Monday, April 30, 2012

More Cranky Monday Blogging

Why, why, why, did the Washington Post give over space to discredited GOP hack Frank Luntz over the weekend?

Luntz, as part of the Post's "five myths" feature, writes about what he calls "five myths about conservative voters." But exactly why should we believe Luntz's version of the truth, which he (for most of the items) pulls from his own polling -- which is well known for producing the answers that he wants. For example: Luntz begins by claiming that conservative voters are not, after all, interested in smaller government -- only in more efficient government. Is that true? I have no idea. What Luntz reports is that the words "'efficient' and 'effective' government clearly beat 'less' and 'smaller' government." But that says nothing at all -- nothing -- about people's real policy preferences, if any. It just tells us what words poll better. I mean, presumably people would also like "good" or "excellent" or "awesome" government, too. So what? There is no policy clash between those who favor effective government and those who prefer ineffective. But there certainly are disputes about the size of government. What do conservatives believe about that -- or about real decisions, such as what should be done about military spending or specific programs? No hint of that from Luntz.

Basically, there's nothing much to take away from his piece. That's true when I suspect he's mostly wrong (in the abstract, I'm pretty sure that self-identified conservatives do in fact prefer smaller government, although that changes when you move to specific programs). It's also true when he's probably right -- as in his claim that conservatives oppose slashing Social Security and Medicare. Granted, his argument there is a bit strange. Is it a myth that conservative voters want to slash these programs, and that "This charge is at the heart of the Democrats’ campaign against the GOP"? Uh, no. What's at the heart of the Democrats' campaign is that Republican politicians plan to slash Medicare. Which is, you know, true. Indeed, the reason that Democrats are campaigning on it is because Democrats believe that it's an issue on which Republican voters do not agree with the plans of Republican politicians. Which is why Luntz advises Republicans to talk about making these programs "work," instead of admitting that they're cutting spending on Medicare.

The problem is that playing with words to find out whether ones test best just doesn't tell us anything interesting, whether it's done by Luntz for the Republicans or his equivalents for the Democrats. Oh, it proves that most voters can be easily manipulated into giving the polling result that the pollster wants to produce -- but that doesn't mean that they can be easily manipulated into actually changing their minds about policy, or that they can therefore be easily manipulated into changing who they vote for. Both parties are suckers when they pay to get this kind of advice, and the WaPo is a sucker to run it.


  1. Certainly "playing with words" won't change voters' underlying policy preferences. No matter how much the GOP says they want to "fix" Medicare and make Social Security "work" better, voters will be angry if the "fix" just turns out to be massive budget cuts. The voters' initial preference - to maintain entitlement spending - would not be affected at all by the semantic games.

    But can Luntz's games manipulate voters "into changing who they vote for"? I don't see why not. A voter who hears Romney saying he wants to "fix" Medicare might think that Romney just wants to tweak reimbursement rates, or maybe he even wants to expand the program! And those would be reasonable beliefs, given how nebulous a word like "fix" is.

    The whole point of Luntz's enterprise, it seems to me, is to fool voters into believing that his candidate-client shares the same values and priorities as they do. That won't work once a pol takes office and actually starts to take action. But it sure might work to get the pol into office in the first place!

  2. Which is, perhaps, why they spend millions of dollars producing and airing commercials which are largely based on using Luntz's polll-selected messaging tips and misleading formulations, combined with a whole host of additional emotion-based messaging tricks (such as music and photo-morphing games), that have long been tested scientifically by advertising companies, as their primary form of campaigning and have completely moved away from calm, rational discussions of policy preferences as a form of campaigning.

  3. This comment has been removed by the author.

  4. Tested scientifically? I suppose it's possible, but most of what I read about advertising, political and otherwise, suggests that's not true.

    Anyway: I'm entirely unconvinced that using Luntz's magic words about policy moves any voters at all. If someone has evidence, I'll be glad to look at it, but I think it's all worthless hokum.

    1. So all their advertising money is wasted? It never has any effect at all?

      I have work to do, I'm not going to stop to search for all the advertising industry literature that talks about testing techniques and verifying techniques, I'm pretty sure I found a bunch of it when I was studying advertising 10 years ago when I was contemplating a career change. Where did the focus group come from? an attempt to test advertising techniques.

      No advertising ever has any effect, am I understanding your position correctly? One of my jobs is at America's largest supermarket chain, a lot of customers come in about every day carrying coupons and looking for advertised sales prices. You don't think there's any evidence that playing creepy music associated with an image produces different emotions than triumphant music? So horror movie producers are all wet too?

    2. I don't have any particular knowledge of the evidence, if any, for or against the effectiveness of political ads. I just have always found the notion that anyone would be persuaded by these ads incredible. It's not the same as commercial ads, which often can be persuasive, or at the very least make you aware of the product and make it stick in your mind. But nobody needs ads to learn who's running for office, and it seems to me you'd have to be a total moron to base your vote on one of these ads.

    3. And in fact, ads (and campaigns in general) in low-information races can make a difference, not only by boosting name recognition but also by informing people about something that they will then like about one candidate or dislike about the other -- which turns out to have a strong effect on House voting (along with, of course, party). But that's not about using words carefully to change peoples' minds about an issue and subsequently get them to change vote choice.

    4. that's not about using words carefully to change peoples' minds about an issue and subsequently get them to change vote choice

      True, but you don't need to change people's minds about an issue in order to get them to change their vote.

      All you need to do is change their perception about what the candidates' positions are.

      Example: John Q. Senior of Ohio is a Medicare and Social Security recipient. He is terrified that these programs will be discontinued or dramatically cut in the name of deficit reduction. Romney runs an ad saying he's going to "fix" these programs. Mr. Senior likes the sound of that - because it sounds like Romney intends to preserve the programs the way they are now.

      So the ad was successful - not in changing Mr. Senior's position on preserving Medicare and Social Security, but in creating a perception that Romney is closer to Mr. Senior's position than he in fact is. And it's all due to the creative (and mendacious, to be sure) use of words.

    5. I just have always found the notion that anyone would be persuaded by these ads incredible. ... you'd have to be a total moron to base your vote on one of these ads.

      I don't see how the one follows from the other. Surely you don't doubt the American electorate's capacity to act like morons?

  5. Since conservative voters tend to be misinformed as much as informed - often claiming their candidate has a position 180 from the stated position of the candidate - I don't think I'd put much on polls of conservative voters.

  6. Of course, this is probably why Republicans love term limits and dark money and hate being compared to Republicans in office.

    I've long since lost count of the number of times I pointed to a vote or position of Republicans in Congress out to a Republican voter who then denied a connection with those views.

  7. Oh Luntz, the hack that keeps on hacking. I came across a great example of the utter folly of trying to take this sort of focus group/polling politics to its logical extreme. It's a This American Life episode about how after 9/11 the Bush Administration tried to use Madison Avenue advertising strategies to "sell" America to the Muslim world and the total train wreck/hilarity that resulted. It's told by one of the guys who was tasked with trying to create commercials to air in the Muslim world and has a great anecdote about how focus groups can be totally pointless when dealing with complex problems like Foreign Relations or the historical interactions of Christianity, Islam and Judaism. It's the 2nd act:


Note: Only a member of this blog may post a comment.