Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

RandySF

(58,884 posts)
Thu Oct 18, 2012, 03:22 AM Oct 2012

Citizens' forecasts point to Obama

Here is a method for forecasting U.S. presidential elections that has been widely overlooked: citizens' expectations of who will win the election. Such vote expectation surveys ask respondents a very simple question: "Who do you think will win the U.S. presidential election?".

Pollsters have regularly asked variations of this question since the 1940s. Yet, such vote expectation surveys have been virtually ignored by election observers and are rarely reported in the media. Likewise, researchers have widely overlooked vote expectation surveys as a method for election forecasting. Little is known about their accuracy compared to established methods.
In a recent study, Andreas Graefe performed this test. He analyzed the accuracy of 190 vote expectation surveys that were conducted between 1932 and 2008. In only 18 of the 190 (9%) surveys did the majority of respondents fail to pick the winner. Across the six elections from 1988 to 2008, vote expectations outperformed four established methods (polls, prediction markets, econometric models, and experts’ judgment) when answering the question that is probably most important to the election observer: who will win? When predicting vote shares, only prediction markets and poll projections were slightly more accurate. The working paper with the complete analysis can be accessed at SSRN.


What do vote expectation surveys predict for 2012?

Since March 2012, when the first survey was published, citizens continuously predicted Obama to win. The respective vote-share prediction derived from the survey ranged from 51.3% to 52.7% for Obama.

The chart below shows the citizen forecasts vs. combined expectations from experts (i.e., Polly's expert survey) and "self-selected experts" at the IEM. Most of the time, the citizen forecasts was within the range of both alternatives. While the IEM tended to be more optimistic for Obama, experts tended to provide a more conservative forecast.



http://pollyvote.forecastingprinciples.com/index.php/pollycomp-home/1/301-citizens-forecasts-point-to-obama.html

Latest Discussions»General Discussion»Citizens' forecasts point...