Paul Scott
Acknowledging a weakness in your dissertation doesn’t cancel out its contribution. A strong dissertation limitations section explains what your evidence can support, where uncertainty remains, and how future research could address it.
You don’t need to apologise for every research decision or produce a catalogue of possible problems. Start with the constraints that change how readers should interpret your findings.
Key Takeaways
- Distinguish limitations, which constrain your conclusions, from delimitations, which define your chosen scope.
- Explain each limitation’s effect on the findings, then identify a suitable response or future research direction.
- Discuss limitations openly, including when your study is strong overall, and keep your conclusions within the evidence.
Limitations and Delimitations Are Different
Limitations Constrain What You Can Conclude
Research limitations are constraints you cannot fully control within your methodology and design. They include weaknesses inherent in your approach, such as imperfect recall in retrospective interviews.
Some also arise from deliberate choices in your approach. Choosing self-reported measures doesn’t remove their potential for inaccurate responses.
Ross and Zaidi’s article, “Limited by our limitations”, links these issues to their effects on research outcomes and conclusions. Every study has them. Concealing them can lead reviewers to reject a paper, so discuss them openly.
Delimitations Define Your Chosen Scope
Delimitations are boundaries you control: the population, location, period, concepts or evidence included or excluded, and your reasons.
For example, studying nurses in Oregon sets a geographical boundary. You can’t automatically generalise those findings to nurses in Arkansas, where working contexts may differ.
These delimitations are deliberate choices that define the study’s scope. Restricted generalisation is one consequence of those boundaries. A focused dissertation research question makes them easier to explain without presenting a manageable scope as a mistake.
Identify the Limitations That Matter
Start with your research questions and consider how your research design limits the answers you can support. Laerd groups research limitations into four types.
These categories help distinguish different problems.
| Type | Example |
|---|---|
| Inability to answer the research questions | A cross-sectional survey cannot establish change over time. |
| Theoretical or conceptual problems | Attendance measures only one aspect of student engagement. |
| Research strategy limitations | Recruitment through one institution restricts coverage. |
| Research quality problems | Missing records reduce confidence in the analysis. |
Choose limitations that affect your conclusions, rather than filling the section with every category. Your research strategy also affects which conclusions the evidence can support.
Measurement and Data Collection
Common limitations include self-reported data, lengthy surveys that encourage fatigue, and closed questionnaires without open-ended responses. Each affects a different aspect of the evidence.
Self-reported data can introduce recall or social desirability bias. Closed questions can miss explanations outside the available options. Secondary data collected many years ago may also poorly reflect current practice.
Describe the methodological effects of time constraints, such as a short observation period or missing follow-up, rather than writing “I lacked time”.
Internal and External Validity
Internal validity concerns whether the study’s conduct and analysis support trustworthy conclusions, particularly causal interpretations. Confounding occurs when another variable helps explain an observed relationship.
External validity concerns whether findings apply beyond the study population or setting. A representative group supports generalisation, whilst convenience recruitment can restrict it.
Keep these distinctions clear: adding participants won’t automatically fix a flawed measure or establish causation.
Where to Put Your Dissertation Limitations
Explain methodological constraints in your methodology chapter, alongside the decisions that created them. For example, justify self-reported attendance when describing your questionnaire.
Then discuss their actual implications in the discussion section, where readers encounter your findings. Avoid copying the same paragraph across chapters.
Limitations concerning the wider nature or scope of the research can appear in the conclusion, linked to future research. However, recommendations shouldn’t replace an explicit explanation of the current study’s restrictions.
A major limitation may also belong in your abstract if omitting it would encourage an unsupported interpretation. Our guidance on writing a clear dissertation abstract explains how to keep the summary within the evidence.
How to Write Dissertation Limitations Clearly
For each substantial issue, connect the methodological constraint to a specific claim. Delimitations describe the study’s chosen scope; research limitations explain what the evidence cannot support. Laerd’s guidance on structuring a limitations section provides a useful framework for organising that discussion.
Use this sequence:
- Name the issue precisely, including the relevant feature of your sample, method or evidence.
- Explain how it could affect the results, including what remains uncertain or unsupported.
- Describe any mitigation you actually used, then offer future research suggestions to address the remaining problem.
A reusable sentence pattern is: “Because [constraint], the findings cannot establish [claim]. Although [actual mitigation], [remaining uncertainty] persists. Future research could [targeted response].”
Don’t claim that an action removed bias unless your evidence supports that conclusion. Use “may” for an uncertain effect, but direct language for a definite restriction.
“The questionnaire may underestimate attendance” expresses uncertainty about measurement. “The design cannot establish causation” states a clear methodological boundary.
Finally, prioritise the issues with the greatest effect on your main conclusions. A detailed explanation of three relevant constraints is stronger than ten unexplained labels.
Worked Example: A Low Survey Response
For a planned target of 100 at a university of 20,000 students, 30 responses raise concerns about coverage and precision. A complete model section could read:
The survey received 30 responses against a planned target of 100 from a university population of 20,000 students. This low participant count limits the precision of estimates, and the narrow respondent group limits external validity. Respondents may differ from non-respondents in ways relevant to the research question, creating a risk of selection bias. The findings should therefore describe respondents’ reported experiences rather than establish university-wide prevalence.
Increasing participant numbers alone would not necessarily remove selection bias. Future research could recruit across programmes and years of study, then compare respondents with available population characteristics to assess possible non-response effects. These restrictions limit generalisation, but the survey provides evidence about participating students and identifies questions for broader investigation.
The distinction matters: sample size and selection bias are separate problems. Thirty responses can produce uncertain estimates, whilst selective participation can distort them.
Also, don’t calculate a response rate from the planned target unless you know how many people received an invitation. A recruitment target and the number approached are different denominators.
The closing sentence preserves a bounded contribution without dismissing the limitation.
Examples for Other Research Designs
Qualitative Interviews and Transferability
For interviews conducted at one university, you could write:
Recruitment within one institution creates constraints on the range of contexts represented. These accounts cannot establish how common reported experiences are across universities. Detailed descriptions of participants and the setting help readers assess transferability. Future research could use interviews in contrasting institutions to examine whether similar interpretations occur elsewhere.
Qualitative research doesn’t usually seek population-wide prevalence claims through representative sampling. Instead, readers assess whether interpretations might transfer to sufficiently similar contexts.
A small participant group isn’t automatically a weakness. Explain whether it restricted relevant perspectives or the depth needed to answer your question, rather than apologising for its size alone.
Secondary Data and Historical Context
For workplace data collected before widespread hybrid working, a suitable paragraph is:
“The dataset predates widespread hybrid arrangements, so its measures may not capture current working practices. The analysis can address relationships within the recorded period but cannot establish their present-day extent. Future research using recent data and explicit measures of hybrid working could assess whether those relationships persist.”
Here, the limitation concerns the match between the data and the claim. Older evidence can still answer a historical question well.
However, missing variables cannot be recovered through interpretation alone. State which relevant concept the dataset doesn’t measure and how that restricts your analysis.
Match the Depth to Your Degree
An undergraduate dissertation can make a modest contribution within a narrow scope. Its limitations section should explain how the study’s question, method and conclusions relate to the constraints that shape its contribution.
Doctoral work needs a deeper account of how limitations affect its original contribution. That may include theoretical assumptions, competing explanations, measurement choices and the boundaries of broader claims.
Neither level needs a fixed number of issues. Give each one space according to its significance.
A feasible dissertation topic reduces avoidable problems before data collection. Yet even a well-planned project requires an honest account of what remains unresolved.
Avoid Apologies and Unsupported Reassurance
Don’t dismiss research limitations with unsupported reassurance such as “the sample was small, but this wasn’t important”. Explain which estimates or comparisons the sample can support instead.
Likewise, don’t claim your work has no limitations. Strong procedures can reduce a problem, but they can’t eliminate it.
Avoid treating “lack of time” as a standalone limitation. If timing prevented follow-up, explain that the study can’t assess whether the observed pattern persisted.
End with a proportionate contribution and a targeted next step. Laerd’s future research suggestions distinguish changes to questions, concepts, research strategy and research quality. Choose the response that addresses your actual constraint.
Conclusion: Keep Your Claims Within the Evidence
A convincing limitations section explains the constraints on your findings. Name the issue, explain its effect, and identify a response that addresses the remaining uncertainty.
Your dissertation doesn’t lose its contribution because it has limits. Accurate claims make that contribution easier to understand and defend.
Frequently Asked Questions About Dissertation Limitations
How Long Should a Limitations Section Be?
There’s no universal length. Give the principal limitations enough space to explain their consequences and possible responses. Minor issues need less attention, especially when they don’t affect your central argument.
Does a Small Sample Make My Study Invalid?
No. Its implications depend on your question, sampling strategy and analysis. A small quantitative sample may reduce precision or statistical power. In qualitative work, assess whether it provides sufficient depth and relevant perspectives for the enquiry.
Should I Mention Limitations If My Findings Are Strong?
Yes. Consistent findings don’t remove sampling, measurement or design constraints. State what the evidence supports and what it cannot establish. This applies even when results match your hypothesis or agree with previous research.