The Meaninglessnes of the Mean

I had an argument about research methods with a psychologist a while ago. Nothing unusual there, eh? While most of what went on was fairly standard stuff, there is one thing that struck me. It seemed to get to the heart of how we do research, what we mean by research, how it can be applied to people, and what it says about us as individuals.

The Psychologist’s position was that one of the main aims of psychological research should be to find the mean of human behaviour and experience- what most of us feel and do. By this, he hopes to come to a greater understanding of who we genuinely are.

I… disagree.

I can see his point. It’s an attractive idea. Finding the genuine (as opposed to perceived) mean of human behaviour could be interesting.

It’s a pity that the mean, by its very nature, simply does not apply to most of us, most of the time. There’s this thing about humans, you see. One characteristic which defines us more absolutely than any other.

We adapt. We change.

The interesting thing about  humans isn’t in our averages. It’s in the stunning diversity of our adaptations, in why and how we do things differently in different- or similar- situations. It’s in how our hardware interacts with environment, in the commonalities and differences between us. It’s in why we do things the way we do.

The mean erases this diversity. And as a method of studying people, it seems odd to pick one which erases that which makes us most human.

But it’s not just that.

Simplification can be necessary for modeling just about any phenomenon you care to study. Since humans are about as complex as it gets, simplification can be incredibly important in the social sciences. But simplifying is also powerful. What we leave out of a question is as important as what we leave in. The framing of a question can drastically affect the answers we get. If we are to be intellectually honest and aware of the impact of what we do, we cannot make these choices lightly.

A one-size-fits-all decision to use a single method of analysis is not just incredibly intellectually lazy. It also misses the point of why we do research in the first place.


3 thoughts on “The Meaninglessnes of the Mean

  1. This annoys me immensely. Any research psychologist should know that a measure of central tendency means very little without at the very least a measure of dispersion, particularly if the phenomena in question does not satisfy the assumptions of normality. Basic intro stats, basic human sense.

    I just spend a term on qualitative research epistemologies* and methods under a very brilliant woman in her field, and am more convinced than ever that it is a crime against science that so many researchers are allowed to operate without ever once considering challenges to their methodological assumptions. Hell, even basic statistical competence is lacking in the field, much less rational and informed critique of one’s philosophical orientation. I’m starting to think that, screw the software license, we should have to pass a licensing exam before we’re allowed to use a copy of SPSS.

    *For some reason I find it telling that Firefox only recognizes the singular form of “epistemology” – damn the objectivist hegemony!

  2. Great points, I work with 3 psychologist in a small office, I have business and health degrees. Psychologists are interesting and our research is not so psychological as opposed to simply methodology based. I have found many interesting things from my colleagues. Loved reading your post.

  3. I’m planning to take Psychology 101 units this year. I’m hoping to learn about the whole, broad, wonderful range of the human psyche, not just the mean.

What do you think?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s