An accounting dissertation’s limitations are the things your chosen data and method genuinely could not show — not confessions of weak effort. The recurring ones are single-country data, a narrow time window, reliance on published figures you cannot verify independently, and a sample of listed companies that will not generalise to private or unlisted firms. State each one plainly, then say what it means for how far your conclusion can be trusted.
Why accounting dissertations have a distinctive set of limitations
Most UK undergraduate and taught master’s accounting dissertations are built on archival, secondary data — annual reports, Companies House filings, FAME or Orbis extracts, share-price series — rather than primary data collection with human participants. That changes what a limitation actually looks like compared with, say, a psychology or nursing dissertation. You are not usually worried about recruitment bias or informed consent; you are worried about data availability, comparability across accounting regimes, and whether published figures reflect genuine economic performance or the choices management made within what the standards allow.
A second distinctive feature is that accounting standards and regulation change, sometimes mid-study. If your data spans a period where a relevant standard was updated (a new IFRS, a change to the UK Corporate Governance Code, a shift in the small-companies audit exemption threshold), that is a real limitation on comparability across your sample period — not a footnote to skip.
Scope, delimitations and limitations: three different things
Markers frequently deduct marks when these three get muddled, so keep the distinction sharp:
- Scope is a decision you state in the introduction — what your dissertation covers and does not cover, by design (e.g. “this study examines FTSE 250 non-financial firms only”).
- Delimitations are the boundaries you chose deliberately, and could have chosen differently — a single country’s reporting regime, a single sector, a single five-year window — stated so the reader knows you narrowed the study on purpose, not by accident.
- Limitations are the constraints you did not choose — the ones built into the data source, the method, or practical access — that qualify how far your findings can be trusted or generalised.
A worked illustrative example: “This study is delimited to UK-listed firms in the retail sector between 2019 and 2024 (scope/delimitation). A limitation of this design is that FAME’s coverage of small-cap retailers is incomplete for this period, which may bias the sample towards larger, better-covered firms.”
Common limitation types for an accounting dissertation, with worked sentences
1. Data availability and database coverage
Illustrative: “Company-level ESG disclosure data was drawn from [named database], which has more complete coverage for large-cap than small-cap firms; the sample therefore over-represents larger companies, and findings may not generalise to SMEs.”
2. A single reporting regime or jurisdiction
Illustrative: “This study uses UK-reporting firms only, under UK-adopted IFRS; findings about discretionary accrual behaviour may not transfer to jurisdictions with different disclosure requirements or enforcement intensity.”
3. Reliance on published, audited figures
Illustrative: “This study assumes reported figures reflect audited, compliant accounts; it cannot detect undisclosed misstatement, and any conclusions about earnings quality are bounded by what external reporting reveals rather than internal management information.”
4. A time-bound sample crossing a standard change
Illustrative: “The sample period spans the transition to a revised leasing standard; pre- and post-transition balance sheet figures are not perfectly comparable, and this is treated as a limitation on the trend analysis in Chapter 4 rather than corrected for statistically.”
5. Secondary data cannot capture managerial intent
Illustrative: “Because this study relies on published financial statements rather than interviews with preparers, it can describe patterns in reported figures but cannot establish why a specific accounting choice was made by management.”
6. Sample size and survey response in management-accounting designs
Illustrative: “Where this study surveys management accountants directly, a modest response rate limits statistical power and may over-represent respondents willing to discuss their own reporting practices, a form of self-selection bias worth naming explicitly rather than ignoring.”
7. Currency, inflation and cross-period comparability
Illustrative: “Nominal revenue and profit figures across the sample period are not adjusted for inflation; year-on-year comparisons should therefore be read as indicative of reported growth rather than real economic growth, and this is noted wherever trend figures are discussed.” Where a sample spans several years of above-target inflation, naming this limitation explicitly, rather than presenting nominal figures as if they were directly comparable, is the kind of detail that separates a competent limitations section from a thin one.
Delimitations: what you chose to leave out
State delimitations as decisions, not apologies. A worked example: “This dissertation does not compare UK findings against a second jurisdiction; a cross-country comparison was considered but was not feasible within the word count and timeline available, and is proposed instead as a direction for future research.” That single sentence does three jobs a marker rewards: it names the boundary, explains it was a choice, and converts it into a recommendation rather than leaving it as an unexplained gap.
Turning limitations into recommendations for future research
Each limitation you name is also, almost automatically, a recommendation once you flip its framing: a single-country limitation becomes “future research could replicate this analysis using a comparable jurisdiction”; a database-coverage limitation becomes “future research with access to a more complete SME database could test whether these findings hold for smaller firms.” Pairing each limitation with its recommendation in the same paragraph, rather than listing limitations and recommendations separately, reads as a more coherent discussion chapter.
A worked recommendations paragraph, built from three of the limitations above: “Three extensions would strengthen this line of research. First, replicating the analysis with a dataset offering fuller SME coverage would test whether the patterns found here hold beyond large, well-covered firms. Second, extending the sample across a second jurisdiction with a comparable reporting regime would test whether the findings are UK-specific or reflect a broader pattern under IFRS. Third, combining the archival analysis with a small number of interviews with preparers would help explain why, not just whether, particular accounting choices were made.” Notice each sentence names a specific extension tied to a specific limitation, rather than a vague call for “further research.”
Mistakes examiners flag most in this section
- Apologising instead of qualifying. “Unfortunately, this study was limited by time constraints” tells a marker nothing useful. Naming the specific data or method constraint, and what it means for your conclusion, does.
- Generic limitations that would apply to any dissertation. “A larger sample would have been better” is true of almost every study and reads as filler. Tie the limitation to something specific about your data source or design.
- Confusing delimitations with limitations. A choice you made on purpose (a single sector, a single country) is a delimitation, stated confidently in the introduction — not a limitation to apologise for in the discussion.
- Undermining your own findings entirely. A limitations section should qualify the scope of your claims, not suggest the whole study is worthless — markers read the latter as a lack of confidence in your own work, not intellectual honesty.
- Listing limitations with no link to validity. Every limitation should say, in the same sentence or the next one, what it means for how far the reader can trust or generalise the finding.
Where this fits with your topic and data sources
Limitations are easiest to write well when you have already been specific about your data sources and sample — if you have read our guide to population, sample and data access for an accounting dissertation, most of the limitation types above should already sound familiar, since they follow directly from how UK accounting dissertations typically source their data. If you are still finalising a topic, our list of accounting dissertation topics with named UK datasets shows which data sources pair naturally with which research questions, and our comparison of data analysis software for an accounting dissertation covers a related, tool-specific limitation: what your chosen software can and cannot test. For the general, cross-subject version of the scope-versus-limitations distinction, see our piece on the scope and limitations of a dissertation.
Writing this section with Tesify
Tesify helps you draft a limitations section that names the actual constraints of your specific data source and sample, rather than a generic template, and pairs each one with a matching recommendation for future research. Over 9,000 students have used Tesify across more than 15,000 dissertation chapters, and every dissertation on the platform is 100% written by you.
Frequently asked questions
What is the difference between a limitation and a delimitation in an accounting dissertation?
A delimitation is a boundary you chose on purpose — a single country, sector or time window — and belongs confidently in your introduction. A limitation is a constraint you did not choose, built into your data source or method, that qualifies how far your findings can be trusted or generalised, and belongs in your discussion chapter.
Should I apologise for my dissertation’s limitations?
No. State the specific constraint and what it means for your conclusion, in neutral, factual language. A limitations section is meant to show critical awareness of your own study’s scope, not to undermine the work you have done.
How many limitations should I include?
Most accounting dissertations do well with three to five substantive, specific limitations rather than a long generic list. Depth on a few real constraints reads better than breadth across ones that do not meaningfully affect your specific study.
Can data availability be a legitimate limitation?
Yes, and for archival accounting research it is one of the most common and most defensible limitations, provided you name the specific database and the specific coverage gap rather than saying only “data was limited.”
Is relying on published financial statements a limitation?
Yes — it means your study can describe patterns in what firms reported but cannot verify managerial intent or detect information not disclosed externally. Naming this explicitly, rather than assuming published figures capture the full picture, is exactly the kind of critical awareness markers reward.
Do limitations only belong in the discussion chapter?
They are normally concentrated there, but scope and delimitations should already be visible in your introduction, so a marker is not surprised by boundaries you reveal for the first time at the end.
What if a change in accounting standards affects my sample period?
Name it directly as a limitation on comparability across your sample period, rather than silently treating pre- and post-change figures as equivalent. You do not need to correct for it statistically in an undergraduate dissertation — stating the constraint clearly is normally sufficient.
Can I recommend future research that just repeats my study with more data?
Yes, provided you are specific — “replicate this analysis with a larger, more complete SME sample once database access improves” is a genuine recommendation; “more research is needed” on its own is not.
Should I list limitations and recommendations in separate sections?
Either structure is normally acceptable — check your department’s template — but pairing each limitation with its corresponding recommendation in the same paragraph usually reads as more coherent than two disconnected lists.
Does a small sample of companies count as a limitation in accounting research?
Yes, if it restricts how far you can generalise your findings — but be specific about what it restricts. A sample of 30 FTSE 250 firms is not “small” in the way a sample of 30 survey respondents is; the relevant limitation is usually that the sample represents large, well-governed firms and may not generalise to smaller or unlisted companies, rather than the raw count being too low.