After a Washington Post poll that gave the "wrong" answer on NSA collection of phone data, the frantic media rushed out two more polls in less than a week showing the "right" answer, i.e. public disapproval. The first one was from Newsweek, and the latest is from USAToday.

Problem with rush jobs is that the reporter looking for the information he wants will tend to skip around a bit.

Look at the first question in the USA Today poll itself.

1. Do you think the Bush administration has gone too far, has been about right, or has not gone far enough in restricting people’s civil liberties in order to fight terrorism?

Too far: 41
About right: 34
Not far enough: 19
No opinion: 6

In other words, the headline could just have accurately read "53 percent of respondents thinks Bush about right or not going far enough in restricting civil liberties to fight terrorism." Or based on question 7: "64 percent of respondents not concerned about government having their phone records."

You might have noticed that the results from one question often conflict with the results from another. Look at 7 and 8 for example.

Question 10 is loaded for bear, putting out complete paranoid speculations for which no evidence has ever been presented.

And finally:

According to question 2, a third of respondents weren't even following this story, and less than a third were following it closely. So, what does that indicate about question 3? Or for that matter, all the rest of the questions?