Bryan Caplan  

Mainstream Marketing vs. the Central Six

PRINT
Audit the Fed, or End It?... Sumner's One-Sentence Class Au...
After faulting economists for ignoring marketing, Geoffrey Miller harshly attacks mainstream marketing for ignoring what he calls the "Central Six" - IQ plus the Big Five personality traits.  At first, you might think he's merely telling marketers, "Here's something useful you've overlooked":
Surprisingly, most marketers have no idea how well the Central Six can predict consumer behavior.  The typical consumer behavior textbook includes a large section devoted to individual differences, but no discussion of general intelligence and the Big Five factors.  Rather, the focus is on diverse "factors" that may influence consumer decision making: wealth, time, knowledge, attitudes, values, self-concepts, and motivations.  The fact that the Central Six efficiently predict individual variation across all these factors remains unknown or ignored... Marketers likewise pay attention to "demographic variables" - age, sex, ethnicity, socioeconomic status - without taking into account their correlations with the Central Six.
But on closer look, Miller position seems to be, "Knowledge of the Central Six makes mainstream marketing obsolete":
So, most current research on marketing and consumer behavior relies on a chaotic grab bag of outdated theories and unreliable findings.  The potent effects of general intelligence are hidden behind its causal effects, empirical correlates, and politically correct euphemisms: education, class, socioeconomic status, consumer knowledge, "cognitive resources," and "cultural capital." Often, marketers think they are studying the effects of class, race, or religion on consumer decision making when they are actually studying the effects of intelligence, which shows different average scores, for whatever reasons, across different classes, races, and religions.  The potent effects of the Big Five are likewise hidden behind their correlates and euphemisms: attitudes, motivations, self-concept, values, lifestyle, culture.  Here, marketers think they are studying the effects of sex, age, or political beliefs, when they are actually studying the effects of openness, agreeableness, stability, or extraversion, which also show different average scores across males and females, young and old, liberals and conservatives.
Miller makes little effort to charitably explain the mainstream marketing position.  But even if we take his caricature as fact, mainstream marketing seems far more predictively useful than the Central Six.  Consider a few random examples:

Hannah Montana.  Mainstream marketers would presumably say that age and sex are the key predictors of interest.  What do the Central Six have to add?  Maybe a little - I suspect that tween girls higher in IQ and Openness would spend fewer dollars and hours on Hannah Montana.  But how many 38-year-old men of any IQ and personality type voluntarily watch her show or listen to her music?

Strollers.  Mainstream marketers with their "chaotic grab bag of outdated theories and unreliable findings" might foolishly suggest that people buy strollers if and only if they have very young children.  What do the Central Six have to add?  I guess Miller might respond that high IQ customers buy Peg-Peregos (I finished assembling one today, that's the ticket!).  Still, only mainstream marketing explains why people are in the market for a stroller in the first place.

iPods.  Mainstream marketers would probably claim that age matters a lot.  Maybe the Central Six could add in IQ and Openness, but they'd probably be marginal.  If you're 80 years old, you probably don't own an iPod, even if your IQ and Openness are in the 95th percentile.

Morton's.  Mainstream marketers would say that people who eat at Morton's have high income.  What's Miller alternative?  That Morton's is really a restaurant for people with high IQs?  Come on - there are a lot more people of average IQ and high income at Morton's than people with average income and high IQ.

Even Miller's favorite example - bumper stickers - doesn't have his back.  Sure, there are lots of IQ and personality-themed bumper stickers.  But aren't there far more political bumper stickers?  Miller might object that political ideologies are mere proxies for personality.  But when you look at the empirics, that's just not true.  Personality tells you something about ideology, but most of the variation remains unexplained.  Bottom line: If a car has an Obama bumper sticker, you know the driver's party with at least 90% probability.  But even if I gave you a .5 SD margin of error, how confidently could you predict the driver's IQ or personality?

I suppose it's conceivable that I'm cherry picking my examples to make Miller lose the Mainstream Marketing - Central Six Cage Match.  So here's my challenge: Pick a random object in plain sight.  (Anything displayed in an open browser window counts).  What would mainstream marketing say about this object?  What would Miller say?  Which approach explains more about consumer behavior?


Comments and Sharing





COMMENTS (6 to date)
fundamentalist writes:

Miller may be right, but so what? How can a marketeer use them? He can't! How do you identify people with the central six traits? You have to survey them first.

Marketeers need easily identifiable traits, such as gender, race or economic status, so they can target those groups. But the central six traits are scattered among all of the easily identifiable groups and with the zip codes or radio/tv stations they listen to. The Census Bureau doesn't do the central six and the media don't have data on their consumers about the central six.

hacs writes:

The problem with the "central six" is beyond the R2 or even the deviance, there are traits of personality are not apprehended by the big five model, biasing the results, and the introduction of the IQ is not a solution.

In an article about the Dark Triad, the big five model cannot account to the commonalities as well as the particularities from those behaviors. In others is showed the big five model is not robust to different cultural backgrounds (in a heterogeneous country as USA, that is a big fault).

In an extreme situation, "scientific" misconclusions could lead the USA to a new progressive era (were its followers ignorants? four names: Karl Pearson, Ronald Fisher, Irving Fisher and James Watson) and its racialist-eugenics policies (is, at maximum, 0.1% of difference in human genome a potential explanation to the observed social and economic differences?), inclusive on immigration.

Hayek's asserted the frustration of the specialist is the major force driving the government interventionism. He continues to be a light in the prow.

David writes:

If you want an example of how marketeers can benefit heavily from use of the six factor model, I have one.
Only @60% of the population in the US is extraverted. If you want to capture the introverted market, particularly if you're selling items of disproportionate interest to an introverted population, you only have to do one thing, relative to your competitors. Don't annoy them. Costco is a good example of an introvert-friendly establishment. Best Buy, on the other hand, most certainly is not.

Walt French writes:

Not unrelated: statisticians — good econometricians, especially — find the notion of “data mining” abhorrent, for the irreproducible and random results that necessarily result, whereas it's the term of art for most marketing trawls through databases.

If you threw in these additional 6 RHS explanatories, you'd expect, given the typical appreciation for “independence” and “significance” in a lot of “applied” (not peer-reviewed) work I have the pleasure of seeing, a multiplication of useless results on how small-town 13–19 year-olds who are both extroverted and closed, are more 30% more likely to buy iPods than their introverted, open peers, after seeing a TV ad for it.

Not that I'm aware of advertising channels for e.g., introvert, disagreeable types within a given geography, age group and family status. Even the Great Google might be challenged on targeting ads that way. Modern Married Junior Skinhead magazine, anybody?

I'm not hostile to the psychometrics or whatever, but by the evidence cited, it wouldn't make the Occam cut.

Steve Sailer writes:

Lots of big money cultural products can be arrayed on IQ scales.

Jay Leno tends to aim for maybe a 100 IQ audience while David Letterman aims for 105. It's not a big difference, but it is noticeable.

Malcolm Gladwell's books are bestsellers because they appeal to the bookbuying sweetspot between, say, 105 and 130 IQ. Extremely smart people, like Judge Posner, consider Gladwell to be a bit of a bozo, but they are too few in numbers to dent his bank account.(Gladwell, himself, is very high in Openness to new ideas, which is his strength as a writer.)

Tom Stoppard's plays, in contrast, probably start around 120 to 130 IQ and go up. For all their expert showbiz razz-ma-tazz, they're simply too fast and complex for the great majority of people to follow. I, for example, always try to read a Stoppard play before seeing it. I gave up taking people to see Stoppard plays because they are too difficult. Heck, I found his quantum mechanics play "Hapgood" impenetrable even in written form. The success of Stoppard's stage career depends upon the elitism of modern theatre and the fact that his plays are so good that people will give them a try, sometimes more than once.

Coen Bros. movies probably have a floor around 110, while Michael Bay's movies probably top out around that number.

hacs writes:

About the heritability of intelligence: http://www.stanford.edu/~jhj1/cgi-bin/blog/?p=295


"So, in a phrase, sure, genes help determine intelligence. But the action of these genes is so fundamentally tied up in environmental interactions that it seems that the explanatory power of simple genetic models for intelligence and other complex social traits such as political and economic behavior or social network measures is very low indeed. Moreover, the predictive power of these models in changing environments is low. Without explanatory or predictive potential, we are left with something that isn’t really science. I applaud efforts to more deeply understand how productive environments, good schools, and healthy decisions can maximize human potential. Heritability studies of IQ (and I worry about these other fashionable areas) seem to provide an excuse for the inexcusable failure to deal with the fundamental social inequalities that continue to mar our country — and the larger world."

Comments for this entry have been closed
Return to top