Free tool · private
Academic writing checker
Checks the mechanical things a marker notices immediately: register, sentence length, passive constructions, contractions and spelling that does not match your country. Everything runs in your browser.
Runs entirely in your browser. Nothing is uploaded, stored or sent anywhere.
Statistics
- Words
- 0
- Sentences
- 0
- Avg sentence
- 0 words
- Longest sentence
- 0 words
- Reading time
- 1 min
- Reading grade
- —
Suggestions
Paste text on the left and suggestions appear here as you type.
Short answer
What makes writing sound academic?
Four things, in order of impact: precision (specific claims over vague ones), attribution (every borrowed idea sourced), measured confidence (hedge where evidence is genuinely uncertain, state plainly where it is not), and register (full forms rather than contractions, specific nouns rather than “things”). Formality is the least important of the four, and the one students over-correct hardest — long words do not substitute for a clear argument.
Step by step
How to check academic writing before you submit
Paste the draft
The whole analysis runs in your browser in JavaScript. Nothing is sent to a server, stored or logged — which matters for unpublished work, and is why this was built without an API call.
Set your spelling locale
UK, US and Australian conventions differ on -ise/-ize, -our/-or and -re/-er. Markers notice inconsistency far more than they notice which variety you chose, so pick one and let the checker enforce it.
Work through the flags in order
Overlong sentences, passive constructions, informal register, contractions and hedging are flagged with the text that triggered them, so you can see the sentence rather than a count.
Judge each flag in context rather than clearing them all
Passive voice is standard and often preferred in a methods section — 'samples were collected' rightly keeps the focus on the procedure. It weakens argument, where naming the actor is stronger. The tool flags; you decide.
Re-read for meaning yourself
No checker can tell you whether your argument answers the question or whether your evidence supports your claim. Mechanics are the floor, not the mark.
Worked example
What the checker actually finds
This paragraph was written to fail, and analysed by the tool above when the page was built. Every flag below names the words that triggered it, because “passive voice detected” with no excerpt is not feedback you can act on.
The paragraph
The data was analyzed by the research team and it was found that there's basically a lot of variation in the results, which might perhaps suggest that the behavior of the participants was somewhat affected by the experimental conditions that were established at the beginning of the study before the second phase of data collection had been completed by the team. These findings are quite interesting and could possibly be really important for future work.
- Words
- 75
- Sentences
- 2
- Longest sentence
- 61 words
- Issues flagged
- 9
| Severity | What it found | In your text | How to fix it |
|---|---|---|---|
| High | Sentence runs to 61 words. Markers lose the argument past roughly 30. | The data was analyzed by the research team and it was found that there's basically a lot of… | Split at the first conjunction, or move the qualifying clause into its own sentence. |
| Medium | "a lot of" appears 1× — informal for academic register. | a lot of | Consider: many / substantial |
| Medium | "basically" appears 1× — informal for academic register. | basically | Consider: (delete) |
| Medium | "really" appears 1× — informal for academic register. | really | Consider: (delete) |
| Medium | "analyzed" is US spelling in a UK document. | analyzed | Use "analyse" spelling throughout. |
| Medium | "behavior" is US spelling in a UK document. | behavior | Use "behaviour" spelling throughout. |
| Medium | 1 contraction found. Most academic styles expect full forms. | there's | Write 'do not' rather than 'don't'. Possessives such as 'the study's findings' are fine. |
| Low | Passive construction. Acceptable in methods sections, weak in argument. | was analyzed | Name the actor: 'The study found' rather than 'It was found'. |
| Low | Heavy hedging (5 hedge words). Over-hedging reads as lack of confidence. | might, perhaps, possibly, somewhat, quite | Keep hedges where the evidence is genuinely uncertain; state findings plainly elsewhere. |
Frequently asked questions
Is this sentence grammatically correct?
Paste it above and the checker will flag the mechanical problems it can detect: length, passive construction, informal register, contractions and locale-inconsistent spelling. It cannot judge meaning — for that, a marker or a tutor has to read it.
Is my text uploaded anywhere?
No. The entire analysis runs in your browser in JavaScript. Nothing is sent to a server, stored, or logged. That matters for unpublished work, which is the main reason we built it this way rather than calling an API.
Should I avoid passive voice completely?
No. Passive voice is standard and often preferred in methods sections — 'samples were collected' keeps the focus on the procedure. It weakens argumentative writing, where naming the actor is stronger: 'Smith argues' rather than 'it is argued'.
How long should an academic sentence be?
Aim for an average of 15–25 words. Anything past about 40 words is where markers start losing the thread. Vary the length deliberately — a short sentence after a long one lands the point.
Can I use contractions in academic writing?
Generally no. Most academic style guides expect full forms — 'do not' rather than 'don't'. Possessive apostrophes such as 'the study's findings' are always fine, and reflective writing is sometimes more relaxed.