Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: comebacknewt

Actually Newsweak does weight their sample for party affiliation and "adjusts" the numbers to make sure they conform to "census numbers" and "voter turnout model" which most likely is from 2000 and out of date.


75 posted on 10/02/2004 4:51:49 PM PDT by COEXERJ145
[ Post Reply | Private Reply | To 70 | View Replies ]


To: COEXERJ145
If adjust the numbers for equal dems and repubs and assume all repubs vote for Bush and all dems vote for Kerry:

Then it comes out even: 46 for bush, 46 for kerry:

If you used the same ratio as the previous poll:
Then Bush is at 50% and Kerry is at 42%:
SO THE CHANGE IS SIMPLE IN THE SAMPLE SELECTION!!!!
77 posted on 10/02/2004 4:59:21 PM PDT by nowheretohide
[ Post Reply | Private Reply | To 75 | View Replies ]

To: COEXERJ145

well then newsweek neds a better scale for weighting, if they go by voter turn out models and year 200 census and self identification history, when they arent even consistant on that, I cant say they are wrong in this poll, some people polled may have changed there self identification but it is an awful huge swing of 11 points and party affiliation see this latest poll. I say Bush is up by 4-5%.


82 posted on 10/02/2004 5:11:37 PM PDT by aft_lizard (I actually voted for John Kerry before I voted against him)
[ Post Reply | Private Reply | To 75 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson