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

To: All

One more time. About polls:

They come in 2 flavors. Constant party mix. Variable party mix.

Rasmussen has a constant party mix. His telephone robots are looking for XX Dems, YY GOP and ZZ Independents. They keep calling until they get that mix. When they do, the results are what they are. When you see movement in Rasmussen, it’s a real trend. People are indeed changing their minds.

Most other pollsters call randomly and ask the respondent what their party affiliation is. If they get 70% Dems, 20% GOP and 10% Independent, too bad. That’s what they report. The next week it might be 20% Dem, 70% GOP and 10% Independent. Randomness can do that. It’s much harder to spot trends in this style poll.

The problem with the constant mix variety is that the target mix may not be right (usually it’s whoever showed up last election). Party affiliation is a VERY powerful determinant of candidate preference. So the mix picked will determine the poll’s results. The numbers reported may be wrong, but a trend is REAL.

The problem with variable mix is that poll to poll you may see huge variation of party mix and this cannot reflect national opinion — because people just don’t change their party affiliation that frequently.


198 posted on 08/20/2008 8:28:11 AM PDT by Owen
[ Post Reply | Private Reply | To 195 | View Replies ]


To: Owen
The problem with variable mix is that poll to poll you may see huge variation of party mix and this cannot reflect national opinion — because people just don’t change their party affiliation that frequently.

Must be an election year, DNC Owen is back to tell us why black is white and down is up.

Boy are you confused Owen. You got it exactly backwards. What you said above is what critics say about Rassmussen's Party Mix methodology. He polls from time to time to find out what the party mix is then uses that in all his polls. So his party mix ratios change based on who answer the phones which he then uses in all his other polls. You criticism is valid, if you level it AT Rasmussen instead of in defense of him

As usual, your DNC Masters talking point is exactly backwards to reality.

201 posted on 08/20/2008 8:33:14 AM PDT by MNJohnnie (http://www.iraqvetsforcongress.com ---- Get involved, make a difference.)
[ Post Reply | Private Reply | To 198 | View Replies ]

To: All

One more thing. Re: Hillary.

She would be a formidable VP nominee.

But more to the point, her planning on 2012 is not a great plan. Losing the nomination this year was a DISASTER for her. 2012 will have her either facing a Democrat president running for re-election, or if McCain runs again, an incumbent opponent perhaps presiding over a recovering economy (the cycle being on an uptrend by then).

This was her year. The future is far less favorable.

And yes, there’s no question that she can not have it this year. She would lose all the black votes. They would never tolerate a backstab to Obama (unless he is taken “ill”).


202 posted on 08/20/2008 8:33:21 AM PDT by Owen
[ Post Reply | Private Reply | To 198 | View Replies ]

To: Owen

The only pollster that does anything close to that is Gallup in some of their polls. Virtually everyone else adjusts the numbers for several demographic variables including party affiliation.


232 posted on 08/20/2008 5:43:54 PM PDT by comebacknewt
[ Post Reply | Private Reply | To 198 | 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