First, on the controversy that has emerged about the new Los Angeles Times (LAT) poll. In essence, the criticism of the poll comes down to this: there are too many Democrats in the poll, which explains how Kerry can be leading Bush by 7, despite losing to Bush among independents and having a smaller margin among Democrats than Bush has among Republicans, and how the Democrats can be 19 points ahead in the generic congressional ballot.

And it is true, as LAT poll Susan Pinkus has admitted, that the current LAT poll has has an unusually large 13 point Democratic advantage on party ID (the sample is 38 percent Democrats, 25 percent Republicans and 24 percent independents). It could be true that there is some problem with the poll that led to oversampling of Democrats (though, since this is an RV, not LV poll, there is no obvious culprit for this problem). Or it could be plain old sampling error--the mean of this poll just happens to be unusually far away from the true population figure. Or it could be there is a surge toward the Democrats that is driving up the number of Democratic identifiers among voters and enhancing the Democrats' party ID advantage. (Or it could be a combination of the second and third explanations: there is a surge toward the Democrats and sampling error in the LAT poll produced an unusually high number of Democrats, even given that Democratic surge.)

Of course, no one can "prove" anything here. But I, for one, find these figures (the Democratic party ID advantage and generic congressional ballot advantage) generally plausible, if perhaps a bit on the high side. There are ample grounds for thinking there is, in fact, a surge toward the Democrats and their positions and away from the Republicans and their positions among the broad electorate. A growing Democratic party ID advantage is a logical consequence of that surge, since party ID does not remain stable as political conditions change.

Indeed, the Democratic party ID advantage has been growing ever since the post-9/11 surge in Republican party ID ended sometime in 2003, so the LAT 13 point Democratic advantage, while seemingly high, fits well with trend. Note also that it is not without recent precedent: a recent ABC News poll gave Democrats a 10 point party ID lead and a January CBS News/New York Times poll gave the Dems a 13 point lead (though note that this included leaners). The same is true of the Democrats' big advantage on the generic Congressional ballot; public polls have been showing that advantage growing steadily for quite awhile and a number of polls have shown the Democrats with double-digit leads (though, admittedly, none has shown a lead as high as LAT's).

Conclusion: there is no good reason to ignore the results of this poll (unless you're Matthew Dowd, of course, who has his own reasons for doing so). Like all polls, it should be taken with a grain of salt and considered in relationship to other polls. But there's no need, in my view, to be any more stringent that that.

So back to the data. I promised to discuss the Bush-Kerry comparisons on traits and issues, which were generated by giving respondents a series of statements of the form "he......." and asking them whether the statement applies more to Bush or Kerry.

It is interesting to compare Kerry's worst area ("he flip-flops on the issue") to Bush's worst area ("he is too ideological and stubborn"). By almost 2:1 (48-25), voters felt flip-flopping applies more to Kerry, but by well over 3:1 (58-16), voters felt being too ideological and stubborn applies more to Bush (57-14 among independents and 64-11 among moderates). Given the pragmatic, problem-solving orientation of American voters, that seems like an exceptionally poor position for Bush to be in.

It's also worth noting that Bush's numbers in his second-worst area ("he has better ideas for handling the problems of cost and access to health care") are also worse (51-24 against him; including 50-19 among independents and 63-11 among moderates) than Kerry's numbers in the flip-flopping area.

Other good areas for Kerry are "he has better ideas for strengthening the nation's economy" (48-37 in his favor, including 46-31 in Ohio) and "he cares about people like me" (47-35, including 45-31 in Ohio), while good areas for Bush are "he would be best at keeping the country safe from terrorism" (50-31 in his favor) and "he shares my moral values" (45-36).

Finally, note that Bush has no advantage at all on "he will be a strong leader for the country" (44-44) and only a one point advantage on "he has the honesty and integrity to serve as president" (41-40). For a president whose stock in trade used to be the strong leader who told it like it was, that's not very encouraging news.

But more encouraging than this: by 52-22, voters say the country is worse off, rather than better off, due to the economic policies pursued by Bush in the last three years. And it's an essentially identical 52-23 verdict on that question in Ohio.

Perhaps instead of wondering whether there's a surge toward the Democrats, we should be wondering why it took so long!