http://www.msnbc.msn.com/id/12771821/site/newsweek/

According to a Newsweek survey, 53% of Americans believe the NSA program is an extreme invasion of people's privacy. The Newsweek survey puts to rest claims by an earlier poll, by ABC/Washington post, that said Americans supported the NSA program by a 2-1 margin. The Newsweek survey had 1007 participants, compared to the 502 by ABC.

I figured once the American public heard the informed criticism, they would become generally opposed to the NSA program. Whether the program is illegal or not is another question, but its definitely another public image screw-up by the Republicans.

__________________
"The person who has nothing for which he is willing to fight, nothing which is more important than his own personal safety, is a miserable creature and has no chance of being free unless made and kept so by the exertions of better men than himself. " — John Stuart Mill

LOL

Sure thing.

That ABC poll only used 500 people? And that is supposed to be statistically significant? I've taken statistics class--its not. You would need at least 10,000 people to even come close to a 5% margin of error at the very least.

Well, I doubt the media care about statistical significance, its not like they are trying to reject a null hypothesis.

__________________
"The person who has nothing for which he is willing to fight, nothing which is more important than his own personal safety, is a miserable creature and has no chance of being free unless made and kept so by the exertions of better men than himself. " — John Stuart Mill

So.......how many Democrats/liberals were 'surveyed' by MSNBC? How was it conducted? Was it phone calls? Was it internet? When were the phone calls made? It doesn't say.

__________________
"This is a time for a national imperative not to fail in Iraq." Condoleeza Rice, January 11, 2007

The two statements are not mutually exclusive. Americans might both think it's an invasion of privacy and support it. That sounds sick to me, but it might be true. People might tbe willing to give up freedom in America.

But good call on the stats, Light.

__________________
"Man lives in the sunlit world of that which he believes to be reality. But unseen by most is an underworld, a place that is just as real... but not as brightly lit... A DARK SIDE!"
-opening from Tales From the Darkside

Heh, I'm amazed at how much this nation has succumb to apathy... Our society is plagued by TV shows and video games, it keeps us busy, we focus on our own insignificant little lives seeking pleasure and instant gratification when the world around us changes into a global despotism. And wow, you think something as important as the NSA prying open our privacy would be of national concern... I'm impressed that a whole '1007' people participated in that poll (sarcasm)

Reread you stats book or get a better one. I design surveys for a living and have a doctoral level stats background. The power of a test (that is the capacity of a test such as a poll to capture specific reality) and the standard error will vary with a variety of factors including the homogenity of the population, the strength of the effect being studied and specific instrument used, but both the power of a test and the standard error decreases very little once your reach about 500 cases and almost none after your reach 1000. Virtually no national polls every have more than a 1,000, most have signficantly less.

__________________
Paranoia strikes deep
Into your life it will creep
It starts when you're always afraid
You step out of line, the man come and take you away

The point is, that your "sample" gets skewed depending on who you interview. For instance, have you noticed the difference between the Prez's popularity ratings in the Fox (Opinion Dynamics) polls, versus the Gallup polls? The reason is, because the OD interviews primarily Republicans, so the sample base is inappropriate to begin with. Your "sample error" doesn't take that into account, at least the way it's usually calculated. It would be very difficult to construct an "accurate" political poll in this context, especially in an age where significant numbers in each party are crossing over to the other(s). If you kind of lump all the polls together in your mind, the most accurate thing you can probably say is that "about 30%, give or take maybe 5%", approve of the Prez, and "over half" disapprove. I wouldn't try to make it any more accurate than that, would you?

You have a point. Whatever difference there is in sampling error, there's no way it makes a sizable difference in 30% vs. half. Any way you go with the error, it's still way less than half that approve.

__________________
"Man lives in the sunlit world of that which he believes to be reality. But unseen by most is an underworld, a place that is just as real... but not as brightly lit... A DARK SIDE!"
-opening from Tales From the Darkside