Research a questionPhase 6 · Check and share
Cite So Others Can Check
A citation's job is to let a reader find the exact thing you relied on, down to the page, paragraph or minute. Cite the original you actually read, with a stable identifier and an archived link.
Purpose
A citation is directions. It tells a reader where to go to see for themselves what you saw. If the directions are vague (“Smith, 2019”), out of date (a dead link), or point to the wrong place (a news story about a study instead of the study), the reader has to take your word for it. That defeats the point.
So a good citation answers three questions. What is it: author, title, date, and who published it. Where exactly: the page, paragraph, section or timestamp that holds the claim. How do I get there: a stable identifier like a DOI, a link, and an archived copy of the link.
APA, MLA, Chicago and other style guides are conventions for arranging that information. Follow the one your teacher, journal or editor asks for. But the style is the packaging. A perfectly formatted citation that points to the wrong source, or to a 400-page report with no page number, still fails the reader.
When you need it
Any claim in your work that a reader might want to check: a number, a quote, a finding, a date.
You’re citing a long work (a book, a report, a court opinion) for one specific point.
You’re citing a web page, a video, a podcast or a social media post, which can change or vanish.
You learned about a study from a news story, a textbook or another paper.
You’re preparing a reference list for a paper, a thesis or an article.
How to do it
1Cite what you actually read
Cite the source you looked at yourself. If you read about a study in a news article and never opened the study, either go get the study or cite the article and say where it got the claim. APA’s guidance is to use secondary citations sparingly, for example when the original is out of print or in a language you can’t read, and to mark them with “as cited in.” Citing an original you never opened passes off someone else’s reading as your own.
2Get the full details from the source itself
Take author names, title, date, volume, issue and page range from the article’s own page or PDF, not from someone else’s reference list, where errors get copied. For journal articles, look for the DOI (Digital Object Identifier), a permanent ID such as 10.1371/journal.pmed.0020124. Adding https://doi.org/ in front of it makes a link that keeps working even if the publisher moves the article. If you have only a title, Crossref’s search (search.crossref.org) or Google Scholar will usually turn up the DOI.
3Pinpoint the exact spot
Point to where the claim lives, not just the work it’s in. That’s a pinpoint citation. Use page numbers for print and PDFs. For web pages without page numbers, APA suggests a paragraph number (count them if needed) or a section heading. For video and audio, give a timestamp. For a long report, name the section, table or figure.
4Add the link and an archived copy
For anything online, give the URL, and for anything without a DOI, also an archived copy made with the Wayback Machine, archive.today or Perma.cc. Web pages change and disappear often enough that a bare URL is a bet. See Archive Your Links for how.
5Note the version and the date
Reports get revised, data get updated, articles get corrected, web pages get edited. Say which version you used: the edition, the release date, or the date you accessed a page that changes. If the source has a correction notice, read it and check whether it affects the part you cite.
6Test your own citations
Before you publish, pick a few citations and follow them as a stranger would: click the link, open the DOI, find the page or timestamp. If you can’t get to the exact passage in a minute or two, neither can your reader.
Try it
Your task
A draft says: “Small studies are less likely to produce true findings (Ioannidis).” Turn that into a citation a reader could check: the full reference with a DOI, plus a pinpoint showing where in the paper the point is made. Then check whether the paper has been corrected since publication, and whether the correction matters for this claim.
Show a hint
The paper is a famous 2005 essay whose title begins “Why Most Published Research Findings…”. It’s open access. Look at the section headings.
Show the answer and the path
The full reference, in APA style:
Ioannidis, J. P. A. (2005). Why most published research findings are false. PLoS Medicine, 2(8), e124. https://doi.org/10.1371/journal.pmed.0020124
The pinpoint: the claim is the first of six numbered corollaries in the section titled “Corollaries.” Corollary 1 says that the smaller the studies in a field, the less likely its research findings are to be true. A good in-text citation names it: (Ioannidis, 2005, “Corollaries” section, Corollary 1). The article has no page numbers online (it uses the article number e124), so a section and corollary number do the job a page number would.
The correction: the article page on journals.plos.org shows a correction dated August 25, 2022 (PLOS Medicine 19(8): e1004085). It fixes a missing set of parentheses in one equation in Table 2. It doesn’t change Corollary 1, so the citation stands, but it’s worth knowing it exists.
The path: search the title in Crossref or Google Scholar, open the DOI, take the details from the journal’s own page, then skim the headings to find where the claim lives.
Common mistakes
Citing the summary, not the source. A press release or news story about a study is a different source from the study. Cite what you read.
Copying references from another paper’s reference list. Typos, wrong years and wrong page numbers spread this way. Take details from the source itself.
Citing a whole book or report for one fact. Without a page or section, a reader has to search hundreds of pages.
Bare URLs. A link with no author, title or date is useless once it breaks, because the reader has nothing to search for.
Linking to a search result, a session URL or a personal login page. The reader will see something different, or nothing. Link the item’s own permanent address.
Treating style as the goal. Fretting over comma placement while the page number is missing gets the priorities backward.
Limits
Some sources can’t be pinned down well: a conversation, an unpublished dataset, a document you were given in confidence, a page behind a paywall. Say what you can (“interview with the author, March 2026”; “internal report, not public”) so a reader at least knows what kind of evidence it is and why they can’t check it.
A citation shows where a claim came from, not that the claim is right. A correctly cited weak study is still weak. Judging the evidence is a separate job.
Identifiers help but aren’t magic. The DOI Foundation itself says persistence depends on the organizations that maintain the links, not on the technology alone.
Sources & further reading
- APA Style, “Secondary sources”: use them sparingly and mark them “as cited in”
- APA Style, “Direct quotation of material without page numbers”: paragraph numbers, headings and timestamps
- DOI Foundation, “What is a DOI?”
- Crossref Search: look up a DOI from a title
- John P. A. Ioannidis, “Why Most Published Research Findings Are False,” PLoS Medicine, 2005
- Correction to “Why Most Published Research Findings Are False,” PLOS Medicine, 2022
End of page · Back to top