Theory Is Misused in Organizational Science

Hand juggling syringe and two other blurry objects.

If you read the academic literature in the organizational sciences, you will see the phrase “informed by theory…” It is a device to give the appearance that the research being discussed is theory-based. How theory has informed the work, however, goes unmentioned so the phrase is no more than a meme. But scientific writing is supposed to be precise, so merely saying something is informed by something else is entirely too vague. This is a symptom of how theory is misused in organizational science and beyond.

Why “Informed by” Is No More Than a Meme

I read a lot of academic articles and often see the “informed by” phrase. In most cases it is a vague statement that fails to tell me how the theory being mentioned is connected to the research being described. Given the statement is not helpful in explaining the basis for the research, it is remarkable how it has become so widespread. The explanation is that it is merely a meme—a catchphrase that gives the reader a feeling of familiarity, and perhaps an illusion of clarity. Here’s a few examples of “informed by” in a different context.

  • Informed by the weather, I decided to eat lunch.
  • Informed by my shirt, I decided to play a video game.
  • Informed by one of my students, I went for a walk.

Each of these phases gives the illusion that my behavior is somehow linked to something else, but how is not at all clear. Such clarification makes each one less nonsensical.

  • The weather report predicted thunderstorms, so I decided to eat lunch early so I could get my walk in before the rain.
  • When I glanced at myself in the mirror, I saw I was wearing an Xbox t-shirt which made me want to play a video game.
  • I had a meeting with one of my doctoral students who asked me if I wanted to discuss our research project while we walked across campus.

Theory is Misused in Organizational Science

The way I learned science is that theories are devices used to explain phenomena. A good theory provides an evidence-based framework that explains something. But theories are just that—someone’s supposition about how something works. They need testing to confirm that they correctly explain the world. In the broader scientific process, we begin with exploratory studies to collect information about the world and how things in the world interact. At some point we come up with theoretical explanations that are based on our exploratory research. Those explanations are tentative—they might or might not hold up under  scrutiny. So, we conduct studies to confirm predictions made by theory. This three-stage process involves three kinds of inference—induction from exploratory studies, abduction to generate explanations, and deduction that tests a theory’s predictions.

Organizational science has gone off the rails in that most articles focus on theory without actually testing it. Rather introduction sections merely name drop one or more theories, claiming they drove hypotheses without explaining how. In this way, theory is misused in organizational science because this leaves the reader to fill in the gaps or just take on faith that the research was theory-driven. There is a better way.

Incorporating Theory into Theory Tests

The testing of theory is a multi-stage process that begins before a paper is written. The paper itself is organized around the theory (or theories) being tested.

  • Start with a Clear Purpose. The first paragraph of a research report should make clear that the purpose is to test a particular theory or compare more than one theory to see which one’s predictions (if any) can be supported. Ideally, one should decide that a specific theory is going to be tested before the study is conducted.
  • Explain the Theory. Once the purpose is stated, the theory or theories in question should be explained. Merely providing a citation is insufficient. The author’s understanding of the theory should be shared with the reader.
  • Derive Hypotheses from the Theory. It is crucial that the paper explain how the theory led to the hypotheses. I rarely see this done, and too often authors rely on memes like “informed by”, apparently assuming that it will be obvious. In my experience, it rarely is.
  • Analyses Should Test Theory Predictions. Most papers do a good job of linking analyses to predictions, although here there can be major gaps as well. The one I see most often is how control variables are sprinkled into papers with no theoretical justification other than they have been shown to relate to the variables of interest. This is a topic for another blog, but suffice it to say that testing a model derived from a theory that includes additional variables is not an accurate test of that theory.
  • Link the Evidence to the Theory. The discussion section of a paper should make clear how the data supported or failed to support the theory. It is not enough to merely say that the hypotheses were or were not supported. Often there are multiple hypotheses with only some supported. What this says about the theory should be explained, and in some cases, it might lead to a modification of the theory, or a statement about the theory’s boundary conditions.

Theory is misused in organizational sciences because it is often not taken seriously enough. Theories are vital scientific devices that should be subject to rigorous tests that go beyond “informing” hypotheses. How theories lead to hypotheses should be explicitly stated, with papers focused on theories themselves rather than mentioning them in passing. True deductive research starts with the theory before methods are chosen and data are analyzed. Reports of that research make it clear how the theory led to hypotheses and how those hypothesis tests support or fail to support that theory.

Photo by Daniel Frank from Pexels

SUBSCRIBE TO PAUL’S BLOG: Enter your e-mail and click SUBSCRIBE

Join 964 other subscribers

1 Reply to “Theory Is Misused in Organizational Science”

  1. Before scholars decide what theory to mention, many scholars first cook their data to make sure the manuscript is publishable (i.e., has lots of supported hypotheses). As long as journals focus on statistically-significant supported hypotheses as a major criterion for publication, our research literature will remain screwed.
    Doi: 10.1002/job.2623

Leave a Reply

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