Looking to the Future: How Political Rhetoric Shapes Policy Opinions

Download data and study materials from OSF

Principal investigator:

Jennifer Jerit

Stony Brook University

Email: jennifer.jerit@stonybrook.edu

Homepage: http://www.jjerit.com

Sample size: 1154

Field period: 12/27/2006-01/04/2007


When political actors debate the merits of a public policy, they often focus on the consequences of a bill or legislative proposal, with supporters and opponents making stark but contradictory predictions about the future. Building upon the framing literature, this project examines how rhetoric about a policy’s consequences influences public opinion. I show that predictive appeals work largely by altering people’s beliefs about the impact of a policy. Following in the tradition of recent framing research, this study also examines how opinions are influenced when people are exposed to opposing predictions. The analysis focuses on two strategies that are common in real world debates—the direct rebuttal (in which an initial appeal is challenged by a statement making the opposite prediction) and the alternate frame (which counters an initial appeal by shifting the focus to some other consequence). There are important differences in the effectiveness of these two strategies—a finding that has implications for the study of competitive framing and the policy making process more generally.


H1: People’s opinions will shift in response to predictive appeals about the consequences of a specific policy proposal. In the context of this study, being exposed to an argument about losing (gaining) money with private accounts should decrease (increase) support for Social Security privatization.
H2: I expect that predictive appeals will affect people’s policy opinions by influencing their beliefs about the likelihood of various outcomes. Thus, rhetoric about a policy’s consequences affects people’s beliefs which in turn influence policy opinions.
H3: Individuals will view the prospect of future losses (i.e., losing money with private accounts) as more likely than future gains (i.e., gaining money).
H4: Direct rebuttals will not influence opinion when people believe a particular outcome (e.g., losing or gaining money) is likely to occur.
H5: When individuals do not have firm beliefs about the likelihood of an event occurring, either strategy (direct rebuttal or alternate frame) should influence opinion.

Experimental Manipulations

In this experiment, respondents were randomly assigned to one of seven conditions (six experimental conditions plus a control group). Everyone received a common introduction, which stated: “A proposal has been made that would allow people to invest some of their Social Security taxes in the stock market.” Individuals in the treatment conditions then read either one or two arguments pertaining to the issue before answering a question about their support for private accounts. There are three basic framing conditions in this experiment: two competitive framing scenarios (featuring either a direct rebuttal or an alternate frame), and a one-sided framing condition. For each framing condition, half the respondents were asked about the likelihood of losing money with private accounts and the other half were asked about gaining money. See Figure 1 in the study codebook for a schematic of the design.


The key dependent variable is support for Social Security private accounts, which is a five point measure ranging from strongly support to strongly oppose.

Summary of Results

This study shows that predictive appeals can be understood in much the same way that we think of conventional framing effects. That is, predictive appeals affect opinions by shaping people’s beliefs about the impact of specific policies. Consistent with prospect theory, people appear to overweight information about losses, and have higher expectations for negative, as opposed to positive, outcomes. This asymmetry has important implications for rhetorical strategy, as I show in the analysis of the competitive framing conditions. When belief likelihood for a particular outcome is high, certain rhetorical strategies (e.g., the direct rebuttal) have no effect on public opinion.


This study shows that predictive appeals can have a significant effect on public opinion, and this raises a host of normative questions about the health of our political system. More to the point, can ordinary citizens, whose collective preferences often play an important role in the formation of public policy, participate meaningfully in the decision making process when predictive appeals abound? Claims about the future are not easily verified. Moreover, by the time predictive appeals can be proven right or wrong, the responsible parties may no longer be in power. In this way, predictive appeals have the potential to obscure democratic accountability.

Additional Information

This TESS module was combined with a similar survey experiment administered in the 2006 Cooperative Congressional Election Study (CCES). The survey experiments are analyzed together in the paper noted under “References.”


“Political Rhetoric and Policy Opinions.” Paper presented at the 2008 annual meeting of the American Political Science Association, Boston, August 28-31.

Jerit, Jennifer. 2009. How Predictive Appeals Affect Policy Opinions. American Journal of Political Science. 53:411–426