There’s an old joke that half of all advertising spending is wasted; the problem for the marketing department is determining which half. The Madison Avenue ad agencies depend on no one ever being able to figure this puzzle out because their business model would collapse.
A similar controversy has been going on in political science for some time; namely, whether political campaigns (and presidential debates, etc) actually change or affect election outcomes decisively. Political consultants and the broadcast media that get rich every election cycle on campaign ads certainly hope so, otherwise their business model also collapses. One might point to uber-consultant Mike Murphy, who spent $100 million promoting Jeb Bush to no effect. But that’s just one data point. Albeit a large one. I’m sure Murphy can explain why none of it was his fault at all.
Up right now on the Social Science Research Network (SSRN) is an advance copy of an article slated to appear in a forthcoming edition of the American Political Science Review (the premier journal in the field) with the typical dense title, “The Minimal Persuasive Effects of Campaign Contact in General Elections: Evidence from 49 Field Experiments.” One of the co-authors is a very sharp Berkeley graduate student I pass in the hallway sometimes and see at the Wednesday methodology workshops I usually attend. Once you hack your way through the academic nature of the paper—a “meta-analysis” of field experiments studying the persuasion efforts of campaigns—you come to this startling finding:
The best estimate for the persuasive effects of campaign contact and advertising—such as mail, phone calls, and canvassing—on Americans’ candidate choices in general elections is zero. Our best guess for online and television advertising is also zero, but there is less evidence on these modes.
The article offers lots of caveats, especially insisting that “Our argument is not that campaigns do not influence general elections in any way, but that the direct persuasive effects of their voter contact and advertising in general elections are essentially zero.” Candidate quality and other factors may be important. Of course, these are the factors that actually might be more important for political scientists to study, but aren’t very amenable to the quantitative techniques that dominate the discipline right now. So we stick with investigating things that can be infinitely regressed. Sigh.
But it is worth taking note of this limited finding because of the key sentence quoted above—”Our best guess for online and television advertising is also zero”—because if this is right, it suggests the Russian attempts to influence the 2016 election with its various bots and fake Facebook posts likely had no effect whatsoever. Which means liberals will have to come up with some other excuse.