Do quotes count in the word count?
Yes. At almost every university that publishes a rule, direct quotations count toward your word count, because anything sitting in the main body counts whether you wrote it or quoted it. What does not count, near-universally, is the reference list or bibliography, and appendices. Everything else, footnotes, tables, captions, the abstract, the contents page, even the cover sheet, is genuinely contested: two departments at the same university can and do publish opposite rules. So the real answer is that your assignment brief wins, always, and the table below shows you exactly how far apart real institutions are so you know what to go and check.
- Quotes count, and here is the one-line version
- What each element does, at six real universities
- Two safe rules, and the ones that are a coin toss
- Is there really a 10% allowance?
- What actually happens if you go over
- Going under is not the penalty you think it is
- Getting a number you can trust in Word and Google Docs
- Where to find your actual rule, in under two minutes
Quotes count, and here is the one-line version
James Cook University's library answers it in a sentence: "In most cases, direct quotations do count toward the word count of your assignment." The same page adds that this "applies to both short quotes and longer block quotations", so an indented block quote is not a loophole.
In-text citations go the same way. Newman University is blunt about it: "quotations are included in your word count, as are in-text citations. The only part of your references that don't count towards your word count is the reference list at the end of your assignment." Every (Rahman, 2024) in your body is words on the meter.
It reaches further than most people expect. The University of Wolverhampton warns that "direct quotations as well as words contained within tables, charts or diagrams would count towards your word limit if they appear as part of the main body of your work". If it is in the body, assume it counts.
One verified exception exists, and it is not a university. National History Day, the US school competition, excludes quotes and their source credits from the 1,200 student-composed words in a website entry. No university policy we could find excludes quotations.
What each element does, at six real universities
Below are ten questions put to six institutions that publish an itemised rule. Every column links to the policy it came from, so you can check any cell yourself. Read across a row and you will see why nobody can hand you a universal answer.
| Element | SOAS | UWE Bristol | Oxford English | Arden | ANU Asia & Pacific | ANU Law |
|---|---|---|---|---|---|---|
| Direct quotations | Counts | Counts | Counts | Counts | Counts | Counts |
| In-text citations | Counts, any style | Counts, bracketed or not | Counts | Counts | Counts | Counts |
| Footnotes and endnotes | Counts | Excluded, if they hold references only | Counts | Excluded | Counts if substantive, not if bibliographic | Excluded |
| Reference list or bibliography | Excluded | Excluded | Excluded | Excluded | Excluded | Excluded |
| Appendices | Excluded, not marked | Excluded | Excluded | Excluded, not marked | Excluded | Counts, unless they only reproduce primary material |
| Tables, figures, captions | Figure and table legends count | Counts, tables, graphs, figures, diagrams | Counts, table contents and figure captions | Excluded, tables, figures, diagrams | Counts, tables and text boxes | Not stated |
| Headings and subheadings | Counts | Counts | Counts | Not stated | Counts | Counts |
| Abstract | Counts | Excluded | Counts, and you are told not to have one | Not stated | Not stated | Not stated |
| Title page, contents page | Contents page excluded | Assessment title excluded | Essay title excluded, contents page counts | Not stated | Not stated | Not stated |
| Cover sheet and declaration | Declared; a wrong one is misconduct | Declared on the title page or cover sheet | Given fields free, other text counts | Declared on the front cover sheet | Cover sheet excluded | Count recorded at the end, or per page |
Look at the last two columns: two schools of the same university. The ANU College of Asia and the Pacific counts "footnotes with substantive text" and leaves out "bibliographic content in footnotes". The ANU Law School says flatly that "The bibliography and footnotes will not be counted", then counts your appendices instead unless they "merely reproduce primary materials for the aid of the examiners". Two opposite rules, both current.
In the United States the question often changes shape, because length is more often set in pages by the instructor than in words by an institution. Florida International University's law library sets a memo at "Five to ten pages, Times New Roman 12-point, double-spaced, one-inch margins", with a research section "not counted toward the page limit". A community college guide teaches the conversion: "In MLA formatting, 250 words = about 1 page." Where a US course does set words it looks like everywhere else: Cleveland Community College excludes "The works cited page, the title page and any other identification". Your syllabus is the document that matters.
Two safe rules, and the ones that are a coin toss
Two things are safe. Your reference list does not count, at every university in the table and every library FAQ we read. Appendices do not count either, with the single exception above. If you are deciding what belongs in one, start with the anatomy of a research paper.
The trap is that your citation style says the opposite. APA's own guidance is to "count all words in text, references, table bodies, table and figure notes, and appendices", and the manual, quoted by Royal Roads University, tells authors to "count every word from beginning to end". MLA and Chicago say nothing about assignment length at all. So if someone tells you APA excludes the reference list, they have it backwards. James Cook gives the consequence: "APA recommends to count everything, even headers and footers. However, the final decision is made by your college or lecturer."
Footnotes are the coin toss. SOAS counts them. So do the Oxford Faculty of English, Edinburgh's School of Social and Political Science, Auckland Law School and Trinity College Dublin's School of English: "The word count includes footnotes but not the bibliography." UWE Bristol excludes them, but only "Provided footnotes and endnotes only include references". Arden excludes them outright. And Leeds Conservatoire lists, among the things your count includes, "Other footnotes", which is exactly as ambiguous as it looks.
Two of the strictest exclusions anywhere are thesis rules, not coursework rules, so do not carry them into an essay. Birmingham's thesis count excludes "Tables, diagrams (including associated legends), appendices, references, footnotes and endnotes". Lancaster's postgraduate research regulations exclude "citations" from the thesis word length altogether.
The abstract splits too. SOAS counts it, UWE excludes it, and Oxford's English Faculty counts it and warns you: "You are advised not to include a dedication, table of contents or an abstract, and warned that these will be counted within the word limit." If you are writing one, how to write an abstract covers the shape. And plenty of institutions publish no element rule at all, which is the most honest answer any of them give.
| Institution | What it says instead of a rule |
|---|---|
| UCL | "The word count must specify whether footnotes, bibliographies, appendices, tables, figures etc. are to be included in the word count." The obligation sits on the assessment, not the university. |
| Durham | "There is no University-wide policy on the penalties for submitting over-length assessed work at undergraduate or postgraduate level." |
| Northumbria | Publishes a menu rather than a rule, listing the abstract, citations, quotations, the title and contents page, tables, footnotes, the reference list, appendices and a glossary as things a programme "should identify". |
| Western Sydney | "Referencing styles including APA do not give guidelines on what is included in a word count. This is something that is set by those who have created the assessment task." |
Is there really a 10% allowance?
No. It is one policy repeated until it sounds like a law. SOAS, Northumbria, Arden, Leeds Conservatoire and Trinity's School of English do all allow up to 10% over, so inside those five it is true, and that is where the folklore comes from.
Everywhere else the number is different, or absent. Auckland Law School allows 5%, so that students "are not penalised for accidentally exceeding the limit by a few words". ANU's College of Asia and the Pacific starts penalties above 110% of the limit. Oxford's English Faculty allows nothing: "Word limits are applied strictly, and there is no acceptable 'buffer' before a penalty is applied for over-length work." Edinburgh deducts a flat five marks "regardless of how many words over the limit the work is (whether it is by 1 word or by 500!)". And UWE has warned that a tolerance must be granted rather than assumed: "if your module guide does not state that 10% tolerance will be given then it does not exist for that module."
So across six real policies the allowance runs from nothing to a tenth of the limit. Assuming 10% because a friend on a different course said so is how people lose marks they never needed to lose.
What actually happens if you go over
Four mechanisms are in use, and the first is the one nobody pictures. The marker simply stops reading. Northumbria says the marker "will stop reading" past 10% over, and does the arithmetic: "for a 3000 word essay, the marker will read only the first 3300 words". UWE applies the same idea with no threshold, and past the limit "no marks will be awarded and no feedback provided". Alliance Manchester Business School calls it a "guillotine" system. If your conclusion is past the line, it does not get read.
Banded deductions come next. SOAS takes 10 percentage points for 10 to 20% over, 15 for 20 to 30%, and zero above 30%. Leeds Conservatoire shows its working: a 2,000 word journal handed in at 2,440 words is 22% over, so "your mark will be reduced to 0". Oxford starts at minus one mark and stops at minus ten.
Flat deductions with a floor are third. Arden takes 10 percentage points past 10% over and will not push you below the pass mark, and UCL Philosophy does the same. Auckland Law School takes five marks per 10% over, but promises "no student will fail an assessment because of a penalty for exceeding the word limit".
The fourth is not a length penalty. Getting your declared count deliberately wrong is an integrity matter. SOAS treats "significant inaccuracies in declared word counts" as academic misconduct. UWE calls a knowingly wrong count "an assessment offence". Auckland Law School names manipulating the count "by deceptive means" as a breach of academic integrity, says Turnitin detects it, and gives "invisible hyphens" as its example. Declare the real number.
Going under is not the penalty you think it is
Most students assume a short essay is punished like a long one. Almost every policy we checked says the opposite. SOAS: "There is consequently no specific penalty for work that is under-length." Auckland Law School: "There is no penalty applied for an assessment falling under the word limit." Edinburgh, York St John, Arden and UCL's Institute of Education all say the same.
The catch is that nothing rescues a thin essay either. Northumbria puts it best, and two other universities use the same word: "In not making use of the full word count, students may have self-penalised their work." A short essay is not penalised. It simply cannot earn the marks, because there is not enough of it to meet the criteria.
Two caveats. UCL's Academic Manual requires modules to publish "any penalties that will apply for over- or under-writing", so a module may have one, and a few do: Coventry's CU Scarborough takes 10% of your mark for work "10 % under the minimum limit specified". Leeds Conservatoire penalises timed work in both directions, so a presentation that runs short is a different question.
Getting a number you can trust in Word and Google Docs
In Microsoft Word, the number in the status bar is not necessarily your number. Click it, or go to Review and then Word Count, and look at the checkbox labelled "Include textboxes, footnotes and endnotes". That one checkbox decides whether your footnotes are on the meter, and sources disagree about how it is set by default: Sussex IT Services tells students to tick it, while Liverpool describes Word as including them already. So do not assume. Open the dialog and read it, every time, which is what Auckland Law School tells its own students to do.
Two details catch people out. The checkbox only affects real footnotes, the ones inserted from the References tab; text that merely sits at the bottom of a page is ordinary body text and always counts. And Word counts more than you would guess, with Sussex listing an automatic table of contents, a generated bibliography, and captions including their labels.
To count only part of a document, select it. Microsoft's guidance is that the status bar shows the count for your selection alongside the whole document. That is how you get an honest number when your brief excludes the reference list: select from your first heading to the end of your conclusion.
Google Docs is simpler and has one hard limit, stated on Google's own help page: "Unless you select a section of the document, word count applies to everything except headers, footers, and footnotes." There is no setting to change it. So if your department counts footnotes, and SOAS, Oxford English, Edinburgh, Auckland Law and Trinity all do, Google Docs is under-reporting your essay. It has no bibliography object either, so a reference list you typed at the bottom is just more body text.
Your word processor and your marker may be counting different documents. Turnitin does not count footnotes, endnotes or text boxes. Word does. The University of Reading measured it: a file that read 2,097 words in Word arrived in Turnitin as 1,983, a gap of 114 words, all of it footnotes. Submit the same content as a PDF and the number moves the other way, with 2,088 words in Word arriving as 2,142. Liverpool, documenting the same mismatch across four platforms, warns that it can cause "unnecessary or unfair penalties".
So take your number from the original document, not from the submission portal. Reading's own advice is the rule to keep: "Word counts should always be taken from the original document and relate to the criteria set in the assessment brief." Then declare that number wherever your brief asks for it.
Where to find your actual rule, in under two minutes
Four places, in order of authority. The assignment brief first, because every policy quoted here defers to it. Then the module handbook, or the module's own page on your VLE, which is where UWE, Northumbria and Edinburgh all say the decision gets recorded. Then your school or university assessment policy, which is what the second table above is drawn from. Then, if it is written down nowhere, ask whoever is marking it, in writing, so you have the answer.
That order matters because the answer changes at every level. A university can leave it open, a school can set it, and a module leader can override both, which is why UCL tells students that "It is your responsibility to check with each of your teaching departments". It is a two-minute check, so do it before you write, not at midnight with 2,400 words on screen and no idea which of them count. If tonight is already that night, how to write an essay fast and how to write a research paper fast are the plans for the hours you have left.