Posted: 8/29/2012 9:09:53 PM
QUOTE Originally Posted by 14daroad:
Romney had a 43-percent favorable and 44-percent unfavorable rating in nine battleground states heading into the convention, according to an average compiled by Real Clear Politics.
A survey conducted by Wilson Perkins Allen Opinion Research in nine battle ground states Tuesday evening found Romney’s favorable rating among likely voters had jumped to 48 percent. His unfavorable rating dipped to 39.
You should also disclose that Wilson Perkins Allen(WPA) is a paid Republican polling firm. One of its 3 partners writes a blog for redstate.com.
Also, its statistical sample was small: 303 people over 9 states with a margin of error of 5.7%(not a good margin for a polling firm). An average of about 34 likely voters were called in each state.
It would also be helpful to know if there was any bias in the selection of participants or was randomness utilized. WPA does not reveal how it chooses its participants, other than stating "likely voters". Since it's a Republican firm, it's appropriate to question whether some selection bias was involved.
Overall, purely from a mathematical/statistics standpoint, it's not nearly as strong a poll as most others, simply because of its small sample size. A 1000 person poll(similar to most Gallup polls) is more accurate, by reducing the margin of error to 3% or less. Certainly a significant reduction, as all bettors can appreciate.
Bottom line, a larger sample size would have yielded more reliable results and significantly less margin for error. From this WPA poll, I am left with little to interpret that is of statistical value.