Critical evaluation in academic writing
The most common feedback in UK marking, and the one most students find hardest to act on. Here is what it means in practice.
"Needs more critical analysis" appears on more UK feedback sheets than any other comment, and it is nearly useless as instruction. Students read it, agree, and have no idea what to change.
The gap is that criticality sounds like an attitude. It is not. It is a set of specific questions you ask of evidence, and the answers go in your paragraphs.
What it is not
- Not being negative. Concluding a study is sound is a critical judgement if you say why.
- Not disagreeing with everything. Contrarianism without evidence scores no better than acceptance without evidence.
- Not more sources. Ten sources described is less critical than three evaluated.
- Not longer words. Vocabulary has nothing to do with it.
The five questions
Ask these of any source. You can answer all of them from an abstract and a method section, without being an expert in the field.
1. How was this produced?
Sample size, method, setting. A finding from twelve interviews and one from a 4,000-person cohort support very different claims, and neither is automatically better: it depends what you are arguing.
2. How far does it generalise?
Who was studied, and who was not? A trial in a private clinic tells you less about NHS practice than its abstract implies.
3. When, and does that matter?
A 2011 study of social media behaviour describes a different internet. In other fields a 1970s finding is still current. Knowing which applies to your topic is itself an act of judgement.
4. Does it agree with other work?
Disagreement between sources is the most productive thing in a literature. Most students hide it; strong writers make it the centre of the paragraph.
5. Who produced it, and does that shape it?
Funding, institutional position and publication venue. Not accusation, just context. Industry funding does not invalidate a study, but it is worth a clause.
What it looks like on the page
Descriptive
Ahmed (2024) found that a school-based intervention reduced anxiety scores by 18%. This suggests school interventions are effective.
Critical
Ahmed (2024) reports an 18% reduction in anxiety scores following a school-based intervention. The effect is substantial, though it was measured immediately after the programme ended and no follow-up was conducted, so it is unclear whether the change persists. Brown (2023), using a six-month follow-up with a comparable cohort, found the effect had largely disappeared. Taken together, the evidence supports short-term benefit rather than durable change.
Note what the second version does. It accepts the finding, identifies a specific limitation, brings in a second source that speaks to that limitation, and reaches a conclusion neither source states alone. That last move is the one that earns the mark.
A sentence pattern that helps
When you are stuck, use this structure. It is a scaffold to be removed once the habit forms, not a formula to leave visible.
[Source] found [finding]. This is [strength or limitation] because [reason]. [Second source] however [contrast]. This suggests [your interpretation].
Four sentences. One paragraph. The final one is yours, and it is the only sentence in the paragraph that could not be written by someone who had read the abstracts.
Where to be critical, and where not
| Section | Expected level |
|---|---|
| Introduction | Light. Establish the problem, do not argue it yet. |
| Literature review | Heavy. This is where evaluation belongs. |
| Methodology | Heavy, directed at your own choices and their alternatives. |
| Results | None. Report findings without interpreting them. |
| Discussion | Heaviest, including of your own work. |
| Conclusion | Light. State what was established and its limits. |
Being critical in the results chapter is a real error, not extra credit. It leaves the discussion with nothing to do and blurs the line between what you found and what you think it means.
Being critical about your own work
Students often treat the limitations section as a confession. Markers read it as evidence of judgement. A limitation you identify yourself demonstrates that you understand your method. One the marker finds first suggests you did not.
Defensive
Although the sample was small, the findings still provide useful insight.
Genuinely critical
The sample of twelve was drawn from a single university, so the findings describe a particular institutional context rather than undergraduate experience generally. They are best read as a basis for wider testing rather than as evidence of a general pattern.
The second version does not weaken the work. It shows you know exactly what your evidence can and cannot support, which is the whole skill being assessed.
A test for your own draft
Take any paragraph and delete every sentence that reports what a source said. If nothing remains, the paragraph is descriptive. If two or three sentences of your own reasoning survive, it is analytical.
Run that on three paragraphs and you will know precisely what your feedback meant, which is more than the feedback itself usually manages.
Last updated: 30 July 2026 · Written by the AssignWise editorial team ·Spotted something wrong?
Frequently asked questions
Judging the quality and limits of evidence rather than reporting what it says. It means asking how a finding was produced, how far it can be generalised, whether it agrees with other work, and what follows if it is right.
It almost always means you described sources rather than evaluating them. The fix is to add a sentence after each piece of evidence explaining its strength, its limits, or how it relates to other findings.
No. Critical evaluation is judgement, not criticism. Concluding that a study is well designed and its findings robust is just as critical as identifying a flaw, provided you say why.
Ask five practical questions: who conducted it and with what funding, what sample and method, when, in what context, and does it agree with other work. You can answer all five from the abstract and method section without expertise in the field.
More than most students allocate. A rough guide for a well-marked paragraph is a quarter claim and evidence, and three quarters analysis. If the ratio is reversed, the essay will read as descriptive.
Want a second pair of eyes on it?
Send us your document for a fixed, no-obligation quote. We reply within one working hour.
Get Your Free Quote