Theories and Hypotheses

By Karen Sternheimer

What’s the difference between a theory and a hypothesis? Which one is absolutely necessary for research and which one is common but not a requirement?

I’ll give you a hint: If you are a sociology major, you might have to take a class called Sociological Theory. You probably don’t have to take a class called Sociological Hypothesis (if you do, I’d like to hear more about it in the comments below because I have never heard of such a class before).

A theory is a system of ideas that has been developed after multiple studies. Theories are constructed by examining the results of research and repeated observations. Researchers begin with a theory, and end by noting how their findings add to that theory, or set of theories.

A hypothesis is an educated guess about how two or more things are related. It might be based on our informal observations and life experiences. Here’s the confusing part: We often use the word “theory” in everyday language when we really mean hypothesis.

For example, you might hear a friend say that they have a “theory” that you struggle on tests because you need to be more confident, but what they are really suggesting is that they have a hypothesis that if your confidence improves, your test scores will, too.

Why does all research need theory but not necessarily a hypothesis?

Sociological theories are explanations about an aspect of society that we are interested in learning more about. Theories are not developed from only one study, but rather are based on multiple studies by many scholars who add to the development of the theory.

Think of it as similar to solving a mystery: You might get the process started by narrowing down possible locations and suspects, encouraging others to pick up where you leave off. Others might begin working on the mystery and narrowing things down further, or even suggesting a new suspect after ruling some out based on their research. Theories are created collectively by scholars who share their work. If you were going to solve a mystery, why would you want to start from square one when others have already spent time, effort, and maybe money in search of a solution?

Theories help us understand the conversation that has been taking place on a particular research topic, as I blogged about in 2018. Starting a research project without reading previous research and theory ignores all of the work that others have done and shared.

For instance, let’s say that you are interested in learning more about gender and appearance. You might begin by reading the work of Candace West and Don H. Zimmerman. Their 1987 article, “Doing Gender” focuses on their theory that gender is a social performance developed within our interactions with others. More than 14,000 other publications referenced this article, and thus this theory has been refined by research many, many times in the past three decades. This tells us that many other studies have found this framework useful and have used the findings from their research to add to and refine this idea.

Here’s another analogy: Imagine you are going to have dinner with a large group of people you don’t know. It’s a lot of work to host and do all of the cooking, and it could be very expensive. Maybe this group of people has eaten together many times before and prefers dishes that others have made. Wouldn’t it be easier to have a potluck dinner and ask what you can make that everyone will approve of? It’s not a perfect analogy, as research and theory construction takes place over years, but you get the idea. It often will save time to acknowledge the work that has already been completed. 

Theories are necessary in sociological research, and while hypotheses are common, they are not required. Why not?

Some research uses inductive reasoning, or gathering information first, and then attempting to draw conclusions. Unlike deductive reasoning, where we test our hypothesis in order to find out if we can find evidence in support, inductive reasoning only requires us to be curious about something without making any predictions. We might begin this way, and later create a hypothesis, but it is not required.

Hypotheses are related to theories and it’s important to test them. Whether or not we find supporting evidence, hypotheses inform the larger theory, and contribute to the body of scholarship created. The next time you or someone proposes a “theory,” ask yourself if it’s really a hypothesis in disguise.