Framing immigration and the sip taste test fallacy

Post by Vipul Naik (regular blogger and site founder, launched site and started blogging March 2012). See:

I’ve been reading through the memos on immigration that have been put out by the Frameworks Institute, which describes itself as “changing the public conversation about social problems” and, from what I can make out, takes broadly progressive positions on various issues (or rather, offers framing advice geared towards progressive causes). (Thanks to Alex Nowrasteh for sending me the link for these memos). These memos were also briefly referenced in Fabio Rojas’s post open borders: what to do about it (part 2). The memos are targeted mostly to a US audience.

Let me begin by pointing out what parts of these memos I find intuitively plausible. First, I think that they’re largely on target with respect to their key suggestions to pro-immigration advocates: put emphasis on prosperity, opportunity, and fairness, don’t activate zero sum thinking, and avoid blaming people or calling them bad names. There are also suggestions — somewhat morally dubious but probably strategically valid — to tailor the message based on the race and ethnicity of the people receiving the message. When talking to whites, for instance, the memo recommends the value of “Fairness between Places.” When talking to blacks, it recommends the value of “Fairness between Groups.” When talking to Latinos, it recommends emphasis on the “value of opportunity.” The morality of offering different messages to different audiences seems questionable to me, but I don’t doubt the research they highlight to show that different framings appeal differently to different racial and ethnic groups on average.

The part where I’m not on board with the Frameworks Institute, however, is when it comes to their Don’t list. For instance, they suggest:

  • Don’t talk about immigration as a legal issue, or mention illegal immigrants.
  • Don’t evoke the Crisis frame by talking about how widely and deeply broken the immigration system is.
  • Don’t talk about securing borders or preventing problems.
  • Don’t begin the conversation by focusing on good immigrants, because that brings to mind all the bad immigrants.

I’m quite skeptical of all these suggestions. The way the Frameworks Institute comes up with these suggestions is by doing controlled experiments — expose people to different framings, see how they respond to each one within the setting. For instance, two people may be presented with otherwise identical paragraphs one of which has a phrase evoking a certain frame, and the other one without that phrase. The audience reactions are then compared.

The problem here, however, is that the research methodology seems to focus on people’s instantaneous responses, rather than their considered responses after reflection. In particular, it doesn’t seem to account for how people would respond after they had time to hear opposing viewpoints. In some ways, it reminds me of Malcolm Gladwell’s observations in his book Blink regarding sip taste tests (quote copied from here):

The Studio “Sip Test”

In his 2005 book Blink, author Malcolm Gladwell explored this phenomenon with the case of the “Pepsi Challenge”, a series of soft drink taste-tests that seemed to conclusively prove that Americans plain-old liked Pepsi better than Coke.

These soda shootouts made Pepsi a serious contender for the first time, and left Coca-Cola hemorrhaging market-share. Gladwell however, presents evidence that Pepsi’s overwhelming success over Coca-Cola in these tests was not evidence of a real preference, but rather a result of the flawed nature of the “sip test” method itself.

His research shows that when offered a quick sip, tasters generally prefer the sweeter of two beverages – even if they prefer a less sweet beverage over the course of an entire can. Just because a taster prefers a single sip of the sweeter beverage, Gladwell argues, doesn’t mean he’d prefer to have an entire case of it at home.

Coca-Cola found this out the hard way when they introduced “New Coke”, a soft drink completely redesigned to match Pepsi’s success in the sip test. The results were catastrophic.

It’s a lesson that shouldn’t be lost on the studio world. Where soda drinkers tend to prefer the sweeter of two beverages in a brief sip-test, listeners prefer the louder or more scooped of two sources in quick, non-contextual listening tests. Bass, treble, and volume are like sugar to our ears.

Now, I don’t know enough about taste to know whether Gladwell is right, though I think there’s probably some truth to the idea that sugar tastes a lot better in the short run compared to how good it makes you feel in the somewhat longer run. My personal experience attests to the observation about sound: louder sounds seem nicer and more prominent in the short run, but if you exclusively focus on the short run prominence and catchiness to the ear, the overall effect is worse (see for instance here). In fact, one of the tips offered to people doing dynamic range compression (a technique used to increase the loudness of sound) is that if the sound appeared to get noticeably louder (and apparently better) you probably did too much. In my first attempts at dynamic range compression, I inevitably over-compressed, and it was only after re-hearing the samples a few times that I started developing an antipathy to over-compressed sounds.

I think that much the same applies to framing — there are probably some framing elements that sound good in the short run, when people hear them, and are asked to respond immediately. But those same framing elements can prove to be a liability once people have had time to reflect on them, and are exposed to opposing viewpoints on the matter.

If you don’t find the sweetness/loudness metaphor convincing, here’s a more economics-type take. In a monopoly, the producer can afford to choose a profit-maximizing price. In a competitive market, on the other hand, the producer needs to basically operate at the market price, give or take a bit (depending on the extent of product differentiation and the extent of competition). A producer in a competitive market who thinks he is operating a monopoly would probably end up choosing a tragically high price where nobody buys his wares. In the same way, an immigration framer who is operating in a monopoly world (where there are no others competing to influence people’s views on immigration) can feel free to ignore all the issues that cause negative connotations associated with immigration. An immigration framer in a world where there are many others competing to influence people’s views on immigration, however, cannot afford this luxury. An immigration framer who believes that she is operating in a monopoly world, but who is in fact operating in a world with many others trying to influence views on immigration, will end up looking completely stupid and out-argued.

Analogies and metaphors are fine, but do they shed any light on immigration-specific issues? I think they are quite relevant to immigration-specific issues. Concretely, I think that immigration advocates generally appear to avoid a lot of the “hard questions” raised by restrictionists. While I don’t think that the Frameworks Institute memo per se is to blame (I don’t know if that many people have read it) I think that the “don’t” subset of their advice has in fact become to quite an extent common wisdom among pro-immigration circles, and it has hurt the cause of open borders and even of modest pro-immigration advocacy. Let’s go back to some of the things that the Frameworks Institute memo suggests one should not do:

  • Don’t talk about immigration as a legal issue, or mention illegal immigrants: This is a luxury that open borders advocates, or other pro-immigration folks, simply don’t have, because restrictionists keep bringing this issue up again and again. In fact, avoiding talk of the legal versus illegal distinction looks really really bad. And this isn’t limited to niche blogs. In mainstream newspapers and magazines, a pro-immigration piece that fails to mention the legal/illegal immigration issue will, in a few minutes of being published, get unfavorable anti-immigration comments with ILLEGAL in all caps mentioned in several places. Even if the average reader wouldn’t raise the legal versus illegal objection, if he/she reads the article more than a few minutes after it’s published, he/she will be exposed to the legal/illegal distinction from the other side. Immigration advocates simply don’t come out looking good when they deliberately try to sidestep the issue of legality. The irony, of course, is that most of the legal versus illegal arguments can be dealt with in a reasonable manner — they’re not the most challenging or cutting-edge arguments against immigration (see our page and blog posts about this distinction). But they do need to be addressed head on, rather than swept under the rug.
  • Don’t evoke the Crisis frame by talking about how widely and deeply broken the immigration system is: While I agree that it’s not a good idea to proactively invoke the Crisis frame, restrictionists again have the upper hand here. They’ve already made the statement that the immigration system is broken and needs fixing. At this stage, one can either agree (and redefine “fixing” to be the opposite of what restrictionists mean), or disagree (and thus undermine one’s own advocacy efforts) or just ignore the point. That said, I think the best thing to do is just ignore, rather than actively endorse or challenge, the “broken system that needs fixing” language.
  • Don’t talk about securing borders or preventing problems: While I agree that securing borders is not the biggest or most important issue for serious restrictionists (who are more concerned about self-deportation) it is an issue that is probably more on the minds of ordinary people. Being blithe about the issue or failing to address it makes it easy for restrictionists to portray one as being supportive of terrorists, hostile armies, diseased huddled masses, and barbarians invading the country. But it’s the “don’t talk about preventing problems” suggestion that I find most puzzling. It’s almost like a producer in a competitive market who is selling at far higher than the market price being advised by a consultant — “don’t think about lowering prices, because this is your monopolistic profit-maximizing price — I did the MR = MC calculations.” Again, even ordinary people who cannot come up with a list of problems with immigration on their own can easily tune in to their TV, radio, or favorite website to get the list, which means that within a few hours or days of a successful framing pitch, people return to their original anti-immigration position with a vengeance, and feeling that you (as the framer of a temporarily successful pro-immigration pitch) duped them with your smooth talk.
  • Don’t begin the conversation by focusing on good immigrants, because that brings to mind all the bad immigrants: I’d have to say I agree here. Still, it makes sense to talk in the conversation about both good and bad immigrants. If people can be reminded of bad immigrants by talk of good immigrants, they can probably also be reminded of bad immigrants the next time they watch TV. A framing pitch that doesn’t proactively address arguments people might expose themselves to tomorrow is falling prey to the sip test fallacy of maximizing persuasiveness in the very short run at the risk of being completely discredited tomorrow.

Over and above these strategic objections, I have another, more meta-level moral objection. Talk of framing and communication presupposes that we already have the right answer, which needs to be spoonfed to a benighted public that, due to ignorance or apathy, continues to embrace the wrong views. This may be true in specific instances, but to embed that as one’s core assumption, and to shortcut the standard processes of intellectual rigor and thoroughness for short-run marketing success seems to me to be profoundly wrong. People in the US, as elsewhere, are very skeptical of the elites, and part of the reason is elites’ habit of talking down to them. Even if the elites are correct in their assessment about the public’s level of knowledge, it seems to me that being more intellectually rigorous and charitable to opposing arguments is a better way to convince people than patent attempts to sweep opposing arguments under the rug and pretend they never existed. Elites who say, “We’re right, the masses are wrong, but it would take too long to explain to the masses just why objections A, B, and C are wrong, so we’ll just avoid talking about them and hope that the masses won’t notice” are doing the wrong thing both strategically and morally.

2 thoughts on “Framing immigration and the sip taste test fallacy”

  1. Well said, though in response to the idea that framing is too paternalistic, while I agree, university teaching has made me more sympathetic to… or maybe, more inclined to fear the quixoticness of my aversion to… a more one-sided approved-sources-only approach to teaching. I’m horrified by how inclined students are to believe everything they read. To provoke critical thinking is like pulling teeth. But there’s certainly not much chance of getting anyone to support open borders *uncritically* in our society, when the public overwhelmingly opposes it.

    1. Nathan, I don’t think that framing per se is paternalistic. My objection is more to deciding framings based on their short run appeal. I think that framings need to be chosen that are intellectually accurate and can withstand scrutiny and opposing arguments. Research that looks at the immediate effect of a framing on people’s opinions is therefore, in my view, potentially misleading if used to draw implications about the best framing.

Leave a Reply