Deciding on a Good Research Question

Before we can create a hypothesis, we need a good research question. A hypothesis is a statement that can be proven or disproved – but the research question is the reason we come up with a hypothesis. It’s almost, but not quite, like a “Question and Answer”. If it helps, you can think of the hypothesis as a provable or disprovable answer to a research question.

#Q: Why is the sky blue?

#H: The sky is blue because of Rayleigh scattering.

# [now you can design a test to see if this is true or false]

Think about the questions you have regarding Public Administration or related subjects. Don’t worry about the controversies or the things you’d like to prove right now, just focus on a question that you are genuinely curious to answer.

(If you cannot think of a question you are curious about – take some time to get curious). Kate Turabian says that “a researcher’s most valuable ability is the knack of being puzzled by ordinary things”. When you look at Public Administration, what puzzles you? If nothing puzzles you, then maybe you are not taking the subject seriously enough.

In this course, we are starting out a bit backwards. We are starting with a dataset and formulating questions from the data. Look for a dataset that could be used to inspire, and then answer, questions you may have.

Narrowing It Down

A lot of questions are good, but too broad:

# What is the effect of sunlight on the ocean?

This is a very important question, but you could spend literally your whole life answering it. For the purposes of this class and your career, you’ll need to be more focused.

Broadly speaking, there are two methods to find focused questions that might be useful:

GAPS and PROBLEMS

Alvesson and Sandberg call the first method “gap-spotting”. Basically, you are looking for a gap in our current knowledge, and trying to ask a question about it:

# We know that sleep is necessary for survival, but what is the purpose of sleep and why do we literally die without it?

# We know that most animals yawn, but why do they yawn? What is the purpose of this reflex?

# We know that everybody likes pizza, but why do some people enjoy pineapple on pizza? Are they crazy?

This method begins with something that is widely known or understood in the literature, followed by an observable “gap” in the literature. The best kinds of questions in this regard come from “gaps” that have not been spotted before (not all gaps are obvious).

Sometimes gaps can be more than just missing context – they can also include confusion:

# Some research says that step-children are more likely to be harmed by step-parents, while other research suggests this is not true. Which is more accurate?

Or “lack of empirical support” (we believe something, but it lacks evidence):

# High cholesterol is widely theorized to be responsible for heart disease, however firm evidence for this is lacking. Is high cholesterol actually a cause of heart disease?

Or “application” (something that exists can have a new application):

# Public libraries are meant to improve the intellectual abilities of the general public, but can they also function as social service agencies?

Regardless of the method, when we discuss “GAPS” we are talking about something that is missing. We ask a question because there is something missing in our understanding about a subject.

There is another method that can be broadly described as “PROBLEMS” or “PROBLEMATZATION”:

# ‘endeavour to know how and to what extent it might be possible to think differently, instead of legitimizing what is already known’ - Foucault (1985), as quoted in Alvesson and Sandberg (2013)

The idea here is that you are setting out to challenge the assumptions of a field of research – including, potentially, your own assumptions.

# The literature suggests that people should drink milk every day for health, because stronger bones is an important health outcome. What if stronger bones is not actually an important health outcome?

Note that the above example is a bit different from gap-spotting. We are not questioning whether milk improves the strength of bones, or whether our analysis of the relationship between milk and bone-strength is correct – rather we are questioning the assumption that underlies the ‘milk-drinking paradigm’: that stronger bones is an important health outcome.

Alvesson and Sandberg (2013) suggest that problematization is more useful for truly ground-breaking research than gap-finding. However, in this course, both types of questioning are fine.

Thinking About Your Question

After you get an idea of your topic area and what angle you’d like to approach it, it helps to think about what you want to accomplish with your research (aside from passing the course).

Turabian (2010) suggests the following formula:

# 1) I am working on topic "X"
# 2) Because I want to discover "Y"
# 3) So that I can help others understand "Z"

In practice, it can look like this:

# 1) I am working on topic "X" (The death penalty in America)
# 2) Because I want to discover "Y" (Does the death penalty actually deter crime?)
# 3) So that I can help others understand "Z" (The actual relationship between capital punishment and crime)

The trick is in this class, we want to find our data first before we try to answer the questions above (otherwise you may take too long to find data).

“Y” is your actual research question, “X” is the topic and “Z” is what you are trying to accomplish.

Once you have an actual question, briefly evaluate it for appropriateness:

# 1) Is the potential answer unknown? (No asking "What is 2+2?")
# 2) Can the answer be plausibly *disproved*? ("How many angels are there?" can't be disproved)
# 3) Can the question be answered with the available data?
# 4) Does the answer matter? ("How many apples did George Washington eat in 1773?")
# 5) Does the answer require analysis? (Answer should be more complex than just "yes" or "no")

If you follow all the above steps, you should end up with a decent research question. After that, you can formulate a hypothesis that can be tested to answer the question.

Remember that a hypothesis is a statement that can be proven or disproven. If you’ve asked a good question, the hypothesis should present itself.

# Q) Does the death penalty deter crime?

# H) Hypothesis: the presence of the death penalty in a state does NOT significantly affect the rates of capital crimes in the particular state.

How to NOT ask a research question

# 1) I want to understand the effect of taxes on crime

# 2) What is the relationship between sunlight and migraines?

# 3) Is government good or bad?

# 4) I want to show that my [favorite policy] is good

The above examples are NOT good ways to approach your research question. They are either too vague, too broad or too subjective. #1 and #4 aren’t even questions.

The best thing you can say about these examples is that they are the beginning of searching for a valid question. They can be expanded or developed into questions that actually make sense in a scientific context, and meet the criteria we discussed previously.

# 1) The relevant literature suggests that increases in property taxes leads to similarly proportioned increases in the property crime rate. Is this true even in areas where property taxes are already very high? [GAP-spotting, neatly focused, testable, can be proven/disproved, subject matter is important]

# 2) A review of medical journals reveals that migraines are strongly associated with sunlight, but could it be possible that migraines are in fact associated with other variables that may only be related to the presence of sunlight, such as pollen? [Problematization, neatly focused, testable, can be proven/disproved, subject matter is important]

CITATIONS

Alvesson, M., & Sandberg, J. (2013). Constructing Research Questions: Doing Interesting Research. SAGE Publications Ltd. https://doi.org/10.4135/9781446270035 

George Mason University. (2018, August 18). How to write a research question. The Writing Center. https://writingcenter.gmu.edu/writing-resources/research-based-writing/how-to-write-a-research-question Accessed September 9, 2025.

Panke, D. (2018). Research Design & Method Selection: Making Good Choices in the Social Sciences. SAGE Publications Ltd. https://doi.org/10.4135/9781529682700 

Turabian, K. L. (2010). A manual for writers of research papers, theses, and dissertations: Chicago style for students and researchers. (8th ed.). University of Chicago Press.