Why the Limitations Section Scares Students (and Shouldn’t)
No section of a dissertation causes more unnecessary anxiety than the limitations. Students read it as an invitation to confess everything wrong with their study, and then either write a page-long apology or skip it almost entirely, hoping the examiner will not notice. Both approaches lose marks.
The truth is the opposite of what students fear. Examiners do not expect a perfect study; at undergraduate and master’s level, a perfect study is not possible, and they know it. What they mark is whether you understand the boundaries of your own work. A limitations section is not a confession. It is evidence of research competence: proof that you know what your findings can and cannot support, and that you will not claim more than your data allows. A student who names their limits precisely looks more trustworthy, not less.
Where the Section Goes and How Long It Should Be
Placement varies, and your department handbook is the final word. In most UK dissertations, limitations sit inside the discussion chapter, after the interpretation of findings and before the suggestions for future research — our dissertation discussion chapter guide follows that structure. The content is the same wherever it sits.
Length is where students most often go wrong. For a 10,000–15,000 word dissertation, the limitations section should run roughly 300–500 words: two or three limitations, each handled in a short paragraph. That is enough to show judgement and short enough to show confidence. A limitations section longer than your interpretation section sends a different message — that you have lost faith in your own findings — and examiners read it that way. If you have written a page of limitations, you have almost certainly listed things that are not really limitations at all, which brings us to the first rule.
The Rule: Name Limits That Bound Your Claims
A genuine limitation is something about your study that restricts what you can legitimately claim. Everything else — the things you wish you had done, the constraints every student faces, the features of your design that were deliberate choices — does not belong here, or belongs in one sentence at most.
Use this test for every candidate limitation: does it change what a reader should conclude from my findings? If a limitation means your results might not apply beyond your sample, it passes. If it means your measure captured only part of a concept, it passes. If it is simply “I would have liked a bigger sample” with no consequence attached, it fails the test until you state the consequence: a smaller sample means wider uncertainty around the effect and less power to detect small differences.
Three categories cover nearly every honest limitation at dissertation level.
1. Sample limitations
Who you studied, and how many of them. A sample of 30 volunteers from one university cannot support claims about all students everywhere, and self-selection matters: volunteers are usually more motivated or more interested in the topic than the people who declined. “Participants were recruited from a single university, so the findings may not generalise to institutions with different student populations” is a complete limitation. “The sample was small” on its own is not; it is a complaint waiting for its consequence.
2. Method and measurement limitations
What your design could not capture. A survey can show associations but not explain why they exist. A short interview study can explore experience in depth but cannot estimate how common that experience is. Self-report measures carry social desirability bias — people overstate good habits and understate bad ones. A questionnaire validated on one population may behave differently on yours. Name the design feature, say what it prevents you from claiming, and stop.
3. Researcher position and context limitations
Your own role in the study, and the conditions around it. If you researched your own institution, your classmates, or a placement setting where you were known, that relationship may have shaped what participants told you. If data were collected during exams, a pandemic, or a placement block, timing may have affected responses. These are real limits, and naming them shows the examiner you thought about your study as a human encounter, not just a procedure.
Examples: Weak vs Strong
The difference between a limitation that earns marks and one that costs them is usually specificity and consequence. Compare these pairs.
Weak: “This study had a small sample size due to time constraints, which is a major weakness.”
Why it fails: it names an excuse (“time constraints”) instead of a limitation, it exaggerates (“major weakness”), and it never says what the sample size actually prevents the reader from concluding.
Strong: “The sample of 32 participants was drawn from a single university through convenience sampling. The findings therefore describe this group rather than the wider student population, and the effect sizes reported should be treated as indicative rather than precise estimates.”
Why it works: it states the facts, names the consequence for generalisability, and calibrates the claim — without apologising.
Weak: “The study was limited because only questionnaires were used and interviews would have been better.” It criticises a deliberate design choice as if it were an accident, and implies a different study rather than bounding this one.
Strong: “The questionnaire design captured the strength of the association between workshop attendance and confidence, but it cannot explain the mechanism behind it. Interview data would be needed to establish whether peer contact, taught content, or both drove the effect.”
Why it works: it states exactly what the method cannot do, and turns the limit into a bridge to future research.
Weak: “There may have been researcher bias because the researcher knew some participants.”
Why it fails: “may have been” hedges the fact away, and no mitigation or consequence is given.
Strong: “The researcher had previously taught some participants, which may have encouraged socially desirable responses in interviews. This was managed by anonymising transcripts before analysis and by using a standardised interview schedule, but the possibility of response bias cannot be fully excluded.”
Why it works: it names the risk honestly, states what was done about it, and admits the residual uncertainty. That is what research competence looks like on the page.
The Mistakes That Cost Marks
Apologising instead of analysing. “Unfortunately, this small, flawed study attempted to…” Stop. You did the work. State limits in neutral language: “The study was limited by…” is fine; “Sadly, due to my inexperience…” is not. Examiners mark what is there, not what you wish were there.
Listing every conceivable weakness. A limitations section that runs to eight items reads as a lack of judgement, not thoroughness. Pick the two or three limits that genuinely bound your conclusions, and leave minor issues out entirely.
Confusing limitations with delimitations. Delimitations are boundaries you chose — your age range, your single case study — and they belong in the methodology, where you justify them. The limitations section is for their consequences, not for re-justifying the design.
Claiming “time constraints” as a limitation. Every student has time constraints. Naming them explains nothing about your findings and reads as an excuse. If time genuinely shaped the study — for example, a follow-up measure had to be dropped — name the consequence (no longitudinal data), not the cause.
Introducing limitations that contradict your results chapter. If your results chapter reports a finding confidently and your limitations section then says the data cannot support it, the examiner will wonder which chapter to believe. Limits should calibrate claims, not cancel them.
Turning Limits Into Future Research
The strongest way to end a limitations section is to connect each major limit to a specific suggestion for future research. A limitation left hanging sounds like a flaw; the same limitation followed by “Future research could test this mechanism with a larger, multi-site sample” sounds like a research programme. Keep suggestions specific and tied to your study — a population your sample missed, a method your design could not use, a mechanism your data could not test. Vague wishes such as “more research is needed” say nothing and should be deleted wherever they appear.
The University of Manchester’s Academic Phrasebank is genuinely useful here: it collects standard phrasing for acknowledging limitations, so you can state limits in calm, conventional academic language instead of inventing apologetic wording under pressure. Scribbr’s guide to writing a results section is also worth a final check, since limitations only make sense against findings reported cleanly and without overclaiming.
A Template You Can Adapt
Use this shape, in your own words, and fill it with your study’s facts: “This study has three main limitations. First, [sample limit] — [consequence for generalisability]. Second, [method limit] — [what it prevents you from claiming]. Third, [position or context limit] — [what was done to manage it and what uncertainty remains].” If your section reads like a list of excuses, rewrite each item until it states a fact and a consequence.
Limitations are the part of a dissertation where honesty is directly rewarded. Name the real limits, state what they mean for your claims, and move on with confidence. If the discussion chapter is the section standing between you and submission, you do not have to wrestle it alone — The Pen Master pairs you with subject-matched writers who help students interpret findings and write examiner-ready discussion chapters every week. See our services, message us on WhatsApp at +49 1521 3962850, or send your brief through the contact form on thepenmaster.com, and we will take it from there.