The Persuasive Effects of Televised Campaign Ads

James Gimpel
Daron Shaw
Texas, United States
2006 - 2006
Target group:
Outcome of interest:
Electoral participation Voter behavior
Intervention type:
Nudges and reminders

Political campaigns in the United States spend millions of dollars on television advertisements, but there is little rigorous evidence on the effect of this spending on voter preferences. Early in the 2006 campaign for governor of Texas, the launch date and volume of television advertisements and the launch date of radio advertisements for the incumbent candidate were randomly assigned across media markets to evaluate the impact of the ads on public opinion. Results indicate that the televised ads had strong but short-lived effects on voting preferences.

Policy issue

In the United States, political campaigns often spend millions, or even billions, of dollars on media advertising. For example, in the 2012 presidential campaign, the two primary opponents—President Barack Obama and Mitt Romney—spent over two billion dollars in media advertising. While many believe that such advertising makes a difference in voter preferences, there is a dearth of rigorous evidence on how much it does or can change the minds of voters. This study looks at the effects of television and radio advertisements on voter preferences, and examines how long these effects last.

Context of the evaluation

The majority of U.S. states elect their governors every four years. General elections occur in November, before which primary elections are often held to determine which candidates will be on the ballot for each major political party. Primaries are generally held the preceding March. Campaigning for the primary election begins as early as January, at which point candidates frequently launch media advertising campaigns—such as TV and radio ads—to sway the opinions of voters in the upcoming election.

Details of the intervention

To investigate the impact of media advertising on voter preferences and behavior in political campaigns, researchers evaluated the TV and radio advertising efforts made by the Rick Perry campaign in 2006, when he was running for re-election for the Governor of Texas. The main TV ad aimed to link positive images of the governor with voters’ pride in the State of Texas as portrayed through sweeping landscapes and shots of everyday citizens in schools and doctors’ offices. The voiceover by Governor Perry indicated that he was proud to call himself a Texan, and highlighted the State’s recent accomplishments. The monologue ended, “Our people are compassionate. Our vision, bold. Our values, strong. The best is yet to come. I’m proud of Texas. How ’bout you?” The radio ads had a similar tone and message, but included more examples of specific accomplishments of the state under the Governor’s leadership.

Throughout January 2006, leading up to the primary election in March, the Perry campaign launched a three-week advertising campaign across 20 media markets in Texas. Researchers randomized the start date of the television and radio advertising campaigns in 18 of these markets. The volume of television ads was also randomized during the three-week campaign period. First, media markets were placed in groups based on their demographic and socioeconomic make-up. Then, researchers randomly assigned each media market to a start date for a TV advertisement campaign. Within each weekly rollout bracket, researchers randomly assigned the amount of airtime for the ads. In media markets with less weekly airtime, the ads appeared mostly during the morning and evening. When the amount of airtime increased, the ads appeared more frequently during entertainment and late night programming. Researchers did not randomize the stations or programs on which the ads were placed.

Researchers also randomly assigned media markets within each rollout group for the TV advertisements to periods of radio advertisements. All media markets received the same volume of weekly radio airtime and the Perry campaign was given discretion about how to place the ads.

Every day during the three-week media advertising campaign (and for a few days before the campaign launched and after it concluded), voters in each media market were randomly selected to participate in a short survey which asked about their candidate preferences (without mentioning the advertising campaign), and who they thought they would vote for. They were surveyed again five weeks later, just before the primary election.

Results and policy lessons

Results of the evaluation suggest that TV advertising had strong, but short-term effects on voter preferences, while radio advertising had no significant impact on voter preferences. The maximum volume of TV advertising led to a 6 percentage point improvement in Rick Perry’s relative standing amongst voters. However, the impact of TV advertising was found to be strongest during the weeks that the ads aired. On average, Perry saw a 4.73 percentage point greater voter standing on the weeks when ads were aired, but the effect on the following week’s opinion was a negligible -0.17 percentage points. In other words, the impact of the TV advertisements did not extend beyond one week of the end of the advertising campaign, and vanished before the March primary.

To shift opinion 5 percentage points in a population of 14 million people, the size of Texas’ electorate, means persuading 700,000 voters. Exposing voters to the maximum volume of TV advertisements costs approximately US$3 million per week. At less than US$5 per vote, TV advertisements seem like a bargain, but the difficulty is that there may be little to show for one’s money a week or two later.

Gerber, Alan S., James G. Gimpel, Donald Green, and Daron Shaw. 2011. "How Large and Long-lasting Are the Persuasive Effects of Televised Campaign Ads? Results from a Randomized Field Experiment." The American Political Science Review 105 (01): 135-150.