Psychology Data and Research

Writings, findings, and breakthroughs in understanding people.

The Unspoken Truth

The study of words. The science of response.


What Marketers Should Know About the Nonconscious Webinar: Q + A

We've compiled the top five questions via email or twitter, and had our VP of Research, Kyle Thomas, respond. Still have more questions? Feel free to reach out 

How does this process differ from previous methods using psychology or psychographic information to create segmentation?

If you’re not familiar with the personality psychology literature our method probably looks pretty similar to previous ones; however, previous methods like the MBTI rely on psychological types, which creates a number of limitations. The way the MBTI works is, you take four of these traits and do a binary split. If you do a binary split on just four traits, you get 16 groups. If we wanted to add another trait, that would be 25, and another would be 36. You get this combinatorial explosion if you’re trying to create these different types and it limits the number of traits you can measure. In our research, we typically use over 20 separate traits and motivations, which would generate hundreds of different types, which is clearly an unmanageable amount. That creates a second problem: if you have all of these different types, it’s not like you can be segmenting them on the fly; you can’t create 16 novel types for every different engagement you have, so you’re basically stuck with these universal types that are supposed to apply over all different product realms. Instead, by using continuous variables - we can use as many as we want - we use econometrics methods to map them onto different things that measure whatever we’re interested in, which makes it more manageable. That way, there is more information because traits are continuously measured, rather than discrete measurements, and we don’t run into the combinatorial issue. We can make different segmentations on the fly because we can just combine whatever’s predicting the actual consumer behavior.


Are psychological traits the same as personality traits?

You’ll notice we’re calling them psychological traits rather than personality traits and that was a very conscientious decision we made. In general, if you mention personality to a psychologist, they’re going to think of the Big Five. Indeed these are very important and in some sense form the most abstract core of individual variation and are extremely important characteristics. However, a trait is just any type of behavioral pattern, belief, motivation, etc. that’s stable over time and across different situations. Anything that has that kind of characteristic can be measured as a trait and assessed, so we are measuring many more things than just the Big 5 personality traits. We measure preferences such as eco-consciousness, concern for diet, and liking unique products as traits, which allow us to infer many more specific motivations. We have also developed techniques to measure things like sense of humor, fashion styles, and music tastes, which can be reliably measured as traits due to their general stability over time. The main point is just that many many more things can be measured as traits than just the Big 5 personality traits, which is really helpful and valuable because it allows for a very broad scope of what this kind of methodology can assess.

Can you share your supporting research on the reconceptualization of traits as an “active description”of chronic motivations?

In some sense, all of our research supports such a reconceptualization because this is a broad kind of theoretical position that isn’t directly falsifiable, and so the criteria one ought to apply to such a claim is whether or not it leads to a productive research program (for any philosophy of science nerds out there, this would be akin to Lakatos’ views on what makes for good science, as opposed to the strict Popperian criterion of falsifiability). So, because this viewpoint has led to a lot of productive research, such a perspective would be vindicated, according to this criterion. Furthermore, while this perspective isn’t itself directly falsifiable, it is the only plausible explanation I can think of for our results. Because traditional conceptions of traits are merely descriptive, they cannot be said to “predict behavior” in any coherent sense, since it is exactly these behavioral patterns that are being described in the first place. Conceptualizing traits as motivations is the only way I can think of to get out of this circularity.

Let me be a bit more specific, and just give you a few examples of our research efforts that have come out of this approach. We have done research on marketing messaging, and found that traits indeed predict what kinds of messages appeal to specific kinds of people. We have done work on motivational segmentation, and found that traits are good predictors of what TV shows people watch, what they read, what websites they visit, where they shop, etc. And, of course, we have done quite a bit of work using traits to understand motivations to specific consumer goods, and have found them to be very predictive of what people buy as well. I also want to add that we’re not the only ones that have productive research coming out of this approach. Sam Gosling and people in his lab have uncovered much of the same kinds of things as we have, and they have done quite a bit of other work on how our traits are reflected in the environments we create for ourselves, such as our living spaces and Facebook pages. And, as I mentioned in their direct test of a specific hypothesis derived from this view, William Fleeson and Kira McCabe found that goal pursuit and motivations predicted 74% of the variance in extraversion scores. This is an astronomical number for psychological research, where we often see effect sizes of less than 15%!

You claim psychological traits are additive in explaining consumer behavior. With that said how meaningful is the contribution?

In our research we have found that traits are anywhere from about as predictive as traditional demographic information up to ten times as predictive. I’ve also seen research by other people with similar findings. The size of the contribution varies a lot depending on the specific product domain, or whatever it is that one wants to look at, but because it’s additive, it means you aren’t forced to just pick either traits or demographics, since they can be combined.

I wanted to note one other thing as well in terms of using traits vs. demographics. While demographics can be used to hone in on a target audience, they don’t give you any further information beyond that. Knowing that middle-aged, urban, educated people are more likely to buy something doesn’t tell you much about why or how to reach out to them. In contrast, knowing that eco-conscious, open, and extraverted people gravitate towards a product not only helps to hone in on a target audience, it also tells you why that audience is interested in the first place, and how you might reach out to them. 

How would you propose we integrate psychological trait data into focus group research efforts?

Psychological trait data is well-suited for adding substantial power to more traditional market research methods, such as focus groups. Imagine the added insight one could gain if they knew the motivational profiles of the people being interviewed. The two types of data could be connected to understand not just what people say, but who says what, and why. It could help understand the motivations people have for being interested in one thing or another, even if these people couldn’t themselves report such motivations. Basically trait data can be connected with any kind of behavioral data, and statistical relationships between traits and specific behaviors are what allow one to infer motivations. This means traits can add significant power to almost any existing market research methodology.


Psychology and Consumer Research Webinar

personality psychology Nonconscious Motivations Research Traits and Scales Data Collection

Posted by Emily Dyess on Apr 22, 2013

Follow Us