What is a “theory”?

Coworker A: Again?! That’s the third time this week someone’s eaten my sandwich! They need to padlock this stupid shared refrigerator for the office.

Coworker B: Any idea who’s taking your food?

Coworker A: Well, I can’t be sure, but I have a theory about that…

The term “theory” is fraught with misuses, variable meanings in different settings, and even petty academic disciplinary disputes. As in the vignette above, “theory” is often used in society and pop culture as anything from a gut feeling to an educated guess. Ask a physicist, and they will have a very rigid definition of “theory” held tightly by the natural sciences. Ask a social scientist, and you will likely get a much less rigid answer than the physicist gave but still one that is veiled in scholarly language. Then go back and ask the physicist what they think of the social scientist’s definition of “theory,” and you might just end up with a fist fight on your hands…

So, perhaps a prerequisite to defining “theory” is establishing who gets to define it. As this is a text on social science research, deference will be given to the social science definition. However, as social science is an approximation of the natural sciences and their methods, it is relevant to consider that definition as well.

Natural Science Definition of “Theory”

As with “the” scientific method, there is no single, all-encompassing definition of theory in the natural sciences. As this is a social science text, this concise explanation of the term “theory” in the natural sciences from the University of California Berkeley Museum of Paleontology will suffice:

The process of science works in much the same way whether embodied by an individual scientist tackling a specific problem, question, or hypothesis over the course of a few months or years, or by a community of scientists coming to agree on broad ideas over the course of decades and hundreds of individual experiments and studies. (2024a, para. 1)

In science, a [theory is a] broad, natural explanation for a wide range of phenomena. Theories are concise, coherent, systematic, predictive, and broadly applicable, often integrating and generalizing many hypotheses. Theories accepted by the scientific community are generally strongly supported by many different lines of evidence-but even theories may be modified or overturned if warranted by new evidence and perspectives. (2024b, para. 1)

More succinctly, a theory within the realm of natural sciences is derived from repeated hypothesis testing through scientific methods, and it is a concise, coherent, and general statement of the conclusion. However, theories are always tentative and subject to revision or complete rejection, should the above-mentioned processes later generate results that lead to such conclusions.

Social Science Definitions of “Theory”

In the social sciences, the term “theory” is commonly used in two ways: descriptive and normative. The primary distinction between these two uses is that normative social science theories introduce value judgments (i.e., how things ought to be). Both commonly utilize inductive and/or deductive reasoning. One way to think about the two is that descriptive social science theories are social but with an emphasis on science, whereas normative social science theories are social but with heavy influence from philosophy and/or the humanities.

Unlike scientific theories that should have no or almost no exceptions, social science theories typically describe patterns but embrace the commonality of exceptions. Humans are fickle creatures!

Descriptive Social Science Theory

A descriptive social science theory is simply an attempt to explain ways that social things generally happen. These should be consistent, tightly bounded for variables and settings, and clear. For example, one might utilize the qualitative methodology of grounded theory to create a model of how cisgendered female high school seniors identify their future collegiate major. There are no ideological commitments (in the politicized sense) and no judgments about how they should select a major. Such theories are purely concerned with how something happens and, normally, the prediction of future happenings.

Normative Social Science Theory: Two Approaches

Normative social science theories really fall into two categories.

First, there is the descriptive-plus-value approach. Extending the illustration from the previous paragraph, the theory would become normative when it goes beyond just describing how those students select their majors and also infuses it with perceptions about causes and/or prescriptions about what ought to be. A critical theorist might look at the results of that grounded theory and perceive how societally-imposed gender norms have a significant—and perhaps preeminent—impact on those decisions (critical theory being, perhaps, the most impactful normative social science theory in the 2010s-2020s). It is unlikely that high school students would understand their major choice in light of broad and vague societal impacts, thus the critical theorist might see the descriptive theory as woefully inadequate. They might then infuse that “empirical investigation” with “utopian speculation” (Brookfield, 2005, p. 27), suggesting that societal gender norms more accurately or completely describe how cisgendered female high school seniors select their future college major than what the participants themselves described. Thus, the grounded theory becomes not just about the participants’ perceptions and feelings but also an infusion of the researcher’s contextualization and interpretation.

Second, there are normative social science theories that either start or perpetually remain only normative, prescribing what ought to be. Now, that is not to say that such normative theories do not have research done that supports them; however, that is more ancillary than causal. Most religions contain this type of normative social theory (ex., Judeo-Christian morality); there are right and wrong ways of doing society and human behavior, but those are not derived from research but rather sources like divine revelation or human spiritual intermediaries. As mentioned in the previous paragraph, critical theory is probably the most impactful normative social science theory, and it falls into this second category. There absolutely are extensive research studies done on topics and aligned in support of critical theory’s tenets; however, as Brookfield (2005) observed, “verification of the theory [critical theory] is impossible until the social vision it inspires is realized” (p. 29). It is not so much about testing and/or refining critical theory as it is about implementing a grand vision of how society ought to be.

How is theory actually used in social science research?

At the most basic level, social science research uses theories to advance knowledge at a larger scale than would naturally happen with isolated studies. Theory helps coalesce knowledge to explain specific phenomena, which in the social sciences is often explaining how or why something social happens.

Descriptive theories often prompt further research and guide practice. Such theories should be relatively consistent across time and place, thus practitioners should be able to apply descriptive theories into their related practice without much difficulty. Research studies using descriptive theories typically generate more data across time, place, and demographics that support the theory.

In normative theories, those extensions might come from conceptual applications to a new population or setting as well as aggregating existing research to demonstrate alignment. For example, Phillips and Lincoln (2017) extended critical theory to apply to military veterans. While that theory originated with socioeconomic classes and then later included race/ethnicity and sex/gender, veterans were a more specific and smaller population subset that was not tied in until decades later. They argued that veterans were an historically marginalized population (a conceptual extension), which they supported with the existing knowledge base (demonstrating alignment).

Theoretical (and Conceptual) Frameworks

While the specifics of how they are used and discussed will vary by the type of research being conducted and the nature of the publication, theoretical and conceptual frameworks are normally a part of social research (though, for example, a practical action research study would probably not use one). The two are closely related concepts, with a framework being theoretical because it uses something formally classified as a theory, while a framework is conceptual if it is using something not formally classified as a theory. However, both theoretical and conceptual frameworks are used similarly in research studies.

Theoretical/conceptual frameworks are used to guide research design, interpret findings, and suggest future applications or extensions of the research. In research design, theoretical/conceptual frameworks are often used to develop questions or define variables based on previous research associated with the theory. So, for example, if a theory suggested that you should find A, B, and C; then in your research design, you would ask questions to find if A, B, and C were present. Then in analysis and/or interpretation, one would specifically look for A, B, and C. If this were a qualitative study, the coding, categorizing, and themeing would be defined by A, B, and C (though that does not preclude the possibility that those items are not found in the new data set). Finally in suggesting future applications or extensions, it is common that the researchers propose ways in which the theoretical/conceptual frameworks might be taken one step further than the present study. Continuing our example, that might be that future studies should look for A, B, and C among some different demographic group.

Theory-less Research?

However, not all research has to create, expand, align to, or interpret data with a social science theory. Such research that does not use theory tends to be at two ends of the research spectrum. On the one side, some basic research (but definitely not all) does not use theory, as it is simply collecting and analyzing data for the sake of knowledge. However, it is then common for secondary research to occur as others use that data and interpret it with social theories. On the other side, many types of applied research (especially, practical action research) do not need theory to accomplish their purposes; they are being done solely to solve a local problem. Once the problem is understood, a solution is tested, and tentative conclusions are made, there is no need for greater context or tying to a grand scheme of explaining human behavior. Additionally, some research is meant to be purely based on participant views without adulteration (such as qualitative description and much phenomenology, described in a later chapter on the types of qualitative research).

Occam’s Razor

In its most basic form, Occam’s Razor, sometimes called the Parsimony Principle, suggests that the simplest explanation is usually the best explanation (originally, “plurality should not be posited without necessity” (Duignan, 2024, para. 1)). The reasoning is that each assumption that is introduced is possibly incorrect. The more assumptions there are in an explanation, then the more probable that there is an error. However, that also does not mean the simplest explanation is always right.

This system of interpretating of data and analysis makes sense in the far more predictable natural sciences. If the temperature of water in a container rises as it sits atop an active heating element, the most reasonable conclusion is that the heat is being transferred from the heating element to the water. Of course, that is not always the case. The room temperature could be rising and be contributing to the rise in water temperature. A malicious lab assistant seeking to skew the experiment could have poured some super heated metal into the water container moments before measurement. We could continue to list reasons why the water temperature is rising apart from the simple explanation of the heating element transferring heat to the water; however, the simplest and most likely explanation for this rising water temperature is that the burner beneath it is on.

The social sciences are not as predictable, at least in the sense of consistently following laws. If that were the case, we would have little need for courts and lawyers! While Occam’s Razor may be best for evaluating individual circumstances, it does little to tie those together for generalized or broad explanations of social phenomena. Thus, social science research has increasingly skewed toward favoring (or demanding) theoretical tie-ins to almost all published research, and so it is easy to assume that social research must have theory. That is not the case, and in much applied research, Occam’s Razor applied by expert knowledge is sufficient (or even best). But, if you seek to publish in scholarly outlets or work in academia, you may find theory to be less optional.