Writing a dissertation is not just about collecting reliable sources and presenting your findings. You also need to build an argument that is clear, balanced, and supported by evidence. This is where logical fallacies can quietly weaken your work. A conclusion may sound convincing while relying on assumptions, incomplete evidence, exaggerated claims, or a misunderstanding of another researcher's position.

The good news is that you do not need to memorise every type of fallacy to improve your reasoning. You simply need to learn how to question the connection between your evidence and your conclusions. In this guide, I will explain the most common logical fallacies in dissertation writing, show you how to recognise them, and share practical ways to strengthen your academic arguments.

What counts as a logical fallacy?

A logical fallacy is a problem with reasoning. It occurs when an argument looks persuasive but the reasoning connecting its evidence to its conclusion does not hold up properly.

That does not necessarily mean the conclusion itself is false.

Imagine a researcher writes:

Students who attended optional revision sessions achieved higher grades. Therefore, attending revision sessions causes higher grades.

The first statement might be completely accurate. The second is another matter. Perhaps students who attended the sessions were already more motivated. Perhaps they had more free time. Perhaps they were more confident academically.

The evidence may show an association without establishing causation.

This is why I think it is more useful to treat fallacies as argument-checking problems rather than a list of labels to memorise. Informal logic focuses on evaluating arguments as they occur in ordinary language and recognises the importance of premises, inference and conclusions. It also notes that there is no universally agreed taxonomy of fallacies. (Stanford Encyclopedia of Philosophy)

For dissertation writing, the practical question is simple:

Does the evidence I have presented actually justify the claim I am making?

Why fallacies creep into dissertations

There is a good reason this happens.

When you have spent months researching a topic, you naturally become familiar with your preferred explanation. By the time you reach the discussion chapter, you may be so focused on explaining your findings that you stop questioning the assumptions underneath them.

There is also a structural problem. A dissertation asks you to move repeatedly between evidence and interpretation. Every time you make that move, there is an opportunity to overstate what the evidence means.

A useful habit is to separate every important paragraph into three parts:

• Evidence: What did the research actually find?

• Interpretation: What do I think those findings mean?

• Conclusion: What can reasonably be claimed from them?

Keeping those stages separate makes weak reasoning much easier to spot.

7 logical fallacies worth checking for

You do not need to search your dissertation for every fallacy ever identified. Start with the ones most likely to appear when you are reviewing literature, interpreting findings or defending a conclusion.

1. Hasty generalisation

This happens when you make a broad claim from evidence that is too narrow to support it.

For example:

Ten undergraduate students preferred recorded lectures. Therefore, university students prefer recorded lectures.

The issue is not necessarily the number ten. A small sample may be entirely appropriate for qualitative research. The problem is extending the finding beyond the population and circumstances actually investigated.

A better version would acknowledge the limits:

Participants in this study generally expressed a preference for recorded lectures, although the small sample limits the extent to which this finding can be generalised.

That distinction matters.

Your conclusion should match your sample, methodology and research question, not the conclusion you wish your study had produced.

2. Straw man reasoning

A straw man appears when you simplify or distort another researcher's position before criticising it.

Suppose a researcher argues that artificial intelligence can be useful for certain academic tasks when students use it responsibly.

You write:

Researchers who support AI believe students should allow AI to write their academic work.

That is not a fair representation of the original argument.

The easiest way to avoid this problem is to read the source carefully enough to explain its position before challenging it.

When reviewing a study, ask yourself:

• What exactly is the researcher claiming?

• What evidence supports the claim?

• What assumptions does the argument make?

• What limitation can I identify?

• Am I criticising the actual position or an exaggerated version of it?

That last question is particularly valuable when writing a critical literature review.

3. False dilemma

A false dilemma reduces a complicated issue to two choices when other possibilities exist.

For example:

Universities must either embrace online learning or accept that their teaching methods are outdated.

That sounds forceful. It is also unnecessarily simplistic.

A university could use a combination of face-to-face teaching, online resources, hybrid delivery and discipline-specific approaches. The best option may differ according to the subject, students and learning objectives.

Be particularly cautious when your writing relies heavily on words such as always, never, only, either, must or completely.

They are not automatically signs of bad reasoning. But they should make you stop and ask whether the evidence really supports such certainty.

4. Confusing correlation with causation

This is one of the most important problems to look for in a dissertation.

Suppose your research finds that employees who work remotely report higher job satisfaction.

It would be reasonable to report an association between remote working and job satisfaction if your methodology supports that conclusion.

It is a much stronger claim to say:

Remote working causes higher job satisfaction.

You would need an appropriate research design and analysis to justify a causal interpretation.

This distinction is taken seriously in scholarly publishing. JAMA's current author guidance, for example, advises that causal language should be restricted to study designs capable of supporting causal conclusions and recommends association or correlation language where appropriate. (JAMA Network)

So before writing words such as causes, leads to, results in or produces, check your methodology.

If your study cannot establish causation, terms such as is associated with, is related to, may contribute to or is consistent with may be more accurate.

Do not weaken your dissertation by pretending to know more than your data can tell you.

5. Cherry-picking evidence

Cherry-picking is particularly tempting when your literature review contains conflicting findings.

Imagine you identify ten relevant studies:

• four report positive results,

• three find mixed results, and

• three find little or no significant relationship.

If you discuss only the four positive studies, your reader receives a distorted impression of the literature.

You do not have to pretend every study is equally persuasive. Instead, explain why the findings differ.

For example:

Evidence for the relationship remains mixed. While several studies report positive associations, others identify weaker or statistically insignificant effects. Differences in sample characteristics and measurement methods may partly explain these inconsistencies.

That is much stronger academically because you are showing the reader that you have considered the wider evidence.

Transparent reporting matters for the same reason. The EQUATOR Network maintains a large library of reporting guidelines covering different study designs, including CONSORT, STROBE, PRISMA, SRQR and COREQ. (EQUATOR Network)

The exact guideline you need depends on your research design, but the underlying principle is useful across disciplines: give readers enough information to understand how you reached your conclusions.

6. Appeal to authority

A famous researcher can be wrong.

So can a professor, government official, industry leader or bestselling author.

Citing an expert establishes that a particular person or organisation has made an argument. It does not automatically establish that the argument is correct.

Compare these two sentences:

Professor Smith argues that the policy will improve productivity, so the policy will improve productivity.

and:

Smith argues that the policy could improve productivity, based on evidence from three longitudinal studies.

The second sentence gives the reader something to evaluate.

In a dissertation, your citations should support your reasoning rather than replace it.

When you cite an authoritative source, explain why the source matters and what the evidence actually demonstrates.

7. Ad hominem arguments

An ad hominem argument attacks the person rather than dealing with their evidence or reasoning.

For example:

This study should be rejected because its authors work for a company that benefits from the results.

A researcher's financial interest may be relevant. But it does not automatically make the research wrong.

Instead, examine the possible implications:

The study was funded by an organisation with a commercial interest in the intervention, which should be considered when assessing the possibility of sponsorship bias.

You have raised a legitimate concern without pretending that the funding source settles the question.

That difference is important in a dissertation. Critical analysis is not the same thing as criticism.

A practical way to find fallacies in your own writing

Once you know what to look for, the next challenge is finding these problems in 50, 100 or 200 pages of your own work.

I recommend doing a separate argument audit rather than relying only on normal proofreading.

Step 1: Mark your strongest claims

Search through your discussion and conclusion chapters for statements containing words such as:

• demonstrates

• proves

• causes

• clearly

• therefore

• must

• always

• never

• significantly

• confirms

None of these words is automatically problematic. They simply identify places where you are making a strong claim.

Step 2: Ask what supports each claim

For every major conclusion, identify the exact evidence behind it.

If you cannot answer that quickly, investigate.

Sometimes you will discover that the statement is based on a single source. Sometimes you will realise that you have accidentally turned your own interpretation into a fact.

Step 3: Follow the reasoning between evidence and conclusion

This is the step students often skip.

Do not ask only, "Do I have a citation?"

Ask:

Does this citation support the sentence I have attached it to?

A source can support one part of a paragraph without supporting everything that follows.

For example:

Several studies have reported an association between X and Y (Author, 2024). Therefore, X causes Y.

The citation may support the first sentence while doing nothing to establish the second.

Step 4: Look for competing explanations

Try to argue against yourself.

If you found a relationship between two variables, ask what else might explain it.

If participants reported a particular experience, ask whether another interpretation is possible.

If previous studies disagree, ask whether differences in methodology, population or context could explain the disagreement.

This process can feel uncomfortable, especially when you have spent a long time developing your argument. But it often produces your best discussion writing.

Step 5: Reduce the claim when necessary

Sometimes the evidence is good, but your wording is too ambitious.

Suppose you originally write:

The findings demonstrate that flexible working improves employee wellbeing.

After reviewing your methodology, you realise that your study only measured an association.

Change it to:

The findings indicate an association between flexible working and employee wellbeing.

You have not made your research weaker. You have made the conclusion more defensible.

Match your reasoning to your research method

Your methodology places boundaries around what you can reasonably conclude.

A qualitative interview study might provide rich insight into how participants understand an issue. That does not automatically establish how common the experience is across an entire population.

A cross-sectional survey might identify relationships between variables. That does not necessarily establish what caused those relationships.

A systematic review can synthesise existing evidence, but its conclusions depend partly on the quality and characteristics of the studies included.

If you are conducting a systematic review, PRISMA 2020 provides a structured reporting framework covering areas such as eligibility criteria, information sources, search strategies, study selection and synthesis. The current PRISMA materials include both a checklist and expanded guidance. (PRISMA statement)

The broader lesson is straightforward: your conclusion should never outrun your methodology.

Build counterarguments into your discussion

One of the easiest ways to make dissertation writing more analytical is to anticipate reasonable objections.

Instead of writing:

The results show that the intervention is effective.

Try asking:

What would someone who disagrees with me say?

You might then write:

Although the findings support the effectiveness of the intervention, the absence of a long-term follow-up period means that the durability of the observed effect remains uncertain.

That is much more useful to the reader.

You are not simply presenting your preferred interpretation. You are showing that you understand where the argument is vulnerable.

This is also closer to how strong academic argumentation works. Informal logic treats arguments as relationships among premises, inferences and conclusions and focuses on assessing whether the reasoning provides adequate support. (Stanford Encyclopedia of Philosophy)

A five-question check before submission

When you finish a chapter, put your normal proofreading aside for a few minutes and run through these questions:

1. What exactly am I claiming?

Can I state my main argument in one sentence?

2. What evidence supports it?

Can I identify the relevant data or sources?

3. Does the evidence support this precise claim?

Or have I quietly expanded the conclusion?

4. What evidence could challenge my interpretation?

Have I acknowledged credible alternatives?

5. Am I more certain than my methodology allows?

If the answer is yes, revise the wording.

This simple check can reveal problems that spelling and grammar software will never find.

What if you need help reviewing your argument?

Sometimes the problem is not knowing what a fallacy is. It is being too close to your own dissertation to notice one.

A second reader can be useful here, particularly when you ask them to focus specifically on the logic of the argument rather than proofreading every sentence.

If you are working on a larger research project and need additional academic support, you can also look at capstone project writing help online. Used appropriately, outside support can be useful for feedback on structure, research direction, editing or understanding difficult sections. Your own university's academic-integrity rules should always determine what kind of assistance is permitted.

The aim should be to improve your understanding of the work, not to hand responsibility for your research over to someone else.

Strong dissertations do not pretend to have perfect evidence

I think this is the most important point to take away.

A convincing dissertation is not one in which every paragraph sounds completely certain. Research rarely gives us that luxury.

A convincing dissertation shows that you understand what your evidence can support, where its limitations lie and why you have chosen one interpretation over plausible alternatives.

Current ICMJE recommendations, updated in January 2026, likewise emphasise responsible reporting and appropriate interpretation of research findings. (ICMJE)

So when you edit your dissertation, do not simply ask whether each paragraph sounds academic.

Ask whether it is fair to the evidence.

If you have generalised too far, narrow the claim. If you have ignored contradictory evidence, address it. If you have confused correlation with causation, correct the language. If you have misrepresented another researcher's position, return to the original source.

And if you cannot explain the reasoning that connects your evidence to your conclusion, stop and work on that connection before polishing the prose.

That is where better dissertation writing begins: not with more complicated vocabulary, but with better reasoning.

Author's note

This article draws on established work in informal logic and current research-reporting guidance, including the Stanford Encyclopedia of Philosophy, ICMJE, EQUATOR Network and PRISMA 2020. Reporting requirements vary by discipline and institution, so dissertation writers should also follow the methodology, style guide and academic-integrity requirements specified by their university.