An International Year of Statistics Event: Estimating Election Outcomes with Different Statistical Methods
Michael Traugott, Research Professor, Center for Political Studies, ISR, University of Michigan
Until recently, public opinion polls conducted before an election were the main tool for explaining voter reactions to the candidates and their campaigns as well as estimating the outcome. Generally, polls conducted closer to Election Day produced better estimates than those conducted earlier, but the 2012 election created problems for some pollsters, especially those conducted by Gallup. For the last few presidential election cycles, forecasters and statistical modelers have produced estimates of election outcomes with a focus on such factors as how much in advance the estimates are produced and with what level of accuracy. With more polling conducted at the state level, data aggregators in particular are producing estimates of the outcome in each state and translating them into estimates of electoral votes as well as shares of the popular vote at the national level. These shifts in statistical techniques are also changing the way that political campaigns are covered by the press.