4.2. Types of Research Questions

As you move from a research topic to a research question, there are some considerations that should guide how you pose your question. First, social scientists are best equipped to answer empirical questions—questions about the facts of the world around us—as opposed to normative or ethical questions—questions about what we as a society should value. Empirical questions can be answered through research, but the answers to normative questions depend on people’s moral opinions. (To say something is “normative” means that it relates to our norms or standards—what we should do.) While research projects can inform how we make decisions about ethical issues, they cannot directly answer normative questions, which are fundamentally a matter of debate within communities and societies about what sorts of principles they want to uphold.

For example, a student in one of our methods classes wanted to research student athletes. Their original research question was: “Should college athletes be paid?” Outside of a research context, this is a great question—the matter of paying or not paying athletes affects the lives of millions of students, and it speaks to critical issues about what we as a society think a college education should entail, and what is a fair reward for the work people do. Unfortunately, this specific question is a normative one that we need to debate, not an empirical question that we can resolve with research. A tip-off here is that it begins with the word “should,” a normative phrasing that you generally want to avoid in research questions. The answer to such a question would be a series of moral arguments, based on the particular values the author and their audience hold in common.

It’s true that research can help us to make moral arguments. For example, if we learn how much money universities make from college sports, or how all the work that athletes put into training and playing shapes their experience of college, that empirical knowledge could help us decide whether we believe student athletes are being exploited by their universities, and whether we believe they have a moral right to be paid for their labor. But then those questions would be our research questions, rather than the normative question of whether athletes “should” be paid.

Let’s consider another ethical question that research can inform but not answer: is SpongeBob SquarePants immoral? In 2012, a Ukrainian government commission began reviewing that cartoon show in response to complaints by a right-wing religious group that its depiction of depraved behaviors—such as SpongeBob’s regular practice of holding his male sidekick Patrick’s hand—amounted to the “promotion of homosexuality” (Marson 2012). (Before the government body was disbanded in 2015, the National Expert Commission of Ukraine on the Protection of Public Morality evaluated media to ensure television shows and other content adhered to the country’s morality laws regarding pornography and other controversial issues.) The agency called a special session to discuss SpongeBob and other suspect kids’ shows, though ultimately the eponymous sponge and his starfish companion stayed on Ukrainian TV. A decade earlier, SpongeBob had also drawn the ire of U.S. conservative groups for appearing with other popular cartoon characters in a music video intended to teach children about multiculturalism—which the advocacy group Focus on the Family said was “pro-homosexual” and served as an “insidious means” of “manipulating and potentially brainwashing kids” (Kirkpatrick 2005).

Can research answer the question of whether SpongeBob SquarePants is immoral? No, because questions of morality are ethical, not empirical. Your family members and pundits on TV can rant about sponge creatures all they want, and they can make better or worse moral arguments for their positions, but this is not a question a social scientist should build a study around. That said, we sociologists could certainly choose to study the public opinions and cultural meanings that surround a popular show like SpongeBob SquarePants. We could conduct experiments measuring the detrimental effects that watching the show has on children’s behavior. We could even use surveys to find out precisely how many people in the United States find SpongeBob and/or Patrick repugnant. But sadly, we could not settle the question of whether SpongeBob is indeed morally reprehensible, given that it is not an empirical question.

As you start designing your study, your choice of a particular empirical question will also be influenced by your study’s general purpose. There are three approaches that a research study will typically take: exploration, description, or explanation. These are not mutually exclusive categories, and a study may fall into multiple categories.

Exploratory research is often conducted in new areas of inquiry, where the goals of the research are: (1) to scope out the magnitude or extent of a particular phenomenon; (2) to generate some initial ideas or hunches about that phenomenon; or (3) to test the feasibility of undertaking a more extensive study. For instance, if the citizens of a country are generally dissatisfied with their government’s policies during an economic recession, sociologists could create and implement new surveys to measure the extent of that dissatisfaction and probe for possible causes of it, such as anxieties about unemployment, inflation, or higher taxes. This research may not lead to a very accurate understanding of the target problem, but it may be worthwhile nonetheless in order to get a preliminary sense of its nature and extent, serving as a stepping stone to more in-depth research.

Descriptive research is directed at making careful observations and detailed documentation of a phenomenon of interest. Because these observations follow the scientific method, they hopefully are more accurate than casual observations by untrained people. Much exploratory research overlaps with descriptive research: we often want to describe the magnitude of an emerging problem as a starting point in understanding it. Yet descriptive research is also helpful to conduct on an ongoing basis, and it can involve well-studied topics.

A common type of descriptive research is the work of government agencies to tabulate statistics about the population. In the United States, for example, the Bureau of Labor Statistics uses survey questions to estimate employment by sector every month. The U.S. Census Bureau regularly conducts demographic surveys that allow policymakers and social scientists to track the growth of a wide range of racial and ethnic groups over many years. In general, government agencies, corporations, nonprofits, and other organizations are in great need of such descriptive research so that they understand the circumstances that their citizens, clients, and members are experiencing. They can use these assessments to create new programs or policies to meet people’s needs or preferences. For that reason, if you decide to use your sociological research skills in a nonacademic setting (as described in Chapter 2: Using Sociology in Everyday Life), you will likely be doing a lot of descriptive research.

That said, descriptive research is also a large component of many studies that academic sociologists do. For example, a sociologist’s ethnographic study of gang activities among adolescent youth in urban areas might entail detailed observations of the children’s activities. A study of religious practices in immigrant communities might chronicle the evolution of those practices over time. In conducting this descriptive research, sociologists are gathering essential information about what is actually going on in the social spaces they observe.

Explanatory research seeks explanations of observed behaviors, problems, or other phenomena. You might think of the difference between descriptive and explanatory research in this way: while the former seeks answers to basic “what,” “where,” “who,” and “when” types of questions, the latter examines questions that are more complex—whether or not one concept affects another, and “why” and “how” those concepts are related. Put another way, explanatory research attempts to “connect the dots” in research, by identifying certain important factors and showing how they lead to specific outcomes. Let’s consider the hypothetical studies we discussed at the end of the last paragraph in this light. For an explanatory study about urban gangs, sociologists might seek to understand the reasons that adolescent youth in urban areas get involved with gangs. For an explanatory study of immigrant religious practices, researchers might examine why these practices evolve in the ways they do within particular local or national contexts.

Most studies you read in the academic literature will be explanatory. Why is that? Explanatory research tries to identify causal relationships that are generalizable across space and time. That means the findings of such research should matter to many people: because we’re learning something fundamental about the relationships between the concepts we’re interested in, our conclusions aren’t limited to a one-off situation or scenario. It also means our findings are actionable: because we know what causes what, we can act individually or collectively to promote, discourage, or alter the phenomenon we’re studying. In other words, explanatory research gives us a better sense of how and why society operates the way it does, rather than just describing what particular aspects of society look like.

Arriving at compelling explanations for social phenomena requires especially strong theoretical and empirical skills. You need to have a sophisticated understanding of how a social process operates and rule out any alternative stories, and you need to collect empirically sound data and rigorously analyze it. For these reasons, sociologists often see explanatory research as a “higher” form of research, one that is exceedingly challenging to do well. At the same time, they will frequently engage in some amount of exploratory and descriptive research for any given study, particularly during its initial phases. Indeed, these other approaches can be especially important in helping us understand a relatively new or hard-to-study phenomenon: without good descriptive research to draw from, any theorizing we do will be built on shaky empirical foundations.

Deciding on the primary purpose of your research will shape the study you ultimately propose and conduct. If you are doing academic work, your instructor or advisor may push you to be less “descriptive” in your approach, and to focus more on seeking out explanations for what you observe. If you are studying a topic that so far has generated only a small amount of literature, however, you may very well want to conduct exploratory research to generate plausible theories, or descriptive research to understand the scale or characteristics of a particular phenomenon.

The overall purpose of your research will also inform the research questions that you pose. Probably the easiest questions to think of are descriptive research questions. For example, “What is the average student debt load of college graduates?” is a descriptive question—and an important one. In this case, you aren’t trying to identify a causal relationship. You’re simply trying to describe how much debt students carry. When you seek to answer a descriptive research question like this one, you might find yourself generating descriptive statistics—counting the number of instances of a phenomenon, or determining an average, median, or percentage. You can also pursue descriptive research questions using qualitative methods. For instance, you might conduct in-depth interviews or focus groups to gauge the public’s view of student debt, describing the range of opinions on that subject.

In the next section, we’ll focus on explanatory research questions. We will detail one strategy for developing questions based on whether your study is using quantitative methods or qualitative methods. We’ll also connect those two types of research questions to two kinds of empirical analysis—deduction and induction.