USELESS POLLING – AT 7:45 P.M. ET:  The New York Times reports the latest Times/CBS News poll, but winds up looking amateurish.

The poll, we are informed, shows that President Obama still commands greater support than the Republicans.  Oh, there are vulnerabilities, the Times concedes, but the Democrats are fighting back.

Then you get to stuff like this:

The boisterous Tea Party movement that has grown out of the strain of discontent so far commands relatively little public support; 18 percent of respondents said they considered themselves supporters of the movement, while 55 percent said they had heard little or nothing about it.

I love the term "boisterous," don't you?  As opposed to what?  The scholarly Democrats? 

Look, when you have a poll that says 55% have heard little or nothing about the Tea Party movement, you have a defective poll.  That movement has been all over the tube for months. 

So what is the problem here?  This is the problem, but you have to wait until the last line of the story to be told it:

The nationwide telephone poll of 1,084 adults was taken from Feb. 5 to Feb. 10, and has a margin of sampling error of plus or minus three percentage points for all adults.

Any poll taken among all adults – not even registered voters – is going to skew Democratic.  Most of those who don't bother even to register are found in sub-groups that tilt toward the Democratic Party.  The best polls, like Rasmussen, are taken among likely voters, the best possible sample. 

I have no idea why The Times and CBS News would poll only among "adults."  Maybe the cost of refining the sample is a factor.  But I wouldn't take this poll too seriously.

February 11, 2010