Lets go through the rough on these two polls then hit some analysis on why there are such grandure contradictions between them.
PPP
http://www.publicpolicypolling…
January 3-4: Margin of Error 3.7%
Caroline Kennedy (D) – 46%
Peter King (R) – 44%
Undecided – 10%
Caroline Kennedy’s Favorability Ratings:
Favorable – 44%
Unfavorable – 40%
Not Sure – 16%
Peter King’s Favorability Ratings:
Favorable – 34%
Unfavorable – 26%
Not Sure – 40%
Party Breakdown for the Kennedy/King Head to Head:
Kennedy:
Democrats – 65%
Republicans – 19%
Other – 35%
King:
Democrats – 25%
Republicans – 74%
Other – 50%
Rasmussen
http://www.rasmussenreports.co…
Caroline Kennedy (D) – 51%
Peter King (R) – 33%
Third Party – 9%
Not Sure – 7%
Why the staunch contrast in the head to head matchup? Favorability Ratings.
Caroline Kennedy’s Favorability Ratings:
Favorable – 63%
Unfavorable – 31%
Not Sure – 6%
Peter King’s Favorability Ratings:
Favorable – 39%
Unfavorable – 27%
Not Sure – 34%
Party Breakdown for the Kennedy/King Head to Head:
In the match-up with King, Kennedy gets support from 74% of Democrats and holds a 12-point lead among unaffiliated voters. King is supported by 73% of Republicans.
Now, between these two polls I’m leaning toward Rasmussen. Why? She split Republicans and Democrats more evenly in that poll, which is realistic because most elections split 90-10 or 85-15 in partisan support. The wild card is the other category. There is a 27 point difference between PPP and Rasmussen and how Kennedy and King split the other vote.
King’s approval ratings and support among Republicans is consistent and crosses over between the two polls. I just don’t see Kennedy’s approvals being closer to 44/40 than 63/31.
I have to go to sleep, I may add more tomorrow after work. Please share thoughts on why there is such a sharp contrast between polls in certain areas and similar findings in others, or which poll is more accurate and why.