Presenting models for making predictions about presidential elections, this brief supplementary text is accessible to students without a statistical background and features discussions of Election 2000 throughout.
Well-grounded in elections theory, this new text introduces students to the major models used to forecast presidential elections and covers a variety of topics through that lens: approval ratings, exit polls, election cycles, nomination process and campaigns, performance of the economy, etc. Lucidly written, it offers an abundance of figures to illustrate concepts to students and include an easy-to-understand explanation of regression so that no prior knowledge of statistics is necessary to read the text. Professor Jones not only summarizes and utilizes the forecasting techniques employed by experts in the past, but brings in new techniques and tools, making a valuable contribution to the methodologies of presidential election forecasting.
- Provides equations and conditional forecasts for the 2000 election so that students can relate forecasting techniques to the most recent presidential election.
- All major forecasting models are presented, including those by Abramowitz, Campbell, Erikson, Fair, Hibbs, Lewis-Beck, Norpoth, Rosenstone and others. They are described in a lucid manner that cuts through stastical jargon.
- An abundance of figures (line plots) identifying win-loss thresholds for an indicator in past elections. Students can visually observe how specific values for an indicator result in wins and losses in an election. No other book presents election indicators in this way.
- An easy-to understand, brief explanation of regression is provided in an appendix. Because of this feature, no prior knowledge of statistics is necessary to read and understand the text.
- Adds breadth to the presidential election forecasting enterprise by illustrating techniques not commonly used for that purpose: identifying bellwether states that closely track the national vote; eliciting predictions from panels of experts, as in the Delhi approach; applying expected utility theory, which has been successfully used by the intelligence community in predicting elections abroad; describing the University of Iowa's online political stock market and documenting its uncanny ability to predict presidential election results.
- A Glossary at the end of the book acquaints students with terminology.
- A terrific supplementary text for courses on voting and elections, campaigning, and the Presidency, as well as Methodologies courses where it can be used to illustrate the use of line plots and the applications of the regression statistical technique.
1. Trial Heat Polls.
Survey Methods.
Lead Time of Surveys.
Linking Surveys to Elections.
Conclusion.
Addendum: Trial Heat Equations for 2000 Election Scenarios.
2. Bellwethers.
Types of Bellwethers.
States: Swingometric and Barometric Bellwethers.
Counties As Bellwethers.
Scenarios Using Bellwethers.
Conclusion.
Addendum: Bellwether Equations for 2000 Election Scenarios.
3. Presidential Approval Ratings.
Rationale.
Establishing the Linkage.
Conclusion.
Addendum: Equations for 2000 Election Scenarios Using Presidential Approval Ratings.
4. Other Public Opinion-Based Techniques.
Exit Polls.
Surveys of Voters' “Likes” and “Dislikes” of Candidates.
Surveys of Past and Present Demographics: Ithiel de Sola Pool's “simulmatics”.
Iowa Electronic Market.
5. Judgment-Based Forecasting Techniques.
Judgment-Based Regression: Budge and Farlie.
Assessments of Key Factors Affecting Elections: Lichtman.
Panels of Experts.
Expected Utility Theory.
Conclusion.
6. Cycles in Presidential Elections.
Cycles of Liberalism and Conservatism: Arthur M. Schlesinger, Sr.
Cycles of Public Purpose and Private Interest: Arthur M. Schlesinger, Jr.
Two-Term Penalty.
Another Approach to Two-Term Cycles.
Conclusion.
7. The Nomination Process and Campaigns.
Primary Elections.
Disorderly Conventions.
Post-Convention Campaigns.
Conclusion.
8. Performance of the Economy.
Retrospective Indicators.
Prospective Indicators.
Conclusion.
Addendum: Equations for the 2000 Election Scenarios Using Selected Economic Indicators.
9. Putting It All Together: Multivariate Forecasting Models.
National-Level Models.
State-Level Models.
Forecast Scenarios for 2000 Election Using Multivariate Models.
Conclusion.
Addendum: A Note on Data Sources.