Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
10 Cards in this Set
- Front
- Back
|
What are 'Forecasting Models'
|
⁃ Mathematical models/equations designed to predict ahead of tim what will happen in elections
⁃ Gives some insight about what drives voters |
|
|
What do 'Forecasting Models' assume about elections
|
⁃ Campaigns themselves do not matter all that much
⁃ Variables that predicted outcomes in past election are also good predictors of the current one |
|
|
What variables do Forecasting Models usually contain?
|
⁃ Economic indicators- unemployment rates, GDP, stock market, housing market
⁃ Political indicators- political environment, presidential approval |
|
|
Conventional wisdom on the Effects of the economy on vote share
|
⁃ Good economy rewards incumbent, or at least incumbent party (prospective/retrospective voting)
|
|
|
Some indicators of the Effects of the economy on vote share
|
⁃ GNP/GDP
⁃ Rate of economic growth ⁃ Inflation rates ⁃ Number of new jobs created ⁃ Consumer confidence ⁃ Forecasts about economic future |
|
|
Some debates on Effects of the economy on vote share
|
⁃ Actual economic performance vs. perceived economic importance?
⁃ Pocketbook vs. sociotropic evaluations? ⁃ Prospective vs. Retrospective evaluations? ⁃ Do people punish more than they reward? |
|
|
Effect of presidential approval on vote share
|
stronger relationship when there is incumbent, vs. no incumbent
⁃ Most variation in presidential approval is a function of what independents and out-partisans think |
|
|
Why did recent forecasts in 2000 election go wrong?
|
Prediction?
⁃ most models predicted a Gore victory (52-60% of popular vote) ⁃ Gore got 48.4% (vs. 47.8% for Bush) ⁃ What went wrong? ⁃ economy had been good so long that little reward was given to the democrats where credit was due ⁃ Gore distanced himself from Clinton's policies to show he could stand on his own, but he might have gone too far ⁃ Idiosyncrasies of Clinton approval ⁃ people thought he did a good job, but that he was a bad person |
|
|
Why did recent forecasts in 2004 election go wrong?
|
⁃ Prediction?
⁃ Bush would win, with 49.9%-57.6% of the vote ⁃ Bush got 50.7% of the vote ⁃ Still an overestimation ⁃ Why did most models overestimate? ⁃ Wayne's argument about campaign? 2000 and 2004 had changes in how campaigning was done ⁃ Turnout was higher than history (models don't normally predict how turnout will effect figures) |
|
|
Why did recent forecasts in 2008 election go wrong?
|
⁃ Mostly Economic Models
⁃ Ray Fair's Model- purely economic model based on objective indicators and news about the economy (predicted Obama would win with 52%) ⁃ John Hibb's "Bread and Peace" Model - economic model that makes correlations for wartime effects (Predicts Obama win with 53-54%) ⁃ Models with Political Variables -- APSA Symposium ⁃ Norpoth- Performance in Primaries (50.1% Obama) ⁃ Lockerbie- Ruling party time in White House, voter's expectations about future financial well being (Obama 58.2%) ⁃ Abramowitz- Growth rate of economy, presidential approval, length of time in white house (Obama 54.3%) ⁃ Erikson and Wlezien- Ecomomic indicator and trial heat polls (Obama 52.2% ⁃ Holbrook- Presidential popularity and average level of satisfaction with current personal finances (Obama 55.7%) ⁃ Campbell- Labor Day Gallup Trial Heal Poll and growth in GDP in 2nd quarter of election year (52.7% McCain) ⁃ Klarner- State and district-level factors (Obama 53%) ⁃ Cuzna and Bundrick- Ration of federal outlays to GDP and share of the 2-party vote to incumbents (Obama 52%) ⁃ Actual Results: Obama 52.9%, McCain 45.7% ⁃ Two Party Results: Obama 53.5%, McCain 46.5% (Always equals 100%) ⁃ Who was closest: Klarner, Cuzna and Bundrick, Erikson and Wlezien |
