David Kirkpatrick

October 21, 2008

Forecasting the House and Senate …

from the American Political Science Association.

The release:

October 21, 2008: Election Forecast Predicts Democrats Will Gain 3 Seats in Senate, 11 in House APSA Press Release
 

For Immediate Release

In forecasts made in July 2008, Democrats seen as gaining in both chambers but falling short of achieving control of the Senate.

Washington, DC—An election forecast model developed by a political scientist 99 days before the 2008 elections and before the recent Wall Street crisis predicts significant Democratic gains in the 2008 congressional elections—including 11 seats in the U.S. House of Representatives and 3 seats in the U.S. Senate.

The predictions are made in an article authored by Carl Klarner (Indiana State University) and published in an election-specific symposium in the October 2008 issue of PS: Political Science and Politics, a journal of the American Political Science Association (APSA).  The full symposium is available online at http://www.apsanet.org/content_58382.cfm.

Traditionally, efforts to call elections rely either on district- and state-level analyses limited to recently collected information (such as polls) or aggregate forecasting models measuring national trends. Klarner notes that “most forecasting models of House and Senate elections have not made predictions at the state or district level” and that “how national factors influence election outcomes is contingent on the distribution of votes across districts or states.” His 2008 forecast refines his own previous work in this area to use a model that combines both approaches.

The House and Senate forecasts were made in late July 2008, and the author’s model focuses on the percent of the major-party vote that the Democratic candidate received in a state or district. Klarner considers three main sets of factors in examining past elections from 1954 onward: district partisan composition, candidate attributes, and national partisan tides. The weighting of these factors is based on a range of historical and empirical data—including most recent votes for Democrats in a district; results of the most recent presidential vote; incumbency; prior experience in candidates; national vote intentions reported in surveys; presidential approval; performance of the economy; and the “midterm penalty” for the president’s party.

The model’s House prediction includes the following items of note:

  • Democrats will receive 247 seats in the House—a gain of 11 seats overall. 
  • There is a 95% probability that Democrats will have between 233 and 266 seats after the election and a 67% probability that they will have between 240 and 255 seats.
  • There is a about a 0% probability that the Democrats will lose control of the House.

The model’s Senate prediction includes the following items of note: 

  • Democrats will have 54 Senate seats after the election—a net gain of 3.
  • There is a 95% probability that the Democrats will win between 12 and 19 seats out of the 35 seats up this election.
  • There is a 2.4% chance the Democrats will lose control of the Senate.
  • There is a 0.3% chance that the Democrats will obtain a filibuster-proof majority of 60 seats.

By integrating long-term data analysis with current local and national political factors, this election forecast model reflects ongoing efforts by political scientists to analyze election dynamics in the US.  Notably, this prediction of the outcome of the 2008 congressional election was made well before the recent Wall Street financial crisis has made the political landscape more favorable to Democrats.

# # #

The American Political Science Association (est. 1903) is the leading professional organization for the study of politics and has over 14,000 members in 80 countries. For more news and information about political science research visit the APSA media website, www.politicalsciencenews.org.

October 17, 2008

Six of nine presidential election forecasts predict Obama

I’ve been running polls and statistical projections. Now here’s the science.

The release:

October 16, 2008: 6 of 9 Presidential Election Forecasts Predict Obama Will Win 2008 Popular Vote APSA Press Release

For Immediate Release

6 of 9 Presidential Election Forecasts Predict Obama Will Win 2008 Popular Vote

WASHINGTON, D.C.—Most of nine forecast models developed by political scientists predict a victory for Senator Barack Obama over Senator John McCain in the two-party contest for the popular vote in the 2008 presidential election. Obama is predicted to win an average of 52% of the vote with an 80% probability that he will gain more than half the total two-party popular vote.

Six out of the nine presidential election forecasts predict an Obama victory with popular vote totals ranging from 50.1% to 58.2%, while two predict a race too close to call and one predicts a narrow McCain victory.  All of the predictions appear in an election-themed symposium in the October issue of PS: Political Science and Politics, a journal of the American Political Science Association (APSA).  The forecasts, published in advance of presidential and mid-term elections, are available online at: https://www.apsanet.org/content_58382.cfm.

The forecasts are based on different combinations of statistical and historical data and differ in their complexity and how far in advance their predictions were made.  The earliest forecast was made 294 days in advance while the latest was made 60 days before the election; however, all were made before the Wall Street financial crisis of the past few weeks.  Together these forecasts use a range of approaches and indicators that are critical to understanding national electoral processes and the dynamics at work in U.S. presidential elections.  Brief summaries of each are provided below:

·         In the earliest completed forecast made in January 2008, Helmut Norpoth’s (Stony Brook University) “Primary Model” uses candidate support in presidential primaries to predict the general election two-party popular vote outcome. Norpoth’s forecast makes Senator Obama the favorite by a razor-thin margin, predicting a 50.1% to 49.9% Obama victory, but also indicates only a 50% chance that Obama will gain a majority.

·         Using data from the second quarter of 2008, Brad Lockerbie (University of Georgia) employs two variables that are decided well in advance of the presidential conventions to forecast the presidential election: the amount of time a party has controlled the White House, and voters’ expectations concerning their financial well-being over the course of the next year. Lockerbie predicts an impressive victory by Barack Obama, with John McCain gaining only 41.8% of the two-party popular vote en route to a loss in the 2008 presidential election.

·         Alan I. Abramowitz’s (Emory University) “Time-for-Change” forecast model, completed in August 2008, is based on the assumption that a presidential election is fundamentally a referendum on the performance of the incumbent president. Abramowitz’s model employs three variables: the growth rate of the economy during the second quarter of the election year, the incumbent president’s approval rating at mid-year, and the length of time the incumbent president’s party has controlled the White House. Abramowitz predicts Senator Barack Obama will receive 54.3% of the two-party popular vote in the 2008 presidential election.

·         Robert S. Erikson (Columbia University) and Christopher Wlezien’s (Temple University) forecast model analyzes a combination of leading economic indicators and trial-heat polls from August 2008 to predict the election’s final outcome. Their model predicts that Senator Obama will win 52.2% of the two-party popular vote, compared with 47.8% for Senator McCain.

·         Michael Lewis-Beck (University of Iowa) and Charles Tien (Hunter College and the Graduate Center, CUNY) “Jobs Model” weighs the sitting president’s popularity and the actual number of jobs created or lost during his term with the powers of incumbency and economic growth. Unadjusted, it predicts that Senator Obama will win 56.57% of the two-party popular vote in what will amount to “the greatest incumbent popular vote loss on record from 1948.” However, with refinements to factor in the impact of race, the Jobs Model predicts a final outcome where Senator Obama will win by a smaller margin, garnering 50.1% of the two-party vote and with a 50% chance that Obama will gain a majority. The forecast was made in August 2008.

·         Thomas M. Holbrook’s (University of Wisconsin-Milwaukee) forecast model emphasizes presidential approval and the average level of satisfaction with personal finances in the summer before the presidential election.  Completed in August 2008, his model predicts Barack Obama will garner 55.7% of the two-party presidential popular vote compared to John McCain’s 44.3%.

·         Symposium editor James E. Campbell’s (University at Buffalo, SUNY) “Trial Heat” forecast model integrates the incumbent’s trial-heat Labor Day Gallup numbers and the real growth in the GDP in the second quarter of the election year. Completed in early September 2008, the Campbell model breaks with the other forecasts to predict that Senator McCain should be expected to receive 52.7% of the two-party popular vote.

·         Alfred G. Cuzan and Charles M. Bundrick (University of West Florida) use the “Fiscal Model” forecast and emphasize the relationship between the ratio of federal outlays to GDP and the share of the two-party vote going to incumbents.  Completed in August 2008, their forecast predicts the Democratic candidate, Senator Barack Obama, will emerge victorious by a 52%-48% margin.

·         Carl Klarner (Indiana State University) employs a forecast model that focuses on broad range of state and district-level factors. Completed in late July 2008, his forecast predicts Senator Obama will obtain 53.0% of the popular vote and 346 electoral votes.

The 2008 presidential election is taking place in an extraordinary environment.  The open-seat nature of the contest, the implications of President Bush’s low approval ratings, Senator Obama’s decision to forego public financing of his campaign, the effect of race on a contest featuring the first black candidate of a major party in U.S. history, the relative levels of party unity, and the impact of the Wall Street meltdown after these forecasts were produced all combine to make the outcome of the 2008 election unusually difficult to predict.  Nevertheless, more of the forecasts predict an Obama victory than not, as the “median of these nine forecasts indicates that Senator McCain will receive 48% of the two-party popular vote,” concludes symposium editor James Campbell.

# # #

The American Political Science Association (est. 1903) is the leading professional organization for the study of politics and has over 14,000 members in 80 countries. For more news and information about political science research visit the APSA media website, www.politicalsciencenews.org.