Fundamental models for forecasting elections at the state level

This paper develops new fundamental models for forecasting presidential, senatorial, and gubernatorial elections at the state level using fundamental data. Despite the fact that our models can be used to make forecasts of elections earlier than existing models and they do not use data from polls on voting intentions, our models have lower out-of-sample forecasting errors than existing models. Our models also provide early and accurate probabilities of victory. We obtain this accuracy by constructing new methods of incorporating various economic and political indicators into forecasting models. We also obtain new results about the relative importance of approval ratings, economic indicators, and midterm effects in the different types of races, how economic data can be most meaningfully incorporated in forecasting models, the effects of different types of candidate experience on election outcomes, and that second quarter data is as predictive of election outcomes as third quarter data.     

政治科学预测方法研究——以选举预测为例

预测是人类的基本认知活动。作为社会科学预测性研究的重要领域,选举预测是促进政治学理论和方法创新的动力。选举预测不仅限于民调,科学的选举预测可以划分为四类范式:意见聚合范式;模型范式;混合范式;大数据范式。每类范式包含若干种预测方法,不同预测方法在准确性、超前性、解释力、可重复性、中立性和预测成本等评判标准上各具优势,预测方法的选择是基于预测目标和应用场景的综合权衡。目前,选举预测领域呈现多元化、交叉互补、跨学科的发展趋势;其未来发展方向是“理论驱动”与“数据驱动”相结合、微观预测变量与宏观预测变量相结合、预测与解释相结合,以政治科学为基础,通过跨学科合作和前沿技术手段推进预测范式的创新,并在预测实践中不断更新升级。预测性研究与解释性研究可以构成“双螺旋”的共生互补关系,共同致力于增进人类对复杂政治活动的理解和研判,提升政治学研究的科学水平和政策应用价值。     

Do party leadership contests forecast British general elections?

When assessing election forecasts, two important criteria emerge: their accuracy (precision) and lead time (distance to event). Curiously, in both 2010 and 2015 the most accurate forecasts came from models having the longest lead time—albeit at most 12 months. Can we increase the lead time further, supposing we tolerate a small decrease in accuracy? Here, we develop a model with a lead time of more than 3 years. Our Party Leadership Model relies on the votes of MPs when selecting their party leader. We assess the forecasting quality of our model with both leave-one-out cross-validation and a before-the-fact forecast of the 2019 general election. Compared to both simple forecasting methods and other scientific forecasts, our model emerges as a leading contender. This result suggests that election forecasting may benefit from developing models with longer lead times, and that party leaders may influence election outcomes more than is usually thought.     

The British general election: Synthetic forecasts

Election forecasting work, in Britain and elsewhere, has been confined mainly to traditional approaches – statistical modeling or poll-watching. We import a new approach, which we call synthetic modeling. These models, developed out of forecasting efforts from the American 2012 presidential election, we are currently testing in a comparative European context. In this exercise, we apply the strategy to forecasting British general elections, in particular the upcoming 2015 contest. These synthetic forecasts come from hybrid models blending structural knowledge with contemporary public opinion. We generate ongoing nowcasts, from six months prior, to one month before, Election Day itself.     

“Wisdom of crowds”? A decentralised election forecasting model that uses citizens’ local expectations

Many studies report the “wonders of aggregation” and that groups (often) yield better decisions than individuals. Can this “wisdom of crowds”-effect be used to forecast elections? Forecasting models in first-past-the-post systems need to translate vote shares into seat shares by some formula; however, the seat–vote ratio alters from election to election. To circumvent this problem, this paper proposes citizen forecasting, which aggregates citizens’ local expectations to directly forecast constituencies. Using data from the 2010 British Election Study, this paper finds (1) that groups are better forecasters than individuals, (2) that citizen forecasting correctly predicts a hung parliament, and (3) that marginality and group size are important predictors for “getting it right”.     

Citizen forecasting: Can UK voters see the future?

The leading approaches to election forecasting have been statistical models, prediction markets, or vote intention polls. This paper explores a different, little used approach - citizen forecasting - a method never yet tried for the case of the United Kingdom. Citizen forecasting simply asks voters in a scientific pre-election survey who they think will win. This aggregated (averaged) result is then assessed for its accuracy in forecasting the overall election outcome. This assessment is carried out on UK general elections from 1951 to 2005, and then applied to 2010. We find, among other things, that generally speaking the majority of the electorate can foretell the winner in advance.     

General election forecasting in Britain: A comparison of three simple models

The advent of three-party politics in Britain with the February 1974 general election has introduced an uncertainty into electoral and parliamentary politics unprecedented in the post-war period. In these circumstances, election forecasting has assumed a special interest and significance for academics, politicians, political commentators, and the like. This article presents and assesses the performance of three forecasting instruments, the ‘incremental’, ‘opinion polling’ and ‘economic’ models. They are estimated over the period 1951–1983 and are then used to predict the share of the vote won by the governing, opposition and Alliance parties in the 1987 general election. All are successful in the sense that they forecast the continuation of the Conservative party's electoral dominance. with Labour and the Alliance a poor second and third. Only the economic model, however, generates a reasonable forecast of the gap separating the major parties and it is used to predict the distribution of parliamentary seats between them. It is seen to be substantially more accurate for the government than for the opposition, which is itself a reflection of the uncertainty introduced into British politics by the emergence of a significant third party in recent elections.     

Forecasting proportional representation elections from non-representative expectation surveys

This study tests non-representative expectation surveys as a method for forecasting elections. For dichotomous forecasts of the 2013 German election (e.g., who will be chancellor, which parties will enter parliament), two non-representative citizen samples performed equally well than a benchmark group of experts. For vote-share forecasts, the sample of more knowledgeable and interested citizens performed similar to experts and quantitative models, and outperformed the less informed citizens. Furthermore, both citizen samples outperformed prediction markets but provided less accurate forecasts than representative polls. The results suggest that non-representative surveys can provide a useful low-cost forecasting method, in particular for small-scale elections, where it may not be feasible or cost-effective to use established methods such as representative polls or prediction markets.     

Forecasting the 2015 British general election: The Seats-Votes model

This paper applies the Seats-Votes Model to the task of forecasting the outcome of the 2015 election in Britain in terms of the seats won by the three major parties. The model derives originally from the ‘Law of Cubic Proportions’ the first formal statistical election forecasting model to be developed in Britain. It is an aggregate model which utilises the seats won by the major parties in the previous general election together with vote intentions six months prior to the general election to forecast seats. The model was reasonably successful in forecasting the 2005 and 2010 general elections, but has to be modified to take into account the ‘regime shift’ which occurred when the Liberal Democrats went into coalition with the Conservatives in 2010.     

The Changing Nature … and Costs … of Election Administration

Changes during the last decade have significantly altered the nature of election administration and driven up its costs. Interviews with practitioner representatives from the elections community reveal a number of issues that continue to influence this field. Expense data from two counties illustrate how policy changes affect election costs. Scheduled ending of federal funding threatens a “perfect storm” in election finance that may undermine hard‐won advances from 2000 to 2010. Possible cost reduction measures as well as critical challenges facing election administration are forecast for 2010 and the years ahead.     

Reordering the French election calendar: Forecasting the consequences for 2002

Abstract. French politicians sometimes change election rules for political advantage. In the Spring of 2001, the ruling Socialists pushed through the inversion of the 2002 election calendar despite stiff opposition. What were the consequences of scheduling the presidential election before the legislature elections? Employing new techniques for French election forecasting, we show that the inversion brought great vote gains to the Socialist-led coalition in both the legislative and presidential arenas. One advantage of this forecasting methodology is that it allows counterfactual comparisons in advance of the actual election contest. Comparing the scenarios 'inversion' versus 'no inversion', the Socialist leadership appeared highly strategic and successful.     

The anglicisation of political marketing: how Blair ‘out‐marketed’ Clinton

Despite increasing academic interest in political marketing, confusion remains over its meaning and scope. Whilst most research focuses on its use in election campaigns, some argue that marketing influences other aspects of political behaviour. This paper contends that a lack of comparative research has contributed to this confusion. Theories derived from country‐specific studies may not be broadly applicable due to the impact of systemic differences. To show this, it analyses the case studies of Clinton in the 1992 US presidential election and Blair in the 1997 UK general election. Comparing the use of marketing in the two cases reveals that while systemic features created the scope for a broader, more coordinated and delivery‐oriented approach in the case of Labour, in both cases marketing influenced the design as well as the presentation of the ‘products’ on offer. This suggests that the potential applications of political marketing are broader than conventional definitions imply. Copyright © 2002 Henry Stewart Publications.     

Party Propaganda in Motion: Finland 1954–1991

This study addresses itself to the election campaigns of political parties in Finland and the changes which have taken place over time. Political parties are no longer strictly defined by ideology-rigid class structures have disappeared; and the primary hypothesis is that the parties have grown increasingly similar with respect to propaganda techniques and voter orientation. The empirical data in this study are based on the election platforms and campaign posters of the four largest parties in preparation for the parliamentary elections of 1954, 1966, 1975, 1987 and 1991. In order to elucidate the election propaganda of the political parties even further, a partial examination has been made of election campaign advertising by individual parties and candidates in four political affiliated newspapers as well as in an independent daily newspaper.
The primary hypothesis is clearly supported by this research: political parties have structured their rhetoric to appeal to all voters as opposed to having earlier directed their message to a traditional voter class/category, such as farmers, manual workers and white collar workers. The Communist party platform of 1975, which displayed a high degree of class/category specific orientation, constitutes an exception from this rule. The use of offensive propaganda has decreased, but there are exceptions from this rule as well. This article serves as a summary of the main findings of my licentiate thesis in political science at the Abo Academy University.     

Collective failure? Lessons from combining forecasts for the UK's referendum on EU membership

Research on election forecasting suggests there are benefits from combining different sources of information. This paper discusses the experience of a combined forecasting method that was developed for the UK’s EU referendum in 2016. The sources included opinion polls, vote expectation surveys, prediction and betting markets, expert and volunteer forecasts, and various forecasting models based on polling and other kinds of data. Averages of sources within each of these categories all, in our final forecast, suggested Remain was more likely to win but with varying degrees of certainty. Combining them produced a forecast that beat some but not others. Opinion polls and citizen forecasts came closest to the true outcome. Betting and prediction market participants and volunteer forecasters were the most overconfident that the UK would vote Remain. This may have been because they were distrustful of the polls following the 2015 general election miss and had too strong an expectation of a late swing towards the status quo similar to those in Scotland in 2014 and Quebec in 1995.     

Piecing it all together and forecasting who governs: The 2015 British general election

This paper discusses a new probabilistic forecasting method that was designed for the 2015 British general election. It proceeds in a series of steps from opinion poll averaging, forecasting national-level vote shares and uncertainty estimates, and subsequent simulation of hypothetical election results, through modelling of constituency polls and survey data to identify and adjust for patterns in the constituency-level variation in party performance, and finally to probabilistic forecasting of seat outcomes and of different combinations of parties commanding relevant governing majorities in parliament.     

Forecasting the Presidential Vote in the States, 1948-2004

SUMMARY

This research updates, revises, and extends a forecasting equation of the presidential vote in the states. The original equation was composed of sixteen predictors available well before the election and estimated with data from 531 state elections from 1948 to 1988. The equation was empirically strong, based on objective predictors, and more parsimonious than previous equations. Reexamining the equation with 200 additional state elections from 1992, 1996, 2000, and 2004 indicates that the equation remains well supported, but suggests several opportunities for improvement. A revised equation has a mean absolute error of 3.2 percentage points and correctly predicts 87 percent of all electoral votes. The extension of the analysis adapts the forecast equation to predict electoral vote winners, conducting a logit analysis that takes into account both the size of the state and the closeness of its previous election. This produces more accurate forecasts of both electoral vote winners in the states and the division of the aggregate national electoral vote.     

Voting and economic factors in French elections for the European Parliament

This article builds and estimates several econometric models that explain and forecast the outcomes of the French elections for the European Parliament. These models show the influence of the change in the local unemployment rate to explain the vote for the moderate Left and the moderate Right in the French elections for the European Parliament. These models appear to be accurate in forecasting the elections of the past, and their behavior for the 2009 French election for the European Parliament is satisfactory.     

Did you see it coming? Explaining the accuracy of voter expectations for district and (sub)national election outcomes in multi-party systems

The expectations of voters regarding election outcomes appear to be mostly influenced by their own political preferences. This raises two important questions. First, once partisan predispositions have been accounted for, how much do other variables like interest in the campaign, election news attentiveness, political knowledge, education or competitiveness help to explain one’s ability at predicting election outcomes? Second, does one’s level of sophistication moderate the link between political preferences and forecasting abilities? To answer these questions, I mobilize data from seven elections taken at the district and (sub)national levels. I also introduce a new measure of forecasting ability—the cumulative Brier score index. In most cases, variables other than preferences and knowledge have little or no influence on the accuracy of voters’ expectations both at the (sub)national and district levels. Political knowledge is positively associated with citizens’ forecasting abilities; however, it does not appear to moderate the preference–expectation link. This result contradicts findings from existing work and holds important implications for the study of citizen forecasting.     

From polls to votes to seats: Forecasting the 2015 British general election

This paper develops a three-stage method to forecast parliamentary election results from vote preferences in British opinion polls: (1) adjusting and aggregating vote-intentions from different polling organizations; (2) forecasting how public support for parties will change in the period before election day; and (3) translating, through simulations, the forecast of election day vote shares into seat totals while incorporating constituency-level information, including local vote-intention polls. Overall, this approach seeks to combine relevant national, regional and local information, and uncertainty about that information, to better reflect the fragmentation and diversity of political contexts found in the new era of five/six-party British politics.     

Nowcasting V. polling: The 2010 UK election trials

This paper first offers a final forecast for the May 2010 UK general election based on our “two-step model” [Nadeau, R., Lewis-Beck, M.S., Bélanger, É., 2009. Election forecasting in the United Kingdom: a two-step model. Journal of Elections, Public Opinion and Parties 19, 333-358.]. That model is then used to explore a new technique, “nowcasting.” We examine our model “nowcasts,” comparing it to the classic forecasting strategy of the vote intention poll, across the contemporary election period, 2005-2010. As is shown, nowcasting offers forecasting advantages that simple polling does not. Most notably, the nowcast provides variation that appears much more subject to substantive explanation of the electoral cycle, and it provides predictions with a three-month lead.