Rasmussen's Biased Voter Model

Just giving a quick glance to the Senate races I saw one where he well off and most where he was relatively close or even undersold what the Republican ultimately got. Is that whay you are referring to?
 
Taking a closer look, Rasmussen's voter model just plain sucked and in the close races he generally missed the spread in the Republican's favor by a few points.
 
NigelTufnel said:
Taking a closer look, Rasmussen's voter model just plain sucked and in the close races he generally missed the spread in the Republican's favor by a few points.
Click to expand...

Looking at the Senate races from a Republican perspective with plus being the amount he overpredicted:

AL -7%
AK +1%
AZ -7%
AR -3%
CA +4%
CO +1
CT +2%
DL =
FL +1%
GA +1%
Hawaii +18%
ID -8%
IL -2%
IN -3%
Iowa -10%
KS -9%
KY -3%
LA -3%
Maryland +2%
MS -2%
NV +4%
NH -4%
NY -2%
NY Special -3%
NC -3%
ND -4%
OH =
OK -3%
OR +2%
PA -1%
SC -4%
UT -1%
VM +1%
WA -2%
WV +3%
WI +1%
 
My math is probably off here but I counted 20 where he undercounted the Republican, 12 where he overcounted and 2 where he called it correctly. He was WAY over on Hawaii but his over large misses he undercounted the Republicans.

I have no idea how accurate pollsters usually are so its quite possible his model sucks but I'm not seeing some partisan bias in these results.
 
cawacko said:
My math is probably off here but I counted 20 where he undercounted the Republican, 12 where he overcounted and 2 where he called it correctly. He was WAY over on Hawaii but his over large misses he undercounted the Republicans.

I have no idea how accurate pollsters usually are so its quite possible his model sucks but I'm not seeing some partisan bias in these results.
Click to expand...


Well, if you look at the close races, the toss ups and leaners, the bias becomes readily apparent. In each of the seven races classified as tossups, Rasmussen erred in favor of the Republican. In two of the three races that Rasmussen classified as leaners, Rasmussen erred in favor of the Republican. So in 9 of 10 "close" races, Rasmussen got it wrong in the Republican's favor.
 
NigelTufnel said:
Well, if you look at the close races, the toss ups and leaners, the bias becomes readily apparent. In each of the seven races classified as tossups, Rasmussen erred in favor of the Republican. In two of the three races that Rasmussen classified as leaners, Rasmussen erred in favor of the Republican. So in 9 of 10 "close" races, Rasmussen got it wrong in the Republican's favor.
Click to expand...

Ok, which races were those?
 
NigelTufnel said:
Well, if you look at the close races, the toss ups and leaners, the bias becomes readily apparent. In each of the seven races classified as tossups, Rasmussen erred in favor of the Republican. In two of the three races that Rasmussen classified as leaners, Rasmussen erred in favor of the Republican. So in 9 of 10 "close" races, Rasmussen got it wrong in the Republican's favor.
Click to expand...

OOOoooooHHhhhhh....I get it.....

You have no point.
You're a pinhead
You're reaching for who knows what.
The voters made an ass of your entire party by repudiating socialism.

:fu:
 
http://www.huffingtonpost.com/alan-abramowitz/bad-night-for-congression_b_778126.html

Gallup's final likely voter poll gave Republicans a 15 point margin on the generic ballot, the largest of any major poll and far beyond the average of 6-7 points. The actual result, based on both the national exit poll and votes tabulated thus far, appears to be between 6 and 7 points.

Not only did Gallup miss the actual vote margin by a mile, but their projections about the composition of the midterm electorate were also way off the mark. Based on the exit poll results, it appears that the actual electorate was not nearly as male, old, Republican, or conservative as Gallup's final likely voter sample.

"the actual electorate was not nearly as male, old, Republican, or conservative "

So, does this mean that Gallop is biased ?
 
bravo said:
http://www.huffingtonpost.com/alan-abramowitz/bad-night-for-congression_b_778126.html

Gallup's final likely voter poll gave Republicans a 15 point margin on the generic ballot, the largest of any major poll and far beyond the average of 6-7 points. The actual result, based on both the national exit poll and votes tabulated thus far, appears to be between 6 and 7 points.

Not only did Gallup miss the actual vote margin by a mile, but their projections about the composition of the midterm electorate were also way off the mark. Based on the exit poll results, it appears that the actual electorate was not nearly as male, old, Republican, or conservative as Gallup's final likely voter sample.

"the actual electorate was not nearly as male, old, Republican, or conservative "

So, does this mean that Gallop is biased ?
Click to expand...


Yes, this means Gallup's likely voter model for its generic ballot question was heavily biased towards Republicans.
 
You must log in or register to reply here.
Back
Top