Why do Texas candidates run such negative ads?

Because they work.

We (voters) make them work.  At least, that’s what campaigns believe. And, we’ve given them reasons to believe.

Recently, the Texas Tribune offered evidence on the effectiveness of attack ads in a poll done February 1-7, 2010. My thanks to the Tribune for doing this test and sharing so many of the detailed results online. There is an 18 page summary and 225 pages of crosstabs linked to the story (“Accentuate the Negative“).

Respondents who said they planned on voting for Senator Hutchison, Governor Perry, or were undecided in the GOP primary were divided into two groups at random. One group was shown a Hutchison ad attacking Perry and the other group saw a Perry ad attacking Hutchison. Before and after seeing the ad the respondents were asked to rate their likelihood of voting for Hutchison on a 100-point scale. It’s the kind of classic experimental design that we’d love to be able to do more often.

Voters who saw the “Bailout” ad attacking Hutchison averaged 37.06 before seeing the ad and 31.63 after.

Voters how saw the “Cha-Ching” ad attacking Perry averaged 35.98 before seeing the ad and 40.43 after.

Several cautionary notes about these results:

  • This method only measures the impact of the ad immediately after viewing. Its seems certain that the impact of the ad diminishes over time as voters forget the ad and/or get new information about the campaign. This means we’re probably measuring an ad at its peak impact and it doesn’t tell us how much impact the ad had on the voter’s final choice on election day.
  • The results were “statistically significant.” This means that the impact can be distinguished from zero. This doesn’t mean that the impact is large. In fact, the researchers note that the impact is small.
  • The 100-point scale may produce some weird results. I’m not faulting the choice of the researchers. Any time you measure citizens opinions you’re going to get some kind of mess. There’s a lot of room between 0 and 100 and how voters locate themselves between those two extremes may vary greatly from person to person (and even from moment to moment). When you ask people to pull numbers out of the air, you need to be careful about evaluating their meaning. If I were the “average” Republican voter and my response shifted from 36 to 40 after seeing the Perry ad–what would that mean in terms of actual voting behavior?

There’s a vigorous debate in political science about the effectiveness of negative ads. However, most campaign consults are absolutely sure that attack ads work and voters aren’t giving them much reason to doubt that.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: