US 2024 -> Models -> Prospective
Issues and Leaders model
- Predicts national popular vote by assessing candidate issue-handling and leadership qualities.
- Highlights decreasing influence of party identification on vote choice, emphasizing rising importance of candidates
- Introduced prior to the 2012 election, strong track record
Model with decision-making implications
The “Issues and Leaders” model, developed by PollyVote member Andreas Graefe, predicts the national popular vote from people’s perceptions of the candidates’ issue-handling competence and their leadership qualities. The model demonstrates that the direct influence of party identification on vote choice decreases over the course of the campaign, whereas issues gain importance. The model has decision-making implications in that it advises candidates to engage in agenda setting and to increase their perceived issue-handling and leadership competence.
Methodology
The model uses two variables, one to capture the candidates’ issue-handling competence and one to capture their expected performance as leaders. Data are collected from polls.
Issues score
The model assumes that voters favor the candidate that they expect to do the better on handling important issues. Three conditions have to be met for an issue to influence vote choice:
- the voter is aware of the issue
- the issue is of some importance to him
- he expects one candidate to do a better job in handling the issue than the candidate.
These conditions have been operationalized by collecting data from two types of polls:
- Issue-salience polls, which ask people which issue they regard as most important, such as ‘‘What do you think is the most important problem facing this country today?’’
- Issue-handling polls, which ask people which candidate / party will do the better job in handling that issue, such as „Who do you trust more to handle the economy, Joe Biden, or Donald Trump?’’
Leadership score
The model further assumes that elections are choices between candidates and that candidates’ leadership evaluations are an important determinant of people’s vote choice. Therefore, the model uses information from polls that ask people who they think would be the better leader, such as: “Regardless of how you intend to vote, who do you think is a stronger leader: Hillary Clinton or Donald Trump?’’
Vote equation
The model’s vote equation reads as:
V= P + bI
+ cL
where I
represents the issues score and L
the leadership score, measured t days before Election Day. The dependent variable V refers to the incumbent’s actual share of the two-party popular vote.
U.S. Presidential elections 2012, 2016, 2020
Past performance
First introduced in 2012, an ex post analysis showed that the model’s ex ante forecasts, calculated three to two months prior to Election Day, were competitive with those from the best of eight established political economy models. Model accuracy substantially improved closer to Election Day. The final Election Eve forecasts missed the actual vote shares by, on average, little more than one percentage point and thus reduced the error of the Gallup pre-election poll by 30%.
One month before the 2016 U.S. presidential election, the model predicted Clinton to win 52.0% of the two-party vote. Clinton ended up with 51.1% of the vote, which means that the forecast missed by only 0.9 percentage points. In 2020, the model forecast published about 2.5 months before the election predicted that Trump would win 48.6% of the two-party vote. Trump ended up winning 47.7% of the vote, again an error of only 0.9 percentage points.