to explain the impossible 43/37 returning Bush/Gore voter split in the Final 2004 National Exit Poll. The claim was that the 600-700 polled produced a result that did not match the RECORDED vote share; the survey "overstated" support for the incumbent.
Here's why the naysayer interpretation of the survey was bogus. There was no false recall.
First, it was based on the recorded vote - not the True vote. There are millions of uncounted votes in any election. Any analysis which does not consider total votes CAST is invalid by definition. That is the fundamental FLAW in all of the naysayer "theories" used to denigrate UNADJUSTED exit polls.
Second, the individual survey MARGIN OF ERROR was 4%; the deviations can vary greatly from one election to the next. As in all statistical analysis, one needs to look at the average deviation to "smooth out" the variations. It's the Law of Large Numbers. A good example of LLN is that a baseball players batting average will fluctuate greatly early in the season but converge to his TRUE average at the end as his total at bats increase.
This analysis will show that the retrospective votes were well within the Recorded and True Vote margin of error. The respondents told the truth. That should end the discussion right there. By not considering TOTAL VOTES CAST, "false recall" is based on the FALSE PREMISE that the recorded vote is identical to the TRUE VOTE. In the 11 elections since 1968, the average Democratic True Vote share exceeded the recorded vote by 3.7%; the retrospective Democratic margin was higher than the recorded margin in every election.
In 8 of the 11 elections, the Democratic True Vote share fell within the NES margin of error.
The average NES winning margin was 11.4% (11.4% for the Democrats and 11.5% for the Republicans).
The average Democratic True Vote winning margin was 10.0%
The average Republican True Vote winning margin was 12.4%
NES vs. True Vote (1968-2008)
The average absolute discrepancy for the 11 elections was -0.40%.
The average Democratic absolute discrepancy was -0.70%.
The average Republican discrepancy was 0.46%.
NES vs. Recorded Vote (1968-2008)
The average absolute vote share discrepancy was -1.75%.
The average Democratic absolute discrepancy was -3.30%.
The average Republican discrepancy was -0.46%