Princeton Election Consortium

A first draft of electoral history. Since 2004

Democrats outperforming national polls in district-level polls

November 2nd, 2010, 1:43pm by Sam Wang


The Gallup generic Congressional preference poll (R+15%) has drawn notice, including from ostensibly statistics-savvy poll geeks. It’s unclear why since individual data points always vary more than averages. Then again, as the saying goes, if it bleeds it leads. But there’s something more surprising.

In addition to random variation, Democrats seem to have a hidden advantage relative to national House preference polls, which are averaging R+7%. But district-level polls (which indicate 230R-205D) are said to be equivalent to R+4%. Can both be right? District polls tend to be concentrated in key districts. If the Republicans’ final House seat advantage matches district polls (a majority of 25 seats), it’s an underperformance relative to national surveys. Lucky for the Democrats – though not lucky enough. Maybe they have a better machine / strategy. Or maybe it’s systematic error when the question is asked generically.

Based on the excellent performance of direct polls in 2004/2006/2008, the hypothesis to beat is listed above. If the House is outside [228, 232] R or the Senate is outside [50, 52] D, then I’ll want to know why.

Tags: Politics

2 Comments so far ↓

  • LondonYoung

    Well, the miss is significant on both intervals, and in opposite directions. Why?

    • Sam Wang

      Not yet it’s not, at least until the Washington Senate race (Murray vs. Rossi) is decided. However, at best that would lead to a numerical agreement with the original Senate CI. Basically the culprit there is Nevada polls, which are a mystery to me. If you look at them they clearly prediced an Angle victory with high confidence.

      House: see the comment thread on the other posting. Not sure yet. My guess is that they were too infrequent to catch a late wave toward the Republicans.

Leave a Comment