(Cross-posted at the Princeton Election Consortium)
National polling averages are contradictory: Pollster gives Romney by +0.4%, TalkingPointsMemo gives Romney by +0.6%, and RealClearPolitics gives Obama by +3.4%. Who's right, and why are these measures fluctuating from day to day?
Much of the discrepancy is explained by my post on the problem with averaging.
These measures bounce around far less when using medians. RCP data since July 18th give a median of Obama +4.0 ± 1.3% (estimated SEM, n=5), consistent with the Princeton Election Consortium's Meta-margin, which is based on state polls. And a few weeks ago, the RCP median was at Obama +2.0%, again matching the Meta-analysis.
A second answer is the use of a long time window. These other sites' engines achieve a smoother average by looking over a longer time window. The use of medians would allow them to achieve better time-sensitivity.
Third and finally, note that Pollster and TPM are dominated by many Rasmussen data points. It is a clear methodological error to rely so heavily on one pollster. At a minimum, all data from the same pollster within a given time period should be combined to make one contributing data point.
Bottom line: Obama is currently ahead by about 4-5%. But national poll aggregators are currently not showing it.