A textbook quotes a famous study, the study makes your point, and it goes into your reference list as if you had read it. You can use an idea met second-hand, but only by saying so. That is a secondary citation, and each style has its own form: "as cited in" in APA, "qtd. in" in MLA, "cited in" or "quoted in" in Harvard and Chicago, and in IEEE, according to the university guides, no form at all.
This guide, published by EdCitation, sets out each style's rule from its publisher or a named university library guide, read in September 2026, and shows why the original is worth the trouble. When your list is final, EdCitation's free Verify references checks every entry against the publisher's record, and Cite a source builds the original's entry from its DOI once you have found and read it. EdCitation never writes any part of your paper.
What is a secondary citation?
Citing a source within a source is a secondary citation: it credits an idea or a quotation to its original author while naming the work in which you actually found it. The American Psychological Association (2022) calls the original the primary source and the work that reports it the secondary source. The rule in every style below is the same at heart: your reference list describes what you read.
The University of St Andrews (2026) puts it in note-taking terms: get the cited source and read it in context where you can, say "as cited in" where you cannot, and keep the two apart in your notes, which is where most accidental cases begin.
Why do styles discourage citing a source you have not read?
Because a claim passed from paper to paper changes on the way, and you inherit every change unseen. Penders (2018) makes reading what you cite the second rule of responsible referencing: only reading shows whether a paper supports your claim.
What happens along a chain
Rekdal (2014) gives the case every referencing tutor should know. A 1995 paper by Larsson, itself about false data spread by poor citation, said the myth of spinach as a rich source of iron dated from the 1930s and came from a misplaced decimal point, citing a 1981 article by Hamblin. When Rekdal read Hamblin, the story was different: Hamblin put the error in the 1890s and its discovery in the 1930s, and gave no source for the decimal point at all (Hamblin, 1981, and Larsson, 1995, as cited in Rekdal, 2014). Rekdal found no sign that the error was ever made. Anyone who cited Hamblin for it after reading only Larsson passed on two mistakes and a rumour. We could not open either older paper, so this paragraph is itself a secondary citation, and Rekdal is the only one of the three in our list.
Rekdal weighs each way of citing the sentence. Citing Larsson alone hides that Larsson was reporting someone else. Writing "Hamblin, 1981, cited in Larsson" is honest and legitimate when the original is hard to get, though here the 1981 BMJ was easy to find. Citing Hamblin without opening Hamblin, Rekdal calls "an academic lie".
How common the problem is
Simkin and Roychowdhury (2003) studied misprints in citations to one paper with about 4,300 citations: 196 misprints, only 45 of them distinct, one repeated 78 times. Identical misprints rarely happen by chance, so most were copied from other lists. Their model put the share of citers who had read the original, or at least a trusted record of it, at about 20%. That is a model's estimate from one paper, not a survey.
Baethge and Jergas (2025) pooled 46 studies of about 32,000 quotations in medicine: 16.9% did not support the claim they were cited for, about half of those seriously, with no improvement over time. An earlier review found indirect references accounted for less than one sixth of such errors (Jergas & Baethge, 2015), so reading the original does not solve everything. It is still the only way to know.
How do I write "as cited in" in each citation style?
Name the original, add the style's linking phrase, and cite and list the work you read. The table uses one example throughout: you read Larsson's 1995 paper, which cites Hamblin (1981).
| Style | In the text | In the reference list | Rule from |
|---|---|---|---|
| APA 7 | (Hamblin, 1981, as cited in Larsson, 1995) | Larsson only | American Psychological Association (2022) |
| MLA 9 | (Hamblin qtd. in Larsson 448–49) | Larsson only | Modern Language Association (2023); University of Queensland Library (2026) |
| Chicago 18 author-date | (Hamblin 1981, quoted in Larsson 1995, 448–49) | Guides differ: Larsson only, or both | University of Chicago Press (n.d.); SMU Libraries (2026); McGill Library (2026) |
| Harvard (Cite Them Right) | (Hamblin, 1981, cited in Larsson, 1995) | Larsson only | University of Westminster Library (2026) |
| IEEE | Not allowed: find and cite the original | The original, once read | Victoria University Library (2026) |
| Vancouver (NLM) | Hamblin (1) as cited by Larsson | One entry: the original, then "Cited by:" and Larsson | University of Western Australia Library (2026) |
| Any style | Build each entry from the record | Check the finished list | EdCitation's Cite a source and Verify references, free |
APA 7
APA keeps secondary sources for the rare case: an original you cannot obtain, one no longer in print, or one written in a language you cannot read (American Psychological Association, 2022). It adds that a textbook or lecture that reports research is not a reason, unless your instructor says otherwise. If the original's year is unknown, leave it out. Only the secondary source gets an entry:
Larsson, K. S. (1995). The dissemination of false data through inadequate citation. Journal of Internal Medicine, 238(5), 445–450. https://doi.org/10.1111/j.1365-2796.1995.tb01222.x
Our guide to APA 7 in-text citations has the narrative form and the rest of APA's citation rules.
MLA 9
MLA marks a borrowed quotation with "qtd. in" before the source you read, and the original author's name can go in your sentence, in the citation, or in a note (Modern Language Association, 2023). The University of Queensland Library (2026) notes that the MLA Handbook advises using the original wherever possible, and that only the work you read goes in the works-cited list. If your sentence already names Hamblin, the citation is just (qtd. in Larsson 448–49).
Chicago 18 author-date
Chicago discourages the practice. Its Q&A quotes paragraph 14.160 of the 18th edition: authors are expected to have examined what they cite, and when an original is unavailable, both it and the secondary source must be listed (University of Chicago Press, n.d.). That paragraph belongs to the notes system. For author-date, guides disagree: SMU Libraries (2026), at Southern Methodist University, list only the secondary source, and McGill Library (2026) lists both. The manual's author-date section is paywalled, so we could not settle it; ask your tutor. EdCitation's Chicago is the 18th edition author-date only.
Harvard (Cite Them Right)
The University of Westminster Library (2026), which follows Cite Them Right, uses "cited in" for an idea and "quoted in" for words, lists only the work you read, and says reading the original is preferable. The Cite Them Right entry, as Cite a source built it:
Larsson, K.S. (1995) 'The dissemination of false data through inadequate citation', Journal of Internal Medicine, 238(5), pp. 445–450. Available at: https://doi.org/10.1111/j.1365-2796.1995.tb01222.x.
IEEE and Vancouver
IEEE's own reference guide (IEEE, 2025) has no rule on secondary sources; we searched it. University guides fill the gap bluntly: Victoria University Library (2026) says IEEE does not allow secondary citations, and that an original you cannot locate should not be cited. For Vancouver, the recommendations of the International Committee of Medical Journal Editors (2026) ask authors to cite original research directly whenever possible. The University of Western Australia Library (2026), in a guide based on NLM's Citing Medicine, says researchers publishing a paper or thesis should not use secondary sources at all, and for students who must, it borrows a form from the AMA Manual: one numbered entry for the original followed by "Cited by:" and the work read.
- Hamblin TJ. Fake. BMJ. 1981;283(6307):1671-4. Cited by: Larsson KS. The dissemination of false data through inadequate citation. J Intern Med. 1995;238(5):445-50.
How do I find and read the original source?
Start from the reference list of the work you read. Then:
- Copy the original's reference exactly from the secondary source, with its page if one is given.
- Search it by author, year and a title word in EdCitation's Find sources, your library search or Google Scholar; the details may have been copied wrongly, as Simkin and Roychowdhury (2003) found they often are. If it will not turn up, my reference cannot be found works through the causes.
- Ask your library when your university does not hold it. Postgraduates at UWA are pointed to a document request service (University of Western Australia Library, 2026); most libraries have one.
- Read the passage the claim comes from, and check that it says what the secondary source says it says. In Rekdal's case it did not.
- Build the entry from the record with Cite a source, from the DOI, and compare it with the printed first page.
- Change the citation from "as cited in" to a direct one, and take the secondary source out of the list if you no longer cite it for anything else.
If the original truly cannot be had, keep the secondary citation; that is what it is for.
I already cited a source I didn't read: what now?
Fix it before you submit, the honest way: read it, relabel it, or remove it. St Andrews' guidance counts listing sources that were not used to prepare the work, or citing correct information from the wrong source, as falsification (University of St Andrews, 2026). Other universities word it differently; your own policy is the one that applies.
Before you hand it in
Mark every entry in your list that you have not opened. For each one:
- You can get it: read it, check the claim, and cite it directly.
- You cannot get it: make it a secondary citation, and list the work you read instead.
- The claim rests on it and you cannot check it: soften the sentence, support it with a source you have read, or cut it.
Do not add "as cited in" to a source you did read; that under-claims your work and confuses the list. Why did I lose marks for referencing covers the comments markers usually write.
After you have handed it in
Follow your university's procedure. St Andrews (2026) tells students who are unsure what is allowed to ask the assignment's teaching staff, which is sound anywhere. No one can promise how a marker will treat it.
What did EdCitation's checks show about secondary citations?
They confirm what can be looked up, not what you read. We ran three checks in September 2026.
A reference list with the chain in it
We pasted five real entries into Verify references. Four came back verified, including Hamblin: "Title, year and first author match a published record." The fifth, Simkin and Roychowdhury's 2003 paper, was marked doubtful, which the page shows as "check this", with the reason "A record with this title exists, but it was published in 2024, not 2003." The journal's own page puts it in volume 14 of 2003; the DOI was registered in 2024, and the record carries that date. "Check this" meant look, and looking settled it.
A paper with "as cited in" in it
We uploaded a short APA text citing (Hamblin, 1981, as cited in Larsson, 1995), with Larsson listed. The summary read "3 in-text citations · 3 entries · 0 orphan citations · 0 uncited entries": the citation was counted against Larsson, the source read. In a second copy we also listed Hamblin. The report said the Hamblin entry "is never cited in the text", with the fix "Cite it where it supports a claim, or remove it", which is APA's rule. In a third, citing and listing only Hamblin, everything passed, as it should: no tool can know whether you opened a paper. Pairing citations with entries is covered in in-text citations that do not match the reference list.
Building the original's entry
Given Larsson's DOI and Hamblin's, Cite a source built both entries; Larsson's is in the APA section above. For Hamblin it gave the title as "Fake", from the record, which has "Fake." with a full stop, while Rekdal (2014) prints it with an exclamation mark. Given the Simkin DOI, it left out the university department the record lists as a third author, but kept the record's year, 2024. A record is only as good as its publisher's entry, so compare the result with the paper's first page.
Which EdCitation tool helps with a secondary citation?
Verify references is the best check for the list you end up with, free with no account, because it looks every entry up in the publisher's record instead of trusting the paper; a chatbot asked to check references only writes more text. Every entry gets one of three verdicts (verified, "check this", not found) and a reason; a retraction is flagged; and when an index does not answer, the entry says could not check rather than pretending the source is missing. Upload the whole paper and it also pairs each citation with its entry, which is how it flagged the listed original in our second copy.
Once you hold the original, the free Cite a source turns its DOI, PubMed ID, ISBN, title or address into an entry in any of the six styles above, with a retraction screen. For a paper still in drafts, Pro ($8 a month) adds References from a file: every reference checked against its record and every in-text citation matched to the list, draft after draft. Max is $24 a month (pricing). No EdCitation tool reads the original for you or writes a word of your paper. That part is yours.
Quick questions
Can I cite something I haven't read?
Only as a secondary citation that names the work you did read, and only when you cannot reasonably get the original. IEEE guides say not to cite it at all.
Does the original go in my reference list?
Not in APA 7, MLA 9 or Cite Them Right Harvard: list only the source you read. Chicago guides disagree, and the Vancouver form names both in one entry.
What is the difference between "cited in" and "quoted in"?
In Harvard, "cited in" is for an idea and "quoted in" for the original's exact words. APA uses "as cited in" for both.
Can I cite a lecture or textbook instead of the study it describes?
APA asks you to find and cite the study itself, unless your instructor says otherwise. Build its entry with EdCitation's free Cite a source once you have read it.
Will a reference checker know if I didn't read a source?
No. Verify references confirms that each source exists and pairs citations with entries; whether you read it is something only you can make true.
References
- American Psychological Association. (2022). Secondary sources. APA Style. https://apastyle.apa.org/style-grammar-guidelines/citations/secondary-sources
- Baethge, C., & Jergas, H. (2025). Systematic review and meta-analysis of quotation inaccuracy in medicine. Research Integrity and Peer Review, 10, Article 13. https://doi.org/10.1186/s41073-025-00173-z
- IEEE. (2025). IEEE reference guide (Version 3.28.2025). IEEE Author Center. https://journals.ieeeauthorcenter.ieee.org/wp-content/uploads/sites/7/IEEE_Reference_Guide.pdf
- International Committee of Medical Journal Editors. (2026). Preparing a manuscript for submission to a medical journal. ICMJE Recommendations. https://www.icmje.org/recommendations/browse/manuscript-preparation/preparing-for-submission.html
- Jergas, H., & Baethge, C. (2015). Quotation accuracy in medical journal articles: A systematic review and meta-analysis. PeerJ, 3, Article e1364. https://doi.org/10.7717/peerj.1364
- McGill Library. (2026, September 22). Chicago (18th edition). Citation style guide. https://libraryguides.mcgill.ca/citations/chicago
- Modern Language Association. (2023, September 20). Where should the original author's name appear when citing indirect sources? MLA Style Center. https://style.mla.org/citing-indirect-sources/
- Penders, B. (2018). Ten simple rules for responsible referencing. PLOS Computational Biology, 14(4), Article e1006036. https://doi.org/10.1371/journal.pcbi.1006036
- Rekdal, O. B. (2014). Academic urban legends. Social Studies of Science, 44(4), 638–654. https://doi.org/10.1177/0306312714535679
- Simkin, M. V., & Roychowdhury, V. P. (2003). Read before you cite! Complex Systems, 14(3), 269–274. https://doi.org/10.25088/ComplexSystems.14.3.269
- SMU Libraries. (2026, August 17). In-text citation. Cite in Chicago style AD (18th edition). Southern Methodist University. https://guides.smu.edu/c.php?g=1433086&p=10638300
- University of Chicago Press. (n.d.). Citation, documentation of sources [Q&A]. The Chicago Manual of Style Online. https://www.chicagomanualofstyle.org/qanda/data/faq/topics/Documentation/faq0021.html
- University of Queensland Library. (2026, July 14). Indirect quotations. MLA 9th referencing style. https://guides.library.uq.edu.au/referencing/mla9/indirect-quotes
- University of St Andrews. (2026). Good academic practice: Guidance for students and frequently asked questions. https://www.st-andrews.ac.uk/policy/academic-policies-assessment-examination-and-award-good-academic-practice/good-academic-practice-guidance-for-students-frequently-asked-questions.pdf
- University of Western Australia Library. (2026, August 5). Indirect citations/secondary sources. Referencing style: Vancouver (based on Citing Medicine). https://guides.library.uwa.edu.au/vancouver/indirect_citations
- University of Westminster Library. (2026, September 9). Harvard style. Referencing your sources. https://libguides.westminster.ac.uk/referencing/harvard
- Victoria University Library. (2026, June 25). Getting started with IEEE referencing. IEEE referencing. https://libraryguides.vu.edu.au/ieeereferencing/gettingstarted