Many believe it. None have shown it. Time to put up or shut up weekend polling tinhats.
Just kidding of course but this seems to me to be an urban legend, Freeper-style.
Glad to see R back up by one. Two would be better. We will break out the champagne when it gets to three.
InterceptPoint,
I agree that it has never been proven conclusively. That is why I always put “many believe” in my thoughts about weekend polling.
However, there have been instances over the summer (and I would need to pull some historical numbers) that the MON and TUE polls (which include SAT and SUN in them) have been lower than the polls for the rest of the week on Rasmussen.
Maybe somone needs to Excel sheet it and put this to rest once and for all
You just have to watch Rassmussen’s number over the past three months. His Sunday and Monday results almost always show a serious rise for Obama, and then if level back on Tues/Wed. No conspiracy, just basic observation skills.
Agreed 110%!
The so-called "weekend effect" may be there for some polls, but not for Rasmussen -- apparently because he adjusts his numbers to cancel out the relatively low weekend response rate from 'Pub supporters.
A year or two ago, I ran the Rasmussen numbers for a multimonth period to see if Ras's Monday reports (representing numbers from Friday, Saturday and Sunday night polling) showed a systematic pro-Obama bias as compared to the Friday reports (which use the numbers from Tuesday, Wednesday and Thursday nights).
Outcome:
There was a very small ANTI-Obama bias in the numbers, but it wasn't statistically significant.
(I don't recall the exact figures, but they were something like 16 Monday swings against Obama, vs. 15 swings in his favor.)