Citizen forecasting is the subject that ties this thesis together. Citizens base their vote choice in part on their forecasts who will win the election. We all should care, therefore, about how accurate the forecasts are and how we may help citizens to form better ones. At first the task of forecasting seems Herculean: for instance, citizens need to overcome bias. measurement problems, and ignorance. But other features simplify the task: the British political system is only slowly changing and the public reflects upon itself by publishing opinion polls. Election fo recasting holds challenges, but citizens can manage them without Herculean effort. Indeed, most citizens correctly forecasted who won in theif constituency in the 2010 British General Election. Groups of citizens are even better at forecasting than individuals, showing a wisdom of crowds ·effect. Aggregating the citizen forecasts for each constituency correctly predicted a hung parliament, with the Conservatives as the largest party. Citizens achieved the accuracy without being unbiased Bayesian learners. Although many researchers already claimed that partisanship biases information processing, none controlled for prior belief certainty. All researchers found that partisan groups differ in their posterior belief, but the finding is not enough to show that partisans are biased. Different posterior beliefs can result from partisans proceSSing information in the same way, but starting with different prior beliefs. It turns out, however, that even after controlling for prior certainty, partisan bias persists. Although citizens are biased, they are efficient information processors. They use Ufast and frugal" heuristics that yield by·and-large accurate forecas ts with very little information. Indeed, a simple path to correct forecasting is to forecast the incumbent party to win again in the constituency, and the party leading in a campaign poll to win the most votes nationally. tdost citizens seem to follow this path.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:601492 |
| Date | January 2013 |
| Creators | Murr, Andreas Erwin |
| Publisher | University of Essex |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0022 seconds