What Does the Psychologist Do?

Like a lot of people we don’t just say what we did, but we add to that why we did it. For example: “I scream at you because I am very angry at you” (what is anger?) or “Because I’m very intelligent I was able to get a high score on that test” (what is intelligence?). We use unobserved entities to explain our behaviour. In psychology we are also concerned with what the causes are of human behaviour.

Psychologists measure all kinds of things that have to do with behaviour. But what exactly do they measure? To answer this question I will make a distinction between manifest variables and latent variables. A manifest variable is a variable that a psychologist is able to directly observe like the response someone gives on a questionnaire. A latent variable is unobserved and that makes it very hard to measure it or even worse, to prove the existence of a latent variable. Examples of latent variables are depression, intelligence, motivation, power, and sadness.

Depression, intelligence, motivation, power, and sadness are all examples of what a psychologist tries to measure. You might think that measuring these things is not that hard. If you think about yourself you might say that you know very well if you are depressed, intelligent or motivated. You might even say, “Come here psychologist, I will tell you how depressed, intelligent and motivated I am”. But then the psychologist will answer that such information is not of any use because it is subjective. And if a psychologist is subjective he cannot work as a researcher at the university.

What the psychologist does very often is indirect measurement of latent variables. If (s)he wants to measure latent variables indirectly (s)he needs manifest variables. Why? Because the psychologist thinks that responses from people that are tested on manifest variables are caused by a latent variable. For example: If someone responds to the statement “I don’t sleep very well” on a seven-point Likert scale with a seven, meaning that someone barely sleeps, we believe that this response is caused by a depression. If we have a collection of statements, we believe we can say something “objective” about a depression (a subjectively constructed latent variable fyi). We do this by putting all the data in a computer so that the latent variable can be calculated. By calculating a number for a latent variable it becomes kind of real.

And how do we do that?

Psychologists have a collection of formal models to their disposal to measure latent variables. A formal model is just a lot of mathematics that have nothing to do with psychology or behaviour.

But how can you measure motivation with something that has nothing to do with motivation? I don’t measure length with a weighing scale do I?

What a psychologist does, is dressing the formal model up with theories, theories about depression, motivation or intelligence. Then he will look if the formal model with all the equations fits the clothes with which the psychologist tried to dress him up. If it does the psychologist can say that his theory is approximately true. We can use the same clothing analogy to show one of the shortcomings.

Shopping stores have a limited amount of sizes they sell their clothes in. Often a lot of different people fit the same t-shirt.


Boris Stapel

Recommended reading:

Borsboom, D., Mellenbergh, G. J., & Van Heerden, J. (2003). The theoretical status of latent variables. Psychological review110, 203.

Borsboom, D. (2008). A tour guide to the latent realm. Measurement6, 134-146.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>