3.1. What Do We Mean by “Theory”?

Theories are systematic explanations of a natural or social behavior, event, or other phenomenon. To be valid, a theory needs to fit the available evidence, or data, we obtain from observations. If it doesn’t fit that evidence, it needs to be adjusted or reworked in a way that does.

Theories and observations are the foundations of scientific knowledge. Theories provide meaning and significance to what we observe, and observations—when rigorously conducted—validate or refine existing theory or construct new theory. We can therefore say that scientific research operates at two levels: a theoretical level and an empirical level. At the theoretical level, we develop abstract concepts that relate to a social phenomenon and identify relationships between those concepts. Here, we’re generating, or building, theory. At the empirical level, we see how well these theoretical concepts and relationships reflect the reality we can observe. Here, we’re testing, or refining, theory. Science advances as scientists shift continually between theory and observations—between theory generation and theory testing—and assemble an inspired but also well-substantiated understanding of a topic. Over time, a theory becomes more and more refined—it fits the observed reality better—and the science matures.

Theories can exist at different levels of abstraction and encompass very different phenomena. That said, it is useful to think of theories as the building blocks of science. For one thing, they come in different sizes: smaller theories can be part of a larger theory explaining a particular phenomenon. For example, in putting forward her theory of emotional labor, Arlie Hochschild (1979) described three types of work that people engage in to regulate their emotions: bodily (doing deep breathing or other practices to change their mood), cognitive (changing their way of thinking in the hopes of modulating their feelings), and expressive (adapting their facial and other expressions to match the desired feelings). Other scholars, in turn, have elaborated and built on Hochschild’s original theory. For instance, Robin Leidner describes how corporations like McDonald’s standardize every possible interaction with a customer. In doing so, they dictate precisely how emotional labor is done—what Leidner (1993) calls the “routinization of service work.”

Like the imagery of building blocks might also suggest, theories allow us to break down a phenomenon of interest into its constituent parts. (This is essentially what we mean by “analysis” in science—the Latin root for “analysis” means “to break up.”) In this vein, sometimes researchers refer to their theories as theoretical models (or just “models”). Theoretical models provide a simplified understanding of some process. Like a model of an airplane, they reduce a complex phenomenon into its most important parts. This simplification allows people to better understand how, for instance, they might change one or a few aspects of a particular situation and bring about different results. In this way, theories help us act on the world for our benefit.