I found the internals of that newsweak poll.
http://biz.yahoo.com/prnews/061007/nysa007.html?.v=73
Newsweek Poll
Congressional Republicans and Election '06
Princeton Survey Research Associates International
Final Topline Results
(10/7/06)
N = 1,004 national adults, 18 and over
Margin of error: plus or minus 4
Interviewing dates: 10/5-6/06
SAMPLE SIZE/MARGIN OF ERROR FOR KEY SUBGROUPS:
483 Men (plus or minus 6)
581 Women (plus or minus 5)
841 Whites (plus or minus 4)
145 Non-whites (plus or minus 9)
296 Republicans (plus or minus 7)
332 Democrats (plus or minus 6)
338 Independents (plus or minus 7)
SAMPLE SIZE/MARGIN OF ERROR FOR REGISTERED VOTERS SUBGROUPS:
899 Registered voters (plus or minus 4)
624 Likely voters (plus or minus 5) (definition noted below)
278 Republicans (plus or minus 7)
302 Democrats (plus or minus 7)
291 Independents (plus or minus 7)
NOTES:
Data is weighted so that sample demographics match Census Current Population Survey parameters for gender, age, education, race, region, and population density.
Reported sample sizes are unweighted and should not be used to compute percentages.
Likely voters are determined based on response to a series of questions about voting intentions, voting history, general interest in politics, interest in the 2006 election, and knowledge of the voting process. Likely voters are the 62% of registered voters with the highest score on an index summarizing response to the likely voter questions.
D2. Now I have just a few more questions so we can describe the people who
took part in our survey ... Regardless of how you might have voted in
recent elections, in politics TODAY, do you consider yourself a
Republican, Democrat, or Independent?
Total RVs LVs
25 Republican 27 29
36 Democrat 38 39
35 Independent 32 29
1 (VOL.) No party/Not interested in politics 1 1
* (VOL.) Other party * *
3 Don't know 2 2
100 100 100
Well .. their numbers do not match the census - because the dems' numbers are not that much higher than the repubs. Even 6 or 7 percent is really high. Actually, the census numbers put us about even since the 2000 election when the conservative voters went out and registered repub. We are almost evenly divided at this point.
And .. the most accurate polls are those done with "likely voters" .. not "registered voters" (because they don't necessarily plan to vote) .. and using just "adults" totally distorts the real numbers.
Again, I've seen the pollsters (especially those who are very friendly to the liberals) totally skew/slant their polls in the dems favor.