Common grammatical errors in academic writing
Eight mistakes that account for most of the corrections we make, with examples and a method for finding them in your own draft.
After enough documents, the same errors keep appearing. Not obscure ones: eight patterns account for most of what gets corrected in student work, and almost all of them cluster in long sentences. That is the single most useful thing to know, because shortening sentences fixes more grammar than studying grammar does.
1. The comma splice
Two complete sentences joined by a comma. The most common error in student writing, by a distance.
Wrong
The sample size was small, the results should be treated with caution.
Right
The sample size was small; the results should be treated with caution.
The sample size was small, so the results should be treated with caution.
The sample size was small. The results should be treated with caution.
Three valid fixes. Use a semicolon when the halves are closely linked, a conjunction when you want to name the relationship, and a full stop when they are separate thoughts. Ourcolons and semicolons guide covers the first option properly.
2. Subject-verb disagreement after a long subject
Agreement is easy over three words and difficult over fifteen. By the time you reach the verb, the wrong noun is nearest.
Wrong
The range of responses collected from participants across all three sites were broadly consistent.
Right
The range of responses collected from participants across all three sites was broadly consistent.
The subject is range, not sites. Strip the sentence to its bones and the error is obvious: the range was consistent.
The data problem
Data is plural in most British academic writing: the data were analysed. Some disciplines now accept the singular. Pick whichever your department uses and hold it for the whole document, because inconsistency is more noticeable than either choice.
3. Tense drift
The usual convention in UK academic writing:
| Context | Tense | Example |
|---|---|---|
| What you did | Past | Participants completed the questionnaire. |
| What earlier studies found | Past | Ahmed (2024) reported a similar effect. |
| Established knowledge | Present | Cortisol rises under acute stress. |
| What your data shows | Present | Table 3 shows a decline across all groups. |
| Your document itself | Present | This chapter examines three variables. |
Wrong
Participants complete the survey in April, and results showed a significant effect which suggests the intervention works.
Right
Participants completed the survey in April, and the results showed a significant effect, which suggests that the intervention works.
4. Dangling modifiers
An opening phrase must describe whatever the sentence names next. When it does not, the result is accidentally absurd.
Wrong
Having analysed the transcripts, three themes emerged.
Right
Having analysed the transcripts, I identified three themes.
Three themes emerged from the transcript analysis.
The first version says the themes did the analysing.
5. Apostrophes
Three rules, and they cover almost everything.
- Possession: the participant's response (one), the participants' responses (several)
- Contraction: it's means it is, and contractions are usually avoided in formal academic writing anyway
- Never for a plural: the 1990s, three PhDs, several NHS trusts
Its versus it's is the one that survives into final drafts most often. If you cannot substitute it is, there is no apostrophe.
6. Which and that
Use that when the clause is essential to the meaning, and which with a comma when it adds detail you could remove.
Different meanings
The studies that used random sampling reported higher effect sizes. (only some studies)
The studies, which used random sampling, reported higher effect sizes. (all of them did)
7. Vague pronouns
This at the start of a sentence, with no noun after it, is the commonest source of genuine ambiguity in student writing. The reader has to guess which of the preceding ideas you mean.
Unclear
Response rates fell and the funding was reduced. This affected the results.
Clear
Response rates fell and the funding was reduced. This reduction in sample size affected the results.
Make it a habit: never let this stand alone. Always follow it with a noun.
8. Sentences that outgrew themselves
Not strictly a grammatical error, but the cause of most of the ones above. Somewhere past about thirty words, agreement slips, modifiers detach and commas start doing work they cannot do.
Before, 44 words
While the findings of this study, which was conducted across three separate sites over a period of six months, appear to support the original hypothesis, it should be noted that the sample was relatively small and therefore the results was not generalisable.
After, two sentences
The findings support the original hypothesis. However, the sample was small and collected across only three sites, so the results are not generalisable.
Forty-four words to twenty-six, and the agreement error disappeared with the length.
How to find these in your own draft
- Have it read aloud to you. Word and most operating systems will do this. Your ear catches disagreement and run-ons your eye skims.
- Search for every comma. Ask whether a full stop would work. If yes, you have found a splice.
- Search for "This " at the start of sentences. Add a noun after each one.
- Find your longest sentences. Anything over thirty words gets read separately and usually split.
- Read the reference list backwards. Out of order you see the formatting rather than the content.
Twenty minutes of that catches most of what a marker would flag, and every pass makes the next draft cleaner.
Last updated: 30 July 2026 · Written by the AssignWise editorial team ·Spotted something wrong?
Frequently asked questions
Comma splices, subject-verb disagreement after long clauses, inconsistent tense between methods and results, dangling modifiers, and apostrophe misuse in plurals. Almost all of them cluster in long sentences, which is why shortening sentences fixes more errors than studying grammar rules.
Spellcheck compares words against a dictionary, so any real word in the wrong place passes: "form" for "from", "trail" for "trial", "casual" for "causal". Grammar checkers add pattern matching but regularly flag correct academic constructions as errors and miss genuine ones.
Not inherently, and this is one of the most persistent myths. Many disciplines now accept first person for methodological choices, particularly in qualitative research and reflective work. Sciences often still prefer the passive. Follow your department guidance rather than a general rule.
Past for what you did and what previous studies found, present for established facts and for what your own data shows, and present for describing the structure of your own document. The common error is drifting between them inside a single paragraph.
Read it aloud, or have your computer read it to you. Errors that are invisible on screen are obvious in speech, because your ear catches disagreement and run-on sentences that your eye skims past.
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