We’ve added new polls today in 14 states. Interestingly, this latest batch is quite strong for Clinton relative to Obama in a number of states, including several of the key swing states. We’re not poll pundits, so will leave the analysis for others, but it is a bit of a paradox that she is polling well as the race winds down with Obama looking like the nominee.
The Obama vs. McCain race seems much more wide open, with 178 electoral votes in states where the polling margin is 5% or less (vs. 112 in Clinton vs. McCain). Since we’re running an electoral college website, we kind of like the idea of a race with an electoral map that gets shaken up a bit — there were only 3 states that differed in 2004 vs. 2000.
{ 3 comments… read them below or add one }
I’ve been visiting this site a lot lately. i like it, however the suggestion i would make is having wither a polls of polls or 3 different polls. evertime i see either the right leaning rasmussen or one of the left ones i take it with a grain of salt.
Thanks for your note. We’re not a huge fan of the ‘poll of polls’ concept. While it seems to make people feel more comfortable, it is statistically suspect to take separate polls and average them. At the same time, there really shouldn’t be much difference in the polls if they are all conducted properly. As for Rasmussen, we haven’t seen much of a bias (although we’ve heard this from others). The biggest issue we see is that they only survey 500 people for each poll — a sample size small enough to yield a 4-5% margin of error. All else equal, we try and use polls where a larger population is surveyed — Quinnipiac is particularly good in this area, usually including about 1,500 respondents in their polls.
Looks like Obama will win the majority!