Common Grammar Mistakes in Academic Writing

A handful of errors account for most of the grammar marks lost in student work, and each takes minutes to learn.

Grammar mistakes rarely change a grade on their own, but they accumulate. A paper carrying the same error twenty times reads as careless, and carelessness colours how a marker reads everything else. The encouraging part is that most students make only a few distinct mistakes repeatedly – learn to catch your own three or four and the problem largely disappears.

Comma splices

A comma splice joins two complete sentences with a comma: “The results were inconclusive, further testing is required.” Both halves could stand alone, so a comma is too weak to hold them. There are four fixes, and the one you choose changes the emphasis.

• Full stop: “The results were inconclusive. Further testing is required.”
• Semicolon: “The results were inconclusive; further testing is required.”
• Conjunction: “The results were inconclusive, so further testing is required.”
• Subordination: “Because the results were inconclusive, further testing is required.”

The test is simple: if the words on each side of the comma form a complete sentence, you need something stronger than a comma.

Dangling modifiers

A modifier dangles when the thing it describes is missing or in the wrong place. “Having analysed the data, the conclusion was clear” implies the conclusion did the analysing. The fix is to name the actor immediately after the comma: “Having analysed the data, we found the conclusion clear.” These are common in academic writing precisely because passive constructions hide the actor, and they produce sentences that are unintentionally comic.

Its and it’s

“It’s” is always a contraction of “it is” or “it has”. “Its” is possessive. Because contractions are discouraged in formal writing, “it’s” should rarely appear in an academic paper at all – which makes the rule easy: if you have written “it’s”, either expand it to “it is” or you meant “its”. The same logic sorts “who’s” from “whose”.

Subject-verb agreement across long subjects

Agreement errors creep in when distance separates the subject from its verb. “The range of responses collected from participants across all three sites were inconsistent” – the subject is “range”, singular, so it should be “was”. Find the actual subject by mentally deleting the intervening phrases. Watch also for “data”, which is plural in formal academic usage: “the data suggest”, not “the data suggests”, though some disciplines now accept the singular.

That and which

“That” introduces information essential to the meaning; “which” introduces additional information, preceded by a comma. “The studies that used control groups showed no effect” narrows to a specific subset. “The studies, which used control groups, showed no effect” says all of them did. The distinction genuinely changes the claim, which is why markers in research-heavy fields pay attention to it.

Apostrophes in plurals

Apostrophes signal possession or contraction, never a plural. “The 1990s”, “several PhDs”, “the participants’ responses” – the last of these is possessive and plural, so the apostrophe follows the s. A quick check: if you are simply describing more than one of something, no apostrophe belongs anywhere near it.

Tense consistency

Academic conventions on tense are more specific than most students realise. Established knowledge takes the present tense (“gravity acts on…”), what a particular study did takes the past (“Chen surveyed 400 students”), and what a paper argues takes the present (“Chen argues that…”). Methods sections are written in the past, and drifting between tenses within a paragraph is one of the most frequent comments markers leave.

How to catch your own errors

Spellcheckers miss most of these, because the words are all correctly spelled. Three habits work better. Read the draft aloud, which catches run-on sentences and awkward agreement almost automatically. Read it backwards a sentence at a time, which stops your eye from smoothing over what you meant to write. And keep a personal list of the two or three mistakes you actually make, then search for them specifically before you submit – targeted checking beats general proofreading every time.

Misplaced “only” and other stray words

Words like “only”, “just”, “even” and “almost” modify whatever immediately follows them, and moving one changes the claim. “The trial only tested two doses” says the trial did nothing but test; “the trial tested only two doses” says how many. In research writing this matters, because the sentence you wrote is the claim you have made regardless of what you intended. When you use one of these words, check that it sits directly before the thing it limits.

Punctuation with citations and quotations

Style guides differ, so check yours, but two conventions are near-universal in author-date systems. The citation belongs inside the sentence’s closing punctuation – “attendance fell sharply (Okafor, 2021).” – not after the full stop. And a quotation integrated into your own sentence takes no capital letter unless the original had one and the grammar requires it. Getting these consistently right is a small thing that markers notice, because inconsistency signals a draft that was never read carefully.

Parallel structure in lists and comparisons

Items in a series should share grammatical form. “The study aimed to measure attendance, improving retention, and how satisfaction changed” mixes an infinitive, a gerund and a clause. Making them parallel — “to measure attendance, improve retention and assess satisfaction” — is both clearer and easier to read.

The same applies to comparisons: “unlike Smith’s findings, this study found” compares findings to a study. It should compare like with like — “unlike Smith’s findings, those reported here suggest”.

Pronoun reference that actually refers

Ambiguous pronouns are common in dense academic prose. “The committee informed the department that it had exceeded its budget” — which one? Rewriting to name the subject is almost always better than adding qualifiers.

Watch particularly for “this” opening a sentence with nothing specific attached. “This suggests that…” is fine only when the referent is unmistakable; usually “this pattern” or “this discrepancy” is needed.

Colons, semicolons and dashes

Each has a defined job, and misuse is visible to markers. A colon introduces what follows — a list, an explanation, an example — and needs a complete sentence before it. A semicolon joins two closely related complete sentences, or separates list items that already contain commas. A dash marks an interruption or an emphatic aside, and should be used sparingly in formal writing.

• Colon: the results were unambiguous: attendance predicted attainment.
• Semicolon: the sample was small; the effect was nevertheless consistent.
• Dash: the finding — surprising given the literature — held across all three cohorts.

Hyphens in compound modifiers

Compound adjectives before a noun take a hyphen: “a well-designed study”, “high-quality evidence”, “second-year students”. After the noun they usually do not: “the study was well designed”. Adverbs ending in -ly are never hyphenated: “a highly cited paper”, not “a highly-cited paper”.

Build a personal error list

General proofreading is inefficient. Most writers make three or four distinct errors repeatedly, and targeted checking catches far more than a slow read-through. Keep a list of the mistakes your markers have actually flagged, and search for them specifically before submitting — the find function will locate every instance of “it’s” or “due to the fact that” in seconds.