The Most Important Ads of the Campaign Are Only Airing Now

by John Sides on November 1, 2012 · 6 comments

in Campaigns and elections

HuffPo’s Jason Linkins asks:

Dunno if @monkeycageblog takes requests, but I’d love to see some poli sci on when, on the campaign timeline, tv ads are most effective.

Here’s what we know, though admittedly we need to learn a lot more.  I’m cribbing from my old “Moneyball” post on 538.

The short story is that the effects of campaign ads decay.  Quickly.  This isn’t necessarily surprising, if you think of campaign messages as being delivered in “doses” that, like most medicines, wear off.  And this makes sense if you think that undecided voters are engaging, at least in part, in what is called “online processing“: updating their assessments of the candidates on the fly in response to new information, but not necessarily remembering the specifics of that information and thus making it possible for the next “dose” to affect them as well.

So how fast is the decay?  In the famous randomized experiment during Rick Perry’s 2006 gubernatorial campaign, Gerber, Gimpel, Green, and Shaw found that:

…the current week’s advertising raises Perry’s vote share by 4.73 percentage points per 1,000 GRPs…a week later, the effects of these ads…receded to −0.17 percentage points.

The same was true in a study of the 2000 presidential campaign by Seth Hill, James Lo, Lynn Vavreck, and John Zaller.  From an earlier ungated version of their paper:

Even when the persuasive effect of ads on candidate preference is large, 50 to 75 percent of the effect dissipates within the first week and almost all is gone by the end of the second week.

The rapid decay of advertising effects makes it hard to understand why, as Ryan Lizza describes, the Obama team was confident that their early advertising in the summer of 2012 would be so effective.  And it makes the Romney campaign’s strategy of waiting and spending a lot of money now seem sensible.

But the ultimate effect of these late ads depends on whether one side will have a definitive advantage.  As I noted in my earlier post, advertising effects emerge most clearly when one side can out-spend the other—and by a lot.  There are reports that Romney and his allied super-PACs will outspend Obama by 2-1 in the final week of the campaign.  But up until now, the cash advantage hasn’t translated into an ad advantage: Romney and the super-PACs have been paying higher rates and not necessarily putting their ads in front of more viewers.

So although this last week of advertising appears to be important, and probably the most important, we don’t yet know whether either the Democrats or Republicans will have enough of advantage to shift the polls—and shift them enough to matter—in the battleground states.

 

{ 6 comments }

Chris@PTS November 1, 2012 at 8:41 am

The concern I have always had with the experiment from the Perry campaign was that it was conducted in January, almost 11 months before the election.

jreff November 1, 2012 at 8:42 am

John — do you have data on whether Obama was out-advertised from 10/21 to 10/28 (similar to your report on WaPo from 10/24 re the 10/14 to 10/21 period), or has that data about total impressions not been released yet?

Scott November 1, 2012 at 12:38 pm

There seems to be vote rigging in presidential elections that can be seen by looking at statistics.
Mitt Romney benefited from it in the Republican Primaries of 2012 and John McCain benefited in the presidential election of 2008 (though he still lost).
Fraudulently, a computer program flips a percentage of votes from one candidate to another but that the percentage of votes flipped varies with the size of the precinct. The rules of the fraudulent vote flipping algorithm are:
Very small precincts don’t have any votes flipped.
The percentage of votes that are flipped is small (such as .01%) for small precincts and large (such as 5%) for large precincts with a *gradual* change in percentage “flipped”.
The reason perpetrators don’t flip as many votes in small districts is because a recount will check a random number of precincts and a smaller precinct is *more* likely to be audited because there are more of them. So if fraudulent vote flipping flips 5% of votes from Democrats to Republicans, a random recount of precincts would show a smaller error – such as 2%.
The authors of the paper show that the effect does not happen in data from some counties presumably because the perpetrators did not have access to those tabulating computers and it is not seen in democratic primaries. The authors of the study look at income and poverty rates which are highly correlated with voter choice but do not correlate with precinct size. This rules out more Republicans living in large precincts as being the cause of the anomalous data.
To find the article, do a google on the following words: vote flipping large precincts central tabulator

Jason MacDonald November 1, 2012 at 1:29 pm

John, does online processing really imply ads at the end of the campaign will be more effective than early spending? I’m no political psychologist but, as I recall from the Lodge article, individuals form impressions but forget the specific information about why they formed their impressions. But the impressions remain. So, early spending would leave some impression, presumably having some effect (notwithstanding the Gerber et. al findings)?

UW November 1, 2012 at 6:03 pm

But early advertizing defines the opponent, and by the time the election rolls around, this is taken as given fact.

DrunkWino November 3, 2012 at 5:01 am

The part about Republicans and their supporting SuperPAC’s having to pay more money for their commercials because Democrats bought their ad space in advance when the price was lower just cracks me up. Styling themselves as financial guru’s who know how to make businesses more profitable…yadda yadda yadda, it’s just too funny to see the party that supposedly spends money like a drunken sailor was much more effective by buying those spots early while the financial masters screwed around and now have to pay a lot more.

Comments on this entry are closed.

Previous post:

Next post: