Issue-specific political advertisements are ubiquitous during elections, and they’re often assumed to be effective in increasing voter turnout. Such ads frequently target specific groups — what political scientists sometimes call “issue publics” — such as veterans, parents or senior citizens.

A 2008 study published in The Journal of Politics, “Messages that Mobilize? Issue Publics and the Content of Campaign Advertising,” looks at the effect of issue-specific television ads on the turnout of targeted groups. The authors, based at the George Washington University and the University of Texas at Austin, combined data from the Current Population Survey over the 1998, 2000 and 2002 election cycles with data on the media markets in which the ads aired.

The study’s findings include:

• Targeted campaign ads appear to have only a small measurable effect on groups: “In three election years, we found no consistent evidence that messages related to Social Security and Medicare were associated with higher turnout among seniors or that messages related to veterans were associated with higher turnout among veterans.”
• Groups such as parents did seem to be mobilized by targeted ads, but the effects may be so small as to be extraordinarily expensive to exploit, with diminishing returns. In one media market, it took more than 4,000 ads to make turnout just 6.4% more likely among parents; in a more lightly advertised market, just 322 spots resulted in a 3.8% increased likelihood in turnout. This means that to achieve a further 2.6 percentage points in likely turnout, the “number of newly mobilized parents yields a cost-per-vote of $282. This is roughly 15 times the average cost-per-vote of door-to-door get-out-the-vote efforts.”
• Because targeted ads appear to have limited effectiveness, they do not exacerbate differences in turnout rates between groups and “do not exacerbate existing inequalities,” at the very least. “The participatory tendencies of senior citizens and veterans do not increase when campaigns focus on entitlements and veterans’ benefits, respectively. In fact, the mobilization of parents may imply that, under specific circumstances, campaign messages can have an equalizing effect.”

The study concludes: “The conventional wisdom often portrays the electorate in terms of discrete groups that are defined in terms of generation, gender, ethnicity, occupation, or a quasi-sociological mash-up (e.g., ‘security moms’). These groups are presumed to have distinct sets of interests and to respond to messages centered on those interests, but our results provide a cautionary tale.” The researchers note that the “scholarly literature on campaign effects has produced clear evidence that campaigns can matter … but this does not suggest that campaigns always matter — a point that even campaign effects scholars are careful to note.”