Skip to main content

Do Campaigns Ads Really Work? The Effectiveness of Targeted Ads in the 2008 Presidential Election: A Controlled Experiment

Over the last two decades, campaigns have increasingly adopted micro-targeting strategies in relation to the Latino community as opposed to only relying on general campaign mobilization efforts. Will the ethnically targeted campaigns of the 2008 election reap the desired reward of wooing the now largest minority group, Latinos? Will Latino voters respond to ethnically targeted messages (and messengers) or are these messages redundant? If so, what is the nature of this response? A broader and more strategic consideration is how does an ethnically angled strategy figure into a larger campaign strategy? Is there a ~Sricochet effect~T among non-Latinos that results from the implementation of an ethnically angled campaign? How do non-Latinos respond to Latino messages and messengers?

In September 2008 we produced 8 campaign commercials, 4 for Obama and 4 for McCain, to test whether or not different campaign appeals work for different types of voters. Using a controlled survey experiment, registered voters in California visited a website and were randomly assigned to view one campaign ad, and then answer a series of questions about the candidates and issues in the election. While the 8 commercials were all similar, we varied things such as an endorser, and the language used in the ad. Half the sample saw a Latino endorser and half saw an Anglo endorser shaking hands with each presidential candidate. Half the sample saw an ad only in English, and half saw an ad that sprinkled in words in Spanish.