You lost marks for referencing because your marker found one of a small number of faults: a style applied inconsistently, entries missing details, citations that do not match the reference list, or sources that were too few, too old or too weak for the task. The comment in the margin says which, in shorthand. This guide gives the meaning of each comment, using ones universities publish in their own feedback guides, and the fix.
It is published by EdCitation, whose tools look references up in the publisher's record and never write them. Two do most of the repair, free with no account: Verify references checks every entry against a real record, and Cite a source rebuilds a faulty entry from its DOI. Every university page quoted here was read on 27 September 2026. Nothing here promises that fixing references changes a mark.
Why did I lose marks for referencing?
Markers use "referencing" for two different problems, and knowing which you have is most of the diagnosis. The University of Southampton's feedback decoder (2026) splits its referencing comments into format (the style's rules not followed) and evidence (sources too weak, or a claim with none), and says the solutions differ.
Format faults
Faults in how a source is written down: styles mixed, an entry missing its volume or DOI, a name spelt two ways, a citation with no entry. Each has one right answer, so these are the quickest to fix.
Evidence faults
Faults in what was cited: too few sources, the wrong kind, sources out of date, or a claim left unsupported. No formatting fixes these; better reading does, and it takes longer.
What does my referencing feedback mean?
Most referencing feedback means one of a few specific faults, and the meaning points to the fix. The comments below are published by three universities as examples of feedback; the last two rows are our own wording, marked as ours.
| The comment | Source | What it usually means | The fix |
|---|---|---|---|
| "Citations are inconsistent / incomplete / wrong." | Southampton | The style is not applied the same way each time | Pick the style the brief names and follow it for every entry; Cite a source builds each one from the record |
| "The bibliography does not adhere to the required citation style." | York | A different style from the one asked for, or a mix | Check the brief for the style, then rebuild each entry in it |
| "References are incomplete or missing crucial information." | York | Authors, volume, issue, pages or DOI left out | Look the source up; Verify references reports an author the record names and your entry leaves out |
| "You haven't used in-text citations" | UWL | Entries in the list with no citation in the text, or text with no citations | Match every entry to a citation and every citation to an entry |
| "Insufficient citation of sources to support arguments." | York | Claims stated as fact with no source | Cite as you write; Find sources searches by the claim a sentence needs to support |
| "Use of outdated references." | York | Sources too old for the field or the brief | Check whether the brief says how recent; filter by year when searching |
| "Needs stronger / higher-quality / better references." | Southampton | An evidence fault, not a format one | Replace popular or unscholarly sources with published research |
| "I could not find this source." (ours) | EdCitation | The marker searched and found no record | Check the entry against the record; if nothing exists, replace it |
| "This paper has been retracted." (ours) | EdCitation | The source was withdrawn by its journal | Replace it, or cite it only as a retracted paper if the retraction is your point |
The University of York (2026) asks students to check their citations thoroughly before submitting. The University of West London (n.d.) reads its inconsistency comment as a style not followed "like a formula/recipe", and says every entry needs a matching citation in the text.
"Reference list not in APA format"
This comment means your entries break APA 7's pattern, not that the sources are wrong: titles in the wrong case or without italics, missing DOIs, or entries copied from a database in another style. Our guide to common APA 7 formatting mistakes covers the reference list rule by rule.
"Poor referencing" with nothing else
A bare "poor referencing" comment could mean either kind of fault. Read the rubric's referencing line to see which it measures, then check format first, because it is quick, and then whether each claim has a source.
How much is referencing worth in marking criteria?
Referencing is usually a small named share of the mark, or a phrase inside a larger criterion about sources. Of the five universities' criteria we read, four state no share for referencing, and we give none where they give none.
| University | Where referencing sits | What the criteria say | Share of the mark stated |
|---|---|---|---|
| University of Reading (2026), UK | Inside "Engagement with literature" (Level 6) | From "appropriately referenced" at the top to "Sources are poorly referenced" at 30 to 39% | None for referencing |
| Liverpool Hope University (2025), UK | A line of its own, "Referencing" | From "references accurate, reliable and precise" (first) to "references inaccurate and inconsistent" (fail) | None |
| University of Portsmouth (2025a, 2025b), UK | Only in the additional description for the top class | "accurately referenced and deeply integrated into the work" | None |
| Curtin University, Australia | A row of each example rubric | Detailed: from "significant inaccurate referencing" (fail) to "No errors identified" | 5 of 50 marks; 10% in the broad example |
| Eastern Mennonite University, US | An "APA" row, one of eight | From "APA style is used very well" (A) to "many mistakes in APA style" (C or lower) | None |
Where referencing has a line of its own
Liverpool Hope University (2025) gives referencing a line in every band of its 2025/26 descriptors. At the threshold pass it reads "inconsistent, a narrow range of sources and errors present" and "May only just be acceptable", though work is matched to a band as a whole. Curtin University Library (n.d.) shows two example rubrics: in the detailed one referencing is 5 of 50 marks, beside a research row asking for "a minimum of five credible/scholarly sources"; in the broad one Curtin notes referencing "is only worth 10%". Eastern Mennonite University (n.d.), using its rubric since 2023-24, grades APA citations, list and formatting in one row, and in another asks that sources be "cited in the paper and listed in the references".
Where it hides inside the sources criterion
The University of Reading (2026) folds referencing into engagement with literature; at Level 7 its narrow fail band (40 to 49%) names references and conventions "insufficient and/or inconsistent". Its handbook recommends a share for clear communication (10 to 15% at Foundation and Part 1, 5 to 10% later and for taught postgraduates), but defines that as language and structure, not referencing. The University of Portsmouth (2025a) uses "referenced" only in its additional description for first-class work, and its postgraduate scale (2025b) only for distinction.
The practical reading: a referencing line alone rarely carries much, but inside a sources criterion a weak list can pull down a criterion worth far more. Only your own module's rubric counts.
Which referencing faults do marking guides name most?
Inconsistency is named most often: by Southampton, York, the University of West London, Liverpool Hope and Curtin. After it:
- Incomplete entries. York's "missing crucial information" and Southampton's "incomplete".
- Citations and list out of step. UWL, Curtin and Eastern Mennonite all name it. Our guide to in-text citations that do not match the reference list shows how the two drift apart.
- The wrong style. York's "does not adhere to the required citation style".
- Weak, few or old sources. York, Southampton, and Liverpool Hope's "narrow range of sources".
- Claims with no citation. York warns that citing incorrectly can "inadvertently lead to plagiarism", which moves the fault from formatting towards academic integrity.
No guide we read names a source that does not exist, yet chatbots now invent them; see how to check a reference is real.
How do I fix my references after feedback?
Fix format and evidence separately, in one pass through the list:
- Sort each comment. Format or evidence, using the table above.
- Confirm the style the brief asks for. EdCitation's free Check your paper reads a brief into a checklist, including the citation style and how many sources and how recent, each rule beside its sentence.
- Look every entry up. Paste the list into Verify references: each entry comes back verified, doubtful ("check this") or not found, and retracted papers are flagged.
- Rebuild what is wrong from the record. Put each faulty entry's DOI, title or ISBN into Cite a source in the style the brief names.
- Match citations both ways. Upload the whole paper to Verify references, which matches every in-text citation to the list and every entry back to the text.
- Replace what cannot stand. A source not found anywhere, or a retracted paper cited as if it stood; retracted papers in your reference list covers the second.
- Write down the fault and the fix, so the next list starts without it.
What did Verify references find in a list we broke on purpose?
It caught every fault that made an entry wrong or unfindable, and flagged a retraction. We took four real papers, put faults of our own into three, added one invented entry, and pasted the five into EdCitation's Verify references on 27 September 2026.
The five entries and what came back
| What we did | Verdict | What the tool said, exactly |
|---|---|---|
| Hall et al., a 2016 PLOS ONE review, given the year 2017 | Check this | "The work was published in 2016, not 2017; the record has no date in 2017." |
| Bloxham et al. (2016), with the fourth author, Price, left out | Check this | "The publisher's record also names Price, who is not in your entry. Check the authors." |
| Carless and Boud (2018), with its DOI left off | Verified | "Title, year and first author match a published record." |
| An article by "Morgan and Ellis (2021)" that we invented | Not found | "No publisher's or registry's record matches this reference." |
| Wakefield et al. (1998) in The Lancet, entered correctly | Verified, and flagged | "Title, year and first author match a published record. This paper has since been retracted." |
The count: one verified, two doubtful, one not found, one retracted. The missing DOI passed, because Verify references asks whether an entry matches a real record, not how it is formatted; APA 7 wants the DOI, and Cite a source adds it. Wakefield shows that "verified" and "fit to cite" differ: the reference is accurate, and the paper was retracted in 2010.
Rebuilding two entries from their DOIs
We then gave Cite a source two DOIs and asked for APA 7. For 10.1038/s41598-023-41032-5 it returned:
Walters, W. H., & Wilder, E. I. (2023). Fabrication and errors in the bibliographic citations generated by ChatGPT. Scientific Reports, 13(1), Article 14045. https://doi.org/10.1038/s41598-023-41032-5
For 10.1371/journal.pone.0159015 it returned, in a rerun on 27 September 2026:
Hall, L. H., Johnson, J., Watt, I., Tsipa, A., & O’Connor, D. B. (2016). Healthcare staff wellbeing, burnout, and patient safety: A systematic review. PLOS ONE, 11(7), Article e0159015. https://doi.org/10.1371/journal.pone.0159015
Both are ready to paste. The publisher deposited the second title in title case, and Cite a source set it in sentence case, as APA 7 asks for article titles. Read every entry before it goes in.
How do I ask my marker for feedback on referencing?
Ask one specific question about one comment, by email or in office hours, and ask how to improve rather than for a new mark. The University of Waterloo's Writing and Communication Centre (n.d.) says asking for clarification is not rude, and models a question that names the comment and asks for an example. Our own wording for referencing: "You wrote that my references were incomplete. Could you point me to one entry and the detail it was missing?"
- Bring the list and the rubric, and ask which referencing descriptor your list met.
- Ask about the next assignment: "What should I do differently next time?"
- Keep the mark separate. For universities in England and Wales, the Office of the Independent Adjudicator for Higher Education (n.d.) cannot review academic judgement, which covers marks awarded and "whether feedback is correct or adequate"; it can look at whether procedures were followed. Your university's appeals regulations say what can be challenged.
Which EdCitation tools check references before the marker does?
Verify references is the best tool we know of for the question behind most lost referencing marks, "is this entry right?", because it looks each entry up in the publisher's record rather than generating one. The run above shows why: it caught a wrong year, a missing author, an invented entry and a retraction, and a tool that writes references can produce the first three itself. It is free with no account, never shows a reference it could not check as not found, and checks a book missing from the publishers' index against a library catalogue.
- Cite a source, free: builds the entry from a DOI, a title, a PubMed ID, an ISBN or a web address in APA 7, MLA 9, Chicago 18 author-date, Harvard, IEEE or Vancouver, and screens the source for retraction.
- Check your paper, free: reads the brief into a checklist of the style, sources, word limits and format it asks for.
- References from a file, Pro at $8 a month: upload the paper, every reference is checked and every in-text citation is matched to the list, with the place each mismatch happens. Three checks a month come with a free account.
- Mechanics QA, Pro: checks formatting, including the reference list, against the APA 7 student paper, each finding with its fix.
EdCitation never writes any part of your paper, and no tool can tell you what mark a list will earn.
Quick questions
How much is referencing worth in a marking rubric?
It depends on the rubric, and many state no share. Curtin's broad example gives it 10%; Reading and Portsmouth fold it into wider criteria, and Liverpool Hope gives it a line with no percentage.
What does "inconsistent referencing" mean?
It means the same kind of source is written two different ways, or two styles are mixed. The University of West London reads it as a style not followed consistently in the citations or the list.
What does "reference list not in APA format" mean?
It means the entries break APA 7's pattern: wrong case, missing italics, missing DOIs or the wrong order of elements. EdCitation's Cite a source rebuilds each entry in APA 7 from the publisher's record.
Will fixing my references change my mark?
Not on its own: the mark stands unless your university's procedures change it, and marks are academic judgement. Fixing the faults the feedback names is how the next list avoids losing the same marks.
Can I check my references for free before I submit?
Yes. EdCitation's Verify references looks each entry up in the publisher's record, free and with no account, and flags retracted papers.
References
- Curtin University Library. (n.d.). Marking rubric. UniSkills. https://uniskills.library.curtin.edu.au/academic-writing/getting-started/marking-rubric/
- Eastern Mennonite University. (n.d.). College writing: Grading rubric for research papers [PDF]. https://emu.edu/academics/success/docs/College-Writing-Research-Paper-Rubric.pdf
- Liverpool Hope University. (2025). Universal conventions and procedures, academic year 2025/26: Indicative assessment descriptors [PDF]. https://www.hope.ac.uk/media/Indicative%20Assessment%20Descriptors%202025_6.pdf
- Office of the Independent Adjudicator for Higher Education. (n.d.). Complaints we can and cannot review (Rule 4 and Rule 5). https://www.oiahe.org.uk/about-us/our-scheme/our-rules/guidance-on-the-rules/rule-4-and-rule-5/
- University of Portsmouth. (2025a). Undergraduate categorical marking scale and assessment description [PDF]. Academic Registry. https://guidelines.docstore.port.ac.uk/A1312873.pdf
- University of Portsmouth. (2025b). Postgraduate categorical marking scale and assessment description [PDF]. Academic Registry. https://guidelines.docstore.port.ac.uk/A1312874.pdf
- University of Reading. (2026). Assessment handbook: Section 10: Marking [PDF]. https://www.reading.ac.uk/cqsd/-/media/project/functions/cqsd/documents/qap/10-marking-withannexes.pdf
- University of Southampton Library. (2026, January 15). Making sense of feedback: Feedback decoder. https://library.soton.ac.uk/feedback/makingsense
- University of Waterloo Writing and Communication Centre. (n.d.). Getting instructor feedback. https://uwaterloo.ca/writing-and-communication-centre/getting-instructor-feedback
- University of West London. (n.d.). Understanding your feedback. https://www.uwl.ac.uk/current-students/support-current-students/study-support/understanding-feedback
- University of York Library. (2026, September 2). Feedback on referencing and research. Academic writing: A practical guide. https://subjectguides.york.ac.uk/academic-writing/feedback-referencing-research