Election Prediction and STEM

By Sheldon H. Jacobson

Every U.S. presidential election attracts the world’s attention, and this year’s election will be no exception. The decision between the two major party candidates, Hillary Clinton and Donald Trump, is challenging for a number of voters; this choice is resulting in third-party candidates like Gary Johnson and Jill Stein collectively drawing double-digit support in some polls. Given the plethora of news stories about both Clinton and Trump, November 8 cannot come soon enough for many.

In the Age of Analytics, numerous websites exist to interpret and analyze the stream of data that floods the airwaves and newswires. Seemingly contradictory data challenges even the most seasoned analysts and pundits. Many of these websites also employ political spin and engender subtle or not-so-subtle political biases that, in some cases, color the interpretation of data to the left or right.

Undergraduate computer science students at the University of Illinois at Urbana-Champaign manage Election Analytics, a nonpartisan, easy-to-use website for anyone seeking an unbiased interpretation of polling data. Launched in 2008, the site fills voids in the national election forecasting landscape. 

Election Analytics lets people see the current state of the election, free of any partisan biases or political innuendos. The methodologies used by Election Analytics include Bayesian statistics, which estimate the posterior distributions of the true proportion of voters that will vote for each candidate in each state, given both the available polling data and the states’ previous election results. Each poll is weighted based on its age and its size, providing a highly dynamic forecasting mechanism as Election Day approaches. Because winning a state translates into winning all the Electoral College votes for that state (with Nebraska and Maine using Congressional districts to allocate their Electoral College votes), winning by one vote or 100,000 votes results in the same outcome in the Electoral College race. Dynamic programming then uses the posterior probabilities to compile a probability mass function for the Electoral College votes. By design, Election Analytics cuts through the media chatter and focuses purely on data. 

Election Analytics allows students to participate in an activity that transforms classroom knowledge into a practical tool used by thousands of people. Indeed, the website represents a learning laboratory for these students. They experiment with new descriptive analytics, participate in the interface design, and analyze the data for posting, all of which showcase their creativity. The website went through several facelifts since its launch in 2008, including a number of interface upgrades, the addition of new descriptive analytics, functionality to measure the impact of third-party candidates, options to exclude certain pollsters and provide various left or right-leaning bias scenarios for undecided voters, and a prediction for which party will control the Senate. The students update the website daily as new polling data becomes available.

Independent of the election, Election Analytics permits students to work on a real-world project with practical implications. Is this not one of the roles of STEM education, to transform technical ideas into practical tools for the benefit of society? 

Election Analytics is one of the more quantitative, data-driven, election forecasting websites available. In addition to their own analysis, The New York Times features several websites that also attempt to make sense of the information disseminated to voters. These site include fivethirtyeight.com, run by Nate Silver; the Daily Kos; Predictwise.com, which uses a market-based approach; the Princeton Election Consortium, run by Sam Wang, which also uses Bayesian statistics; and the following three sites, which provide qualitative measures of the state of the election: the Cook Political Report, the Rothenberg and Gonzales Political Report, and Sabato’s Crystal Ball. Other available sites include 270 to Win and electoral-vote.com. 

The Election Analytics website may indeed provide an accurate snapshot of which candidate will win the White House and control the Senate in November; its track record since 2008 compares very favorably to the other sites listed above. However, the students involved in designing and maintaining the Election Analytics website are the real winners. Election Analytics is an activity that will launch their STEM learning in ways that make a difference far beyond this year’s election.  

Sheldon H. Jacobson is professor and director of the Simulation and Optimization Laboratory at the University of Illinois at Urbana-Champaign. Visit his website for more on his work.