ok, maybe this is the first post that really is directly relevant to the title I chose for this blog: I just love a quote from the paper I am currently reading (and will be continuing to read, because it is rather long) by Daniel Cervone and others:
If by “traits” we are referring to between-person classifications of inter-individual differences, and if by “operate” we mean to exert influence within the psychological dynamics of the individual, then the answer is “They don’t operate.” … What grounds are there, then, for positing a conscientiousness operating in your head? (p. 344)
To put it into the context of this blog: I am — like Cervone et al. — rather sceptical about the idea that personality traits like conscientiousness and the like have anything to do with mechanisms operating in the brain; however, the Big Five “theorists” have published so many papers stating that their pet constructs are causally operating, culturally universal, genetically determined and so on. I do not deny the relevance of trait descriptions — after all, the empirical evidence really IS overwhelming. However, note that I (and I am surely not the first to do that) spoke of trait descriptions; that is, I believe that a lot of what mainstream personality psychology is concerned with at the moment is finding more or less adequate categories for description, but you can not really use it for explanation.
The point is beautifully illustrated by another excerpt taken from the Cervone et al. paper; they show that if one uses a path diagram approach to signify causal influences, there is something wrong if trait descriptions are inserted:
sportiness → engine size → max. horsepower → max. speed (p. 343)
Such path diagrams are rather common in psychological and other empirical-statistical papers who often apply factor-analytic methods, the most recent spawn of which is structural equation modeling (SEM). In the case above, some thinking makes it clear that the first arrow (from sportiness to engine size) is of a different quality than the others: “sportiness” is a descriptive term (like “extraverted”, “open to experience”, “agreeable” etc.) but surely not in the same way causally relevant as max. horsepower when it comes to determine maximum speed of a car – the developers of the car wanted to have a “sporty” one, which “caused” them to build in a large engine, but that’s a different kind of “causation” than the “causation” of max.horspower by engine size. (BTW, it should be clear that even without “sportiness” the whole diagram is kind of absurd when it comes down to causal determination of a car’s max. speed, but you will find a lot more silly “causal” diagrams in the psychological, sociological and economics literature).
Let me present another example inspired by a paper by Denny Borsboom:
Jane’s intelligence caused her to find the right solution for the problem:
Birds – Air : Fish – (choose one of the following: soil, can, water, salmon, swim)
At first sight, the statement “Jane’s intelligence caused her to find the right solution” may seem reasonable. However, now consider the statement:
Jane’s intelligence caused her not to find the right solution for the problem:
Vögel – Luft : Fische – (wähle eine der folgenden Alternativen: Erdboden, Dose, Wasser, Lachs, Schwimmen)
That’s the same problem as above, only in German. Now, if we assume that the problem is as difficult in English as it is in German and that Jane’s intelligence did not change in between the two trials, it obviously wasn’t her lack of intelligence that caused her not to solve the second problem – it was her lack of German.
The catch is: Even if intelligence might be a rather useful description of differences between individuals, it is not useful as an explanation of why a single individual is able or unable to solve a specific problem. And we can take this last statement to the general: Factor analytic results for interindividual comparisons tells us nothing about intraindividual processes, in general (even if there may be specific cases to the contrary).
Borsboom, D., Mellenbergh, G.J., & Van Heerden, J. (2003). The theoretical status of latent variables. Psychological Review, 110, 203-219.
Daniel Cervone, William G. Shadel, Ronald E. Smith, Marina Fiori (2006) Self-Regulation: Reminders and Suggestions from Personality Science. Applied Psychology, 55, 333–385.