What Makes a Good Dissertation Question?

What Does a Dissertation Question Need to Accomplish?

At doctoral level, “Is this an interesting question?” is not enough. A topic can be fascinating yet produce a weak dissertation question because the proposed inquiry is already well answered, theoretically thin, methodologically impossible, too broad to resolve convincingly, or too dependent on conditions the researcher cannot control.

The difficulty is that doctoral research has to balance two pressures that can pull in opposite directions. The question needs enough originality and significance to support a contribution to knowledge, yet it must remain sufficiently bounded that one researcher can investigate it rigorously.

That makes a good dissertation question less about sounding ambitious and more about identifying a consequential uncertainty that can realistically be turned into a defensible scholarly contribution.

A Good Dissertation Question Connects an Important Unknown to a Defensible Contribution

Six Tests for a Strong Dissertation Research Question

The question identifies something that is genuinely unresolved

A dissertation begins with uncertainty. If the answer is already adequately established in the literature, repeating the question in a new study does not automatically create a doctoral contribution.

This does not mean that nobody can ever have investigated the topic before. In fact, a completely untouched topic can be difficult to position theoretically. Originality may instead arise from addressing a contradiction, testing an assumption under conditions where it may not hold, extending an explanation, examining a neglected mechanism or population, developing a methodological approach, producing new evidence, or reconsidering an established interpretation.

The important question is not simply, “Has anyone studied this?” It is: What remains uncertain, and why would resolving that uncertainty change what scholars can reasonably know, explain, interpret, or do?

The question creates a plausible contribution to knowledge

A research gap and a contribution are not identical. Finding that a particular population, location, variable, or technology has received little attention establishes an absence in the literature. It does not by itself establish why filling that absence matters.

A strong dissertation question should allow you to explain what becomes possible if the question is answered. Perhaps the study could refine a theory, adjudicate between competing explanations, reveal conditions under which an established relationship changes, challenge a prevailing assumption, provide evidence unavailable to previous researchers, or develop a useful conceptual or methodological account.

The form of contribution differs substantially by discipline. Doctoral work in history, computer science, education, mathematics, engineering, philosophy, and molecular biology will not demonstrate originality in the same way. Your disciplinary conventions and doctoral regulations therefore matter.

The question is specific without predetermining the answer

A dissertation question needs boundaries. Readers should be able to determine what phenomenon, relationship, process, problem, text, population, system, period, or context is actually under investigation.

Yet specificity should not turn the research question into a disguised conclusion. Consider the difference between asking why a particular intervention “improves” learning and asking how, whether, or under what conditions it affects a defined learning outcome. The first wording may already assume what the study is supposed to establish.

The question implies evidence that could actually answer it

A question is not researchable merely because data can be collected about its topic. The evidence and research design must be capable of addressing the particular claim embedded in the question.

If you ask about causal effects, for example, collecting opinions about whether participants believe something caused an outcome does not necessarily establish causality. If you ask how a process unfolds over time, a single cross-sectional measurement may be inadequate. If you ask about institutional practices across a country, studying one conveniently accessible organization may not support the breadth of the question.

This is a useful stress test: Imagine the data you will have at the end of the study. Can those data actually answer the question as written?

The question is feasible at doctoral scale

Doctoral research is expected to be substantial, but substantial does not mean unlimited. Western University's epidemiology and biostatistics doctoral guidance, for example, describes an appropriate thesis question as one that generates a study small enough to complete in a reasonable timeframe while remaining substantial enough to constitute a contribution. That tension captures an important feature of doctoral question design.

Feasibility includes time, funding, access, recruitment, ethics, equipment, data availability, analytical demands, and expertise. It also includes the dependencies built into the design.

A theoretically elegant question may therefore be a poor dissertation question if answering it requires five organizations that have not granted access, a population that is exceptionally difficult to recruit, specialist methods unsupported by the supervisory team, or longitudinal data extending beyond the candidature.

This is why how ambitious a dissertation question should be cannot be separated from the resources and research environment in which the dissertation will actually be conducted.

The question is capable of sustaining a dissertation-length argument

Feasibility has a lower boundary as well as an upper one. A question can be manageable but intellectually insufficient.

If the question can be answered adequately through a simple factual lookup, a routine descriptive analysis, or an observation whose interpretation requires little scholarly argument, it may not support doctoral-level inquiry. A dissertation question should ordinarily create room for sustained engagement with relevant scholarship and for analysis or reasoning capable of producing the contribution claimed.

This does not require a complicated question. Some excellent research questions are remarkably simple to state. The complexity belongs in what must be investigated and argued to answer them.

The question and the method should constrain each other

Researchers sometimes formulate the question first and treat methodology as a later implementation decision. Conceptually, however, the relationship is iterative.

Your question determines what evidence is relevant, but the availability and quality of credible evidence may force you to revise the question. Likewise, methodological limitations may require a change in the claims the question asks you to establish.

This does not mean choosing a question merely because a familiar method can answer it. It means avoiding a mismatch between the epistemic demands of the question and the evidence the dissertation can produce.

The question should remain meaningful if the result surprises you

A useful test is to imagine several plausible outcomes. Suppose the expected relationship is strong, weak, absent, or reversed. Would each result still tell you something relevant to the underlying scholarly problem?

If only one result would make the dissertation interesting, the question may be too dependent on confirming a preferred hypothesis. Strong questions often matter because resolving the uncertainty is useful regardless of which credible answer emerges.

From an Interesting Topic to a Dissertation Question

A possible resulting question might ask: Under what conditions does generative AI-supported revision influence students' subsequent ability to revise comparable academic writing independently?

This wording is not automatically a good dissertation question. Its quality would still depend on the state of the literature, theoretical framing, operational definitions, proposed design, and doctoral expectations. The improvement is that it identifies a more precise uncertainty and makes the evidentiary requirements easier to interrogate.

Common Ways Dissertation Questions Go Wrong

Stress-Test the Question Before Building a Dissertation Around It

Write your proposed question at the top of a page and interrogate every important phrase. Which concept does it invoke? What claim would an answer require? What evidence would support that claim? Which literature establishes that the uncertainty exists? Why would resolving it matter?

Then conduct a second test from the opposite direction. Ask what the smallest credible study capable of supporting the intended contribution would look like. Doctoral ambition should not be reduced indiscriminately, but unnecessary complexity deserves suspicion.

Finally, compare the question with your institution's current doctoral regulations and disciplinary expectations. Generic criteria can help you diagnose a question, but the degree-granting institution ultimately determines what must be demonstrated in the dissertation.

Before Committing to a Dissertation Question, Check This

Frequently Asked Questions About Dissertation Research Questions

A Dissertation Question Must Be Worth Answering and Possible to Answer

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.