Hart's televised focus groups are always interesting and may provide political campaigns with helpful clues about how many voters are viewing issues. But in the end, every focus group leaves me shaking my head at how many voters "go on gut" and inform their views with very little political philosophy or practical knowledge.

No wonder a President's approval ratings can race up and spin down over the course of a few weeks even when the President has been on the same course throughout that period.

Often pollsters group voters by political philosophy: very liberal, liberal, moderate, conservative, very conservative. I don't find this useful because voters who call themselves liberal or conservative are unlikely to place the same meaning on these terms.

For purposes of predicting outcomes of elections or measuring the political pulse of the country -- I find the following divisions more useful:

Yellow Dogs (65%). These are the folks who will vote for one party or the other "all of the time." The term comes from an old southern saying: I'd vote Democrat if they put up a yeller dog!" In this polarized environment when Republicans are in a down mode, the yellow dogs probably make up about 70% of the electorate. About 35% of the electorate will approve of and vote for Obama's re-election no matter how successful or unsuccessful he is. About 30% will vote Republican even if the Party nominated Rush Limbaugh.

Blue Dogs (20%). This is a newly popular term that has come to represent voters and their representatives that "could go either way" based on mixed liberal and conservative views. Many of them are traditional values Americans who also respond to populist messages. Others are truly independent or moderate voters who pick and choose from a menu of issues and vote a split ticket.

The Rest (15%). Although many of the yellow and blue dogs are not well versed in politics or government, they are at least influenced by political attitudes, opinions, loyalties, or information. Apparently, the same cannot be said for the rest of America -- those voters who are for health reform one day and against it the next. Those who think Sarah Palin and Barack Obama are both terrific in October, and have turned on one or the other by November. These voters account for the irrational swings in voter opinion and these voters are targets of 90% of political advertising and rhetoric

I am afraid there will always be a large contingent of voters whose two favorite politicians are Ted Kennedy and Mit Romney. These are people who become afraid when an announcer says in an ad: if you vote for candidate X you are voting for corruption in Washnigton -- even if the candidate is clean as a whistle. These are mainly people who hate politics but vote anyway.

But mostly, these are the people who just don't read newspapers, or discuss societal issues -- unless they are in one of Peter Hart's focus groups. And it shows.

Sadly, John, you are right about "the rest." Around these parts they're called "low-information voters." Interesting question is, with such low information, and exemplified by that, a implied disinterest in gathering more, why do they bother to vote? Out of a sense of patriotic duty to march into that booth? Because they'd got nothing better to do on a Tuesday afternoon? So they can feel part of a national or local conversation that they're not following? Or, of course, being so low on information they can't realize they're low on information? Whatever it is, contrary to the remainder of voters, "the rest" should be discouraged from voting by all reasonable voters, regardless of ideology. Many theorize that these voters don't matter because, being so low on information, they either break 50-50 in a close election, or, blowing with the wind, add to the margin of victory in a decisive win or landslide. But really, after 2000 in Florida, who wants to take their chances with these morons?

As for the "yellow dogs," you miss an important distinction within their ranks. Between what might be called the "delusional" (term coined by Lewis Black) and the "defaults." Example: a delusional Republican would vote for any candidate (Limbaugh, Billy Graham, even Sarah Palin) with an R on their chest. A default Republican would not. Defaults prefer one party over another for philosoophical reasons but are not brainwashed ideologues about it. As a liberal I'm a default Democrat, but if the Republicans ever ran somebody more liberal than the Democrats for president (unlikely as being struck by lightning but could happen), I'd vote for that candidate. I'd happily vote for Obama for President again, but I'd never support Joe Leiberman. Or that cartoon dog (Huck Fin?) he resembles if they ran him.

Since Bush's approval rating bottomed out at about 30%, we can safely assumed that about 30% of the population is delusional Republicans. That no Dem could win the presidency by a margin greater than 70 to 30 even if the Republican president started a nuclear holocaust.

On the other hand, I can't imagine that, being on average more educated, the number of delusional Democrats is that high. I'm sure that if their party screwed up as badly as the Bush administration has, they would jump ship in higher numbers. Look, for example at the margin of victory for Reagan over Carter (the last complete failure as POTUS for Ds) - think it was 60% to 40%? - vs. the margin of victory for Obama over McCain (standing in for Bush, arguable the worst president ever), which was only 53% to 47%. Yes there were complicating factors (like McCain not being the incumbent and Obama's race) but that margin proves the obvious: Republicans have WAY more nutjobs in their party.

CutterMcCool;

From one yellow dog democrat to another, I thank you for your response and wish to reply. I like the term "low information" voters, but am certain that they will never be discouraged to participate. This country has an unbending belief in participation in elections even when a voter has no perception.

More to the point, neither Democrats nor Republicans really want them to go away because they are up for grabs. By watching any political season of advertisements it is clear that the vast percentage of campaign dollars are aimed at people without a clue. The only limit to this kind of cynical campaigning is the risk of losing independents (blue dogs) by going to far. For example, if Obama's re-election campaign faces a well financed effort by the "birthers" he wins hands down.

I would also disagree that the approval ratings of Presidents reflect the percentage of yellow dogs who would vote for that president's re-election. Many yellow dogs will express disapproval of a president but when faced with the opposite party's option will still vote their traditional party. There are occasions, however, when yellow dogs migrate to the independent column when their views change or when the party changes underneath them. This hasn't happened in decades. It may be happening in a generational sense in favor of democrats over the next decade.

I would also venture that under no current circumstances would the vote ever split 70-30 either way. Obama's approval ratings may return to his high approval ratings or higher, but he will not win all of those votes in a head to head. The only way Obama hits even 60% is if the Republicans go completely wacky and nominate a Palin or a Gingrich. Romney, Huckabee, or Pawlenty, on the other hand would easily win 45% of the vote even against a high riding Obama.

The three biggest presidential margins in the last half century have, as you implied, averaged 60-40. (Nixon-McGovern, 61-38; Johnson-Goldwater 61-38; and Reagan-Mondale 59-41.) The important point is that the last time there was such a blowout was 25 years ago, and the reason for this is that the electorate changed and polarization kicked in.

The country will remain very split for the time being, with those low information voters calling the shots. Fortunately, Obama is probably better than anyone at appealing to low information voters.

He's a celebrity.

John