Study Suggests That Polling Is Still Best Predictor Of Election Outcomes

  • comments
  • print
  • email
Feb 06, 2017 08:57 AM EST

Although some people believe that Donald Trump's election as president of the United States is an evidence that electoral polling is no longer functional, researchers recently reported that they have developed models using global polling data which has the ability to correctly predict up to 90 percent of election outcomes around the world.

The researchers focused more on executive elections whereby voters cast ballots directly for the person who will hold executive office, instead of having the potential leader elected by parliament. They report that their discovery offers strong evidence that polling data, used correctly, is the best predictor of election outcomes.

"This study suggests polling data can be utilized not just in the United States but globally to predict election outcomes," lead author and political scientist, Ryan Kennedy of the University of Houston's Center for International and Comparative Studies said.  He added that it would be a mistake to abandon the enterprise as the future really is in trying to make better quantitative predictions, according to Science Daily.

The study authors tested the model as part of a project sponsored by the Intelligence Advanced Research Projects Activity - part of the office of the Director of National Intelligence that supports studies with applications for the intelligence community. They submitted forecasts in Latin America two weeks before election in 2013 and 2014 and correctly predicted the winners in 10 out of 11 elections (90.9 percent of the time).

Another test which involved live predictions for all global direct executive elections starting in 2013 delivered a success rate of 80.5 percent. The researchers developed the model using an election dataset comprising of more than 500 elections in 86 countries, as well as a separate dataset that included polling data from 146 elections.

They found that economic growth had little influence on election outcomes. They believe that even though there is a long literature on the impact of economic growth in elections, they found only little to suggest a global rule. The researchers found only minor impacts for inflation with less democratic institutions favoring the incumbent party.

Kennedy said open democratic elections and whether a candidate was an incumbent increased the likelihood of accurate predictions as persons in power is able to use their power and name recognition to win subsequent elections. However, he noted that the most valuable predictor is the polling results as it has proved true across the globe.

Polling techniques used for decades in the United States also are successful in countries without a history of effective polling, Kennedy said, according to Eurekalert. The researchers said that their findings are salient in light of Trump's victory, which prompted predictions of the death of data and polling. Kennedy said noting that he and his collaborators predicted an 84 percent chance of a Clinton victory which means a 16 percent chance of a Trump victory.

The researchers stated that they often believe that anything over 50 percent absolutely means an outcome is going to occur but that is not necessarily the case. They published their findings Feb. 2 in the journal Science.

Join the Conversation
Real Time Analytics