Updated election forecasting model—based on economic data—still points to Romney win

October 5, 2012

(Phys.org)—An update to an election forecasting model announced by two University of Colorado professors in August continues to project that Mitt Romney will win the 2012 presidential election.

According to their updated analysis, Romney is projected to receive 330 of the total 538 Electoral College votes. President is expected to receive 208 votes—down five votes from their initial prediction—and short of the 270 needed to win.

The new forecast by political science professors Kenneth Bickers of CU-Boulder and Michael Berry of CU Denver is based on more recent economic data than their original Aug. 22 prediction. The itself did not change.

"We continue to show that the economic conditions favor Romney even though many polls show the president in the lead," Bickers said. "Other published models point to the same result, but they looked at the national popular vote, while we stress state-level economic data."

While many forecast models are based on the popular vote, the model developed by Bickers and Berry is based on the Electoral College and is the only one of its type to include more than one state-level measure of . They included economic data from all 50 states and the District of Columbia.

Their original was one of 13 published in August in PS: Political Science & Politics, a peer-reviewed journal of the American Political Science Association. The journal has published collections of models every four years since 1996, but this year the models showed the widest split in outcomes, Berry said. Five predicted an Obama win, five forecast a Romney win, and three rated the 2012 race as a toss-up.

The Bickers and Berry model includes both state and national unemployment figures as well as changes in real per capita income, among other factors. The new analysis includes unemployment rates from August rather than May, and changes in per capita income from the end of June rather than March. It is the last update they will release before the election.

Of the 13 battleground states identified in the model, the only one to change in the update was New Mexico—now seen as a narrow victory for Romney. The model foresees Romney carrying New Mexico, North Carolina, Virginia, Iowa, New Hampshire, Colorado, Wisconsin, Minnesota, Pennsylvania, Ohio and Florida. Obama is predicted to win Michigan and Nevada.

In Colorado, which Obama won in 2008, the model predicts that Romney will receive 53.3 percent of the vote to Obama's 46.7 percent, with only the two major parties considered.

While national polls continue to show the president in the lead, "the president seems to be reaching a ceiling at or below 50 percent in many of these states," Bickers said. "Polls typically tighten up in October as people start paying attention and there are fewer undecided voters."

The state-by-state economic data used in their model have been available since 1980. When these data were applied retroactively to each election year, the model correctly classifies all presidential election winners, including the two years when independent candidates ran strongly: 1980 and 1992. It also correctly estimates the outcome in 2000, when Al Gore won the but George W. Bush won the election through the Electoral College.

In addition to state and national unemployment rates, the authors analyzed changes in personal income from the time of the prior presidential election. Research shows that these two factors affect the major parties differently: Voters hold Democrats more responsible for unemployment rates, while Republicans are held more responsible for fluctuations in personal income.

Accordingly—and depending largely on which party is in the White House at the time—each factor can either help or hurt the major parties disproportionately.

In an examination of other factors, the authors found that none of the following had a statistically significant effect on whether a state ultimately went for a particular candidate: The location of a party's national convention, the home state of the vice president or the partisanship of state governors.

The authors also provided caveats. Their model had an average error rate of five states and 28 Electoral College votes. Factors they said may affect their prediction include the timeframe of the used in the study and that states very close to a 50-50 split may fall in an unexpected direction due to factors not included in the model.

"As scholars and pundits well know, each election has unique elements that could lead one or more states to behave in ways in a particular election that the model is unable to correctly predict," they wrote.

All 13 election models can be viewed on the PS: & Politics website at http://journals.cambridge.org/action/displayJournal?jid=PSC

Explore further: CU-Boulder analysis of election factors points to Romney win

Related Stories

Recommended for you

New paper answers causation conundrum

November 17, 2017

In a new paper published in a special issue of the Philosophical Transactions of the Royal Society A, SFI Professor Jessica Flack offers a practical answer to one of the most significant, and most confused questions in evolutionary ...

Chance discovery of forgotten 1960s 'preprint' experiment

November 16, 2017

For years, scientists have complained that it can take months or even years for a scientific discovery to be published, because of the slowness of peer review. To cut through this problem, researchers in physics and mathematics ...


Adjust slider to filter visible comments by rank

Display comments: newest first

not rated yet Oct 05, 2012
Forecasts are a dime a dozen and not worth it.
not rated yet Oct 05, 2012
Recent polls show Obama well ahead in Virginia, Michigan, and Pennsylvania. Berry's model predicts 46, 51 and 48%.
not rated yet Oct 05, 2012
Polls are about as accurate as opinions. It's so easy to adjust poles based on the households called, their district or even the time of day the calls are made.

These forecasts may be a bit more scientifically accurate over time, but there will always be variables that will be difficult to calculate, if not even seemingly chaotic.
not rated yet Oct 06, 2012
Dudes, you don't need to convince me; let's see you convince Nate Silver.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.