When people ask me what I do for a living, I tell them I’m a statistician. Their eyes typically glaze over, but that’s because they can’t imagine anything more boring, not because they don’t understand what “statistician” means. (Okay, most people don’t really understand what a statistician does, but they have an inkling.) But “statistician” is only half the story. The other half is “social science research methodologist,” but I don’t routinely impart that information because, if I do, people’s eyes glaze over because they don’t have a clue what a “social science research methodologist” is or does.
Without getting too esoteric, the research methodologist part of my job description deals with the design aspect of research projects–things like questionnaire design, survey sampling and project implementation. This is a critical part of any type of research, because if the design and implementation parts of the process aren’t sound, the data that’s generated is worthless (or, in some cases, downright harmful). And what’s more, it’s impossible to tell that your data is worthless after the fact. It can be a real trap, and if you mess up in the first two stages of this three-step process, you’re sunk, no matter how good a data analyst (read: statistician) you are. Basically, it’s a garbage in/garbage out sort of thing.
One of the key components to designing a well-crafted social science research project is what is known as “operationalization”—a never-ending stumbling block in the world of the social sciences. Operationalization is essentially the process of turning “constructs” or concepts into concrete, tangible, measurable placeholders. Clear as mud? Perhaps an (admittedly oversimplified) example will help.
Say you’re interested in understanding what relationship, if any, exists between social class and education. You might, for instance, posit that social class and educational attainment are positively related and you want to see if that supposition is true. That’s a perfectly valid research goal. So, how do you go about determining whether a relationship exists? Among other things, you need to deal with the fact that “social class” and “education” are concepts (or constructs, if you prefer). The only way to measure these things is to define them in a tangible way. Social class might be operationalized by the real world concept of “annual household income.” Educational attainment might be operationalized as “years of education completed.” Now, just about any social scientist worth his/her salt would readily admit that these real world placeholders don’t fully capture the constructs that they represent. Social class implies far more than household income. Educational attainment is more than simply how many years of schooling someone has completed. But it’s difficult, bordering on impossible, to completely flesh out concepts as elusive (not to mention politically charged) as social class and educational attainment.
The operationalizing process is, to a greater or lesser extent, necessarily a compromise and, in the end, frequently leads to complaints that any relationships between variables—in the case of the illustration above, social class and education—is unproven because the operationalization was incomplete or inadequate or, in certain cases, simply wrong. It’s the bane of social science research because anyone with a bone to pick with the results can always say that the project was the victim of a poor design. Ultimately, it’s the very subjectivity of operationalization—the fact that two different people can have very different ways of expressing an intangible construct—that makes social science research so “soft.”
At this point, you’re undoubtedly saying “that’s all very interesting,” while stifling an exceptionally large yawn, “but why are you telling us this? I thought this was a column about photography.”
Well, bear with me and prepare for a tortured analogy.
When it comes to social science, a consensus can be reached in defining a construct. The problem lies in the operationalizing process discussed above: identifying a measurable real life placeholder for the definition. But when it comes to art—including, but not limited to, photography—the very problem of reaching a consensus on a construct definition is where the process grinds to a halt.
Consider the terms by which we describe art: beautiful, provocative, disturbing, emotion-laden and so forth. Now try to devise a substantive, consensus-based definition for these terms without resorting to a meaning that is inherently subjective. It’s virtually impossible…and there’s nothing wrong with that.
But this inherent subjectivity is what makes so much of the discussion and debate about art so superfluous—at least to me. People have a tendency to toss their opinions about art around as though they’re really objective facts. But they’re not. So much of the discussion that I hear and read about art amounts to, when boiled down, an argument over whose opinion is “right”—basically an oxymoron.
There’s no more compelling reason to make your photography (or whatever form of art you choose to engage in) about yourself. Art is, after all, about individual expression. If your principal goal is to have others comprehend the meaning that infuses your art, you may feel the need to make some alterations to that expression, but you simply can’t expect everyone to understand what you’re saying, let alone to agree with or approve of your meaning. To do so is a fool’s errand.
Thursday Tips is written by Kerry Mark Leibowitz, a guest blogger on 1001 Scribbles, and appears every other Thursday. To read more of his thoughts on photography, please visit his blog: Lightscapes Nature Photography.