01 · The Question
How Much Should a Dissertation Try to Accomplish?
A doctoral dissertation is expected to contribute something original. That expectation can create a tempting inference: if the contribution must be substantial, the research question should also be large.
That is not necessarily true. A dissertation can be intellectually ambitious while investigating a tightly bounded problem. Conversely, a question spanning several populations, countries, theories, methods, and outcomes may look impressive while leaving the researcher unable to investigate any of them deeply enough.
The practical challenge is to identify a scope large enough to sustain doctoral-level contribution but contained enough that the resulting claims can be supported rigorously within the conditions of the doctorate.
03 · What You Need to Know
Doctoral Ambition Is About the Contribution You Can Defend
Originality does not require maximum breadth
A dissertation needs to satisfy the originality requirements of the relevant doctoral degree, but originality can arise in many ways. A researcher might develop or refine an explanation, challenge an established assumption, identify important boundary conditions, produce new empirical evidence, advance a method, reinterpret existing material, or connect bodies of scholarship in a way that changes understanding of a problem.
None of these necessarily requires a geographically large, methodologically elaborate, or multi-study project.
A tightly specified question may actually make originality easier to demonstrate because the researcher can state precisely what uncertainty is being addressed and what changes once it is answered. This is one reason the quality of a dissertation question should be evaluated through its contribution and researchability rather than its apparent size.
Every expansion of scope creates an evidentiary obligation
Adding a population, context, variable, comparison, method, or time period does more than lengthen the proposal. It changes what evidence must be produced and what analytical work must be completed.
| Expansion |
Potential scholarly value |
Potential cost |
| Additional populations |
Tests variation or boundary conditions |
Recruitment, sampling, measurement equivalence, additional analysis |
| Additional sites or countries |
Examines contextual variation |
Access, coordination, comparability, ethics and logistical demands |
| Additional outcomes |
Provides a broader account of consequences |
Measurement burden, analytical complexity, fragmented interpretation |
| Additional methods |
May address complementary dimensions of the problem |
Training, collection, analysis, integration, methodological justification |
| Longitudinal design |
Can investigate development and temporal relationships |
Attrition, delayed results, repeated data collection, longer dependency chains |
| Additional studies |
Can build a cumulative argument |
Multiple protocols, datasets, analyses, manuscripts or chapters, and failure points |
The relevant question is not whether each addition would be interesting. Most of them probably would be. Ask whether the additional component materially strengthens the central doctoral contribution enough to justify the work and risk it creates.
A dissertation should contain depth that could not be achieved in a smaller project
Doctoral scope should not be confused with simply doing more of the same work. A larger sample or an extra study can strengthen a dissertation when it serves the research logic, but doctoral depth may instead come from theoretical development, demanding analysis, methodological innovation, engagement with competing explanations, or unusually careful examination of a difficult phenomenon.
A useful question is: What will the dissertation allow me to understand, explain, establish, interpret, or create that a substantially smaller project could not?
If the answer is merely “there will be more data,” reconsider where the doctoral contribution actually resides.
Scope should follow the contribution, not precede it
Students sometimes design a dissertation structurally before they have defined its intellectual core: three studies, several sites, mixed methods, perhaps an international comparison. The structure then begins determining the research question.
A stronger sequence is often the reverse. Identify the unresolved scholarly problem, clarify the intended contribution, determine what evidence would be required to support it, and then decide how many studies, methods, sites, or datasets are actually necessary.
For some dissertations, one substantial study can support the central contribution. Others genuinely require several connected studies because the argument depends on sequential evidence or distinct but complementary investigations.
Feasibility remains a doctoral criterion
A doctorate provides more room for ambitious inquiry than many smaller student projects, but it does not provide unlimited time or resources. A question can be intellectually excellent and still be poorly designed for a dissertation if answering it depends on conditions unlikely to hold.
Feasibility includes access to participants, datasets, archives, equipment, laboratories, field sites, organizations, computing resources, funding, methodological expertise, and supervision. Ethical and regulatory requirements may also shape what can realistically be completed.
Importantly, feasibility should include the entire research cycle. Data collection is only one part. Analysis, interpretation, writing, revision, examination preparation, publication requirements where applicable, and unexpected setbacks all consume time.
Some complexity creates knowledge; some merely creates dependencies
Imagine a dissertation that requires cooperation from six institutions, three participant groups, longitudinal recruitment, administrative records, interviews, survey data, and a specialist analytical technique. Any of those elements might be justified.
Together, however, they form a chain in which several independent things must succeed. If the intellectual contribution truly requires that architecture, the complexity may be warranted. If the same central claim could be supported with two institutions and fewer dependencies, the additional scale deserves scrutiny.
This distinction matters because completion risk is part of choosing a research question, even at doctoral level.
Ambition should be concentrated where failure is informative
There is an important difference between intellectual risk and logistical risk. Intellectual risk means investigating a genuine uncertainty whose answer may challenge your expectations. That is often desirable. Logistical risk means the dissertation cannot proceed because an organization withdraws, a specialist is unavailable, recruitment collapses, or a required dataset never arrives.
The first kind of risk can produce knowledge. The second may produce only a revised Gantt chart.
A well-scoped dissertation therefore does not eliminate uncertainty. It concentrates uncertainty in the research problem rather than unnecessarily placing it in the project's basic ability to operate.
Your claims provide a useful test of scope
Write down the strongest claim you hope the dissertation will support. Then ask what evidence that claim requires.
If you want to make claims across national systems, your research architecture must justify that level of generality. If you want to explain a mechanism, the design must provide evidence capable of examining that mechanism. If you want to establish change over time, the study must observe or otherwise credibly address temporal change.
Sometimes the correct response is to expand the project because the intended claim genuinely requires more evidence. At other times, the better response is to narrow the claim and question.
There is no universal dissertation-sized question
A laboratory-based doctorate, archival dissertation, theoretical dissertation, practice-based doctorate, computational project, qualitative ethnography, engineering design project, and article-based dissertation can have radically different research architectures.
Institutional regulations also vary. Some doctoral programs permit or encourage thesis-by-publication formats, while others expect a monograph. Disciplines differ in what constitutes originality and sufficient contribution.
Scope should therefore be calibrated against the actual standards under which the dissertation will be examined, not against generic assumptions about how many studies or chapters a doctorate should contain.
04 · A Practical Example
Making a Doctoral Question More Focused Without Making It Less Doctoral
Hypothetical Example
A dissertation on generative AI and higher education
A doctoral researcher initially proposes: How is generative AI transforming teaching, learning, assessment, academic integrity, and faculty practice across higher education?
The question certainly sounds important. It also contains several dissertation-sized problems. “Higher education” is extremely heterogeneous, “generative AI” covers multiple technologies and practices, and each listed outcome has a substantial literature of its own.
Identify the underlying scholarly problem Students can use generative AI to improve performance on academic tasks, but improved assisted performance may not necessarily indicate improved independent capability.
Specify the unresolved mechanism The researcher becomes interested in whether different patterns of AI-supported revision influence what students subsequently do without AI assistance.
Define the intended contribution The dissertation will seek to explain conditions under which AI-supported performance does or does not translate into subsequent independent performance.
Choose the necessary research architecture The researcher determines which populations, tasks, measurements, and studies are required to examine that mechanism rather than automatically retaining every originally proposed dimension.
Align the claim with the evidence The resulting question is bounded to the contexts actually investigated rather than making claims about higher education as a whole.
A resulting central question might ask: Under what conditions does generative AI-supported revision contribute to students' subsequent independent revision performance?
This question covers far less territory than the original one, but it may support a more substantial doctoral argument because the dissertation can investigate one consequential uncertainty in depth. Whether it is sufficient would still depend on the literature, theoretical framing, methods, evidence, and requirements of the doctoral program.
07 · A Quick Checklist
Check Whether Your Dissertation Has the Right Level of Ambition
Before committing to the scope, check:
State the original scholarly contribution the dissertation is intended to make.
Identify the evidence actually required to support that contribution.
Justify every major population, site, dataset, method, comparison, or study by the intellectual work it performs.
Verify essential access, recruitment routes, data availability, equipment, funding, and specialist support.
Estimate the complete workload, including analysis, integration, writing, revision, and realistic setbacks.
Identify which dependencies could prevent the central question from being answered.
Test whether a smaller research architecture could support essentially the same contribution.
Check that plausible null, contradictory, or unexpected findings would still leave an intellectually meaningful dissertation.
Verify the proposed scope against current institutional regulations and disciplinary expectations for the doctorate.