Site Meter

Monday, August 02, 2004

The Gallup anomaly returns

In 2000 when Gallup used their likely voter filter they tended to get results which are good for Bush. Polling report reports that 2004 is deja vu all over again



CNN/USA Today/Gallup Poll. July 30-31, 2004. N=763 likely voters nationwide. MoE ± 4.
Bush/Cheney Kerry/Edwards Neither(vol.) Other(vol.) NoOpinion

50% 47% 1% - 2%

Among ALL registered voters:
47 % 50% 1% - 2%.

Newsweek Poll conducted by Princeton Survey Research Associates International. July 29-30, 2004. N=1,010 registered voters nationwide. MoE ± 4 (for all registered voters).
Bush/Cheney Kerry/Edwards Nader/Camejo Other (vol.)/Undecided

42 % 49% 3% 6 %

Zogby America Poll. July 26-29, 2004. N=1,001 likely voters nationwide. MoE ± 3.2 (total sample).

Bush/Cheney Kerry/Edwards Other(vol.) Unsure

43% 48% 2% 8%



update: I left out the best part. a 6 point swing from Gallup likely voters to Gallup registered voters. How does Gallup decide who is a likely voter ? Have they ever explained ?

No comments: