The RAND 2016 Presidential Election Panel Survey (PEPS) uses a unique approach to study the evolution of public opinion, voting intentions, and voter behavior.
Rather than surveying a new cross section of respondents each survey, PEPS contacts the same respondents over the course of the election cycle.
PEPS uses the ongoing RAND American Life Panel (ALP) survey, which allows responses to be linked to previous surveys on voter intentions, opinions, behavior, and life circumstances.
RAND's Presidential Election Panel Survey, like other polls, overpredicted the popular vote. But since it focused on the decisionmaking process and how that translated into behaviors, the data could provide deep insights into what happened and how it took pollsters by surprise.
The ALP consists of a panel of U.S. respondents ages 18 and older. Respondents complete surveys on the web, however, in contrast to most Internet panels, ALP respondents are supplied with Internet access if needed, so the panel represents the entire U.S. adult population.
PEPS includes a subsample of 3,000 ALP members, who are asked about their opinions about political issues in the news, attitudes towards potential candidates, voting intentions, candidate preferences, underlying attitudes towards societal groups, political affiliation, prior voting behavior, and perceived personality traits of candidates and the respondents themselves.
The PEPS methodology has two primary advantages over traditional polling.
First, each new survey can be compared to previous survey responses. This allows us to gain a better understanding of changes in voting intentions and behavior. Not only will we investigate how voting intentions change leading up to the election, but, for many respondents, we will be able to look at past behavior.
Second, we employ a methodology called probabilistic polling. Rather than asking respondents if they intend to vote (yes/no) and for whom, we ask respondents about the probability of voting at all and the probability of voting for each candidate. Our methodology builds on the work of Manski and Delevande (2010) and Gutsche et al. (2014), which demonstrate that the approach can shed greater insight into undecided voters.
Researchers used the RAND Presidential Election Panel Survey to estimate the popular vote in the U.S. election and determine which candidate Americans think will win.
Voters' opinions of both major-party candidates remain low, but their attitudes on two issues have shifted. Support for building a fence along the Mexican border has decreased while support for providing a legal way for illegal immigrants already in the United States to become citizens has increased.
Clinton and Trump supporters disagree on many topics. But they are united in their dissatisfaction with the candidates. Data show that if Americans could vote against the candidates, many would do so.
Data from primary season suggest that Hillary Clinton will likely succeed in uniting her supporters and in-party castaways during the general election. Donald Trump may face a more challenging situation.
RAND's panel survey examines voter attitudes, intentions, and choices, and how these change throughout the 2016 U.S. presidential election. Some sizeable shifts in positions occurred in survey results from December to March.
RAND's panel survey examines voter attitudes, intentions, and choices, and how these change throughout the 2016 U.S. presidential election. What sets this effort apart from public opinion surveys and political polls is that it surveys the same people over the course of the election.
In this Events @ RAND podcast, our panel of experts discusses the latest results of the RAND Presidential Election Panel Survey (PEPS), and what they might portend for Election Day.
In this video Q&A, Michael Pollard discusses the 2016 RAND Presidential Election Panel Survey (PEPS), which is designed to examine voter attitudes, intentions, and choices, and how these change throughout the 2016 presidential election cycle.