Editor’s note: Today we continue our series on the seven sins of consumer psychology from the presidential address of professor Michel Tuan Pham at the recent conference of the Society for Consumer Psychology. Read the introduction here.
A serious problem in our field is a pervasive confusion between what I call “theories of studies” and “studies of theories.” What I mean by this is that what passes off as an empirical study testing theoretical propositions about genuine consumer behavior phenomena is often no more than a conceptualization of a very narrow phenomenon that is created by the study itself.
The studies that I am referring to go like this:
“Past research has shown that when consumers are hungry, they tend to purchase more when they go to the supermarket. However, suppose that we prime them with either the concept of indulgence or the concept of self-control. We predict that hungry consumers primed with the concept of indulgence will purchase more than consumers who are not hungry. However, hungry consumers primed with the concept of self-control will not purchase more than consumers who are not hungry. This because … “
[You can fill in the blank]. That study might be followed-up with another study going like this: “Now suppose that half the participants are asked to remember a 7-digit number and the other half are asked to remember a 2-digit number. Now we predict a 3-way interaction showing that among participants asked to remember a 2-digit number, the results of the previous study would be replicated; in contrast, among participants asked to remember a 7-digit number, hungry participants would purchase more, regardless of whether they are primed with indulgence or self-control.”
I don’t think that this study has been done yet – and I hope that did not unintentionally describe one of our own studies here. However, you will recognize that this pair of studies has some of the qualities of what often gets published in our main journals today: (a) it leverages previous established findings; (b) it combines them in a way that has not been done before, and (c) it yields predictions that are very plausible. If the researchers are reasonably competent, (d) the studies will be conducted in a way that is free from major methodological confounds; and (e) as a result, the reviewers will have a hard time coming up with a better explanation than the one that authors propose. This is a type of paper that is not easy to reject on methodological grounds. From my own observations, it often does not get rejected on “theoretical” ground either, because the conceptualization seems logical and the data pattern is not open to obvious rival interpretations. The only way that such a paper would get rejected is if some of the reviewers are blunt enough to state that “the research is not interesting,” which most of us are generally reluctant to do, at least openly.
However, let’s have a closer look about what such a pair of studies would actually show. I don’t think that it would tell us anything meaningful and general about genuine consumer behavior. At best, what it would only show is a sound and logical explanation of a very narrow phenomenon that can only happen under the very contrived conditions that the researchers created in the lab. Any conceptual claim for such studies is just a mini-theory (with a lower case t) of the authors’ particular studies. They are the behavioral equivalent of analytical marketing models that are based on very implausible assumptions: the math may be correct, but the results tell us very little beyond the model itself.
As authors and reviewers, we need to pay more attention to this distinction and recognize when a paper consists of genuine studies of consumer relevant theories as opposed to mere “theories of studies.”