Posted 9:05 AM by Robert

Davis Descending XXXIV: He's Still, Still Not Dead, Yet

The Los Angeles Times says that the poll results described below indicating support for the recall at only 50% already assume a disproportionate Republican turnout. That adjustment to the raw polling data was made by the pollsters because Republicans seem more "eneregized" at the moment and more likely to go to the polls. That's a perfectly appropriate adjustment for a pollster to make. But is is also something a good campaign by the Governor can change - if he can "energize" his support by, say, demonizing his opponent. Interestingly, although Mr. Davis is widely reported to have alienated the liberal wing of his own party, the Times reports: Only the liberal wing of the Democratic party is overwhelmingly against the recall. Nearly nine out of ten of that group plan to vote to allow the Governor to finish his term.

Most strikingly: If this poll is accurate but there were not disproportionate Republican turnout, then the recall might not pass, because even with the assumption of disproportionate Republican turnout, only 50% of voters favor the recall. So relaxing that "disproportionate Republican turnout assumption or allowing for Davis' campaign changing even a relatively tiny number of minds, might well mean that he survives. Cautions are in order: The Times does not say how big an adjustment was made by the pollster on the basis of this assumption - although the Times rhetoric suggests it was a fairly big adjustment. And that 50% translates into slighly more than 50% of voters who say they have made up their minds, since 5% are still undecided.

It also is worth recalling that there are other measures on the ballot which could affect turnout - most notably Ward Connerly's measure that would ban collection of racial data by the state and therefore potentially affect affirmative action calculations. Oddly, the Times' poll and article don't explore any of that. Many people have suggested that minorities would flock to the polls to defeat this measure - although the last time I looked it was favored by 3-to-2.