Well: "For this NEWSWEEK poll, Princeton Survey Research Associates interviewed 1,002 adults aged 18 and older on Jan. 16 and 17. The margin of error is plus or minus 3 percentage points"
Since some of you seem to know that this poll was scientifically valid:
from what area were the people asked? What gender, profession, age etc did they have? How many people refused to answer? Was the poll done in person at home, by telephone. or by mail?
Walter, I can't speak to the specifics of the particular poll in question, but typical "Scientifically Valid Polling Protocol" is highly structured, commonly involves a computerized random selection of a sample structured to encomapas statistically Nationally-representative demographics, results are compiled only from completed responses, and telephone or mail are employed in greater frequency than are "Face-toFace" interviews, though I infer this poll was conducted by telephone. In short, "Non-Respondents" are not factors and painstaking care is given to the matter of statistically valid sampling. It is also not infrequent that the respondents chosen to establish the poll results are further winnowed from a much larger universe of completing respondents prior to calculating the talley.
Of course, "There are lies, damned lies, and then there are Statistics".
timber
