How to Cite a Statistic

The cite-what-published-it rule

No style guide has an entry format for a statistic, because a number is not a source: it is a finding inside a source. The citation therefore names whoever published it. In APA 7, a Bureau of Labor Statistics unemployment figure is cited in text as (U.S. Bureau of Labor Statistics, 2024) and in the reference list as the report or data table where the figure appears, with its title in italics and a stable URL. MLA 9 names the agency in prose, along the lines of According to the Bureau of Labor Statistics, and locates the figure by table or section number when the source has no pages. The how-to-cite-sources guide carries the general anatomy. What changes for a statistic is only which source earns the entry, not the entry's shape itself.

The rule has a corollary that graders check first: cite the report you actually opened. If a database aggregator displayed the number, the aggregator is a container, not a source, and the reference goes to the agency's own document, the same container logic the how-to-cite-a-website page applies to web sources. This is also the cite-what-you-read rule in its most testable form, because a misquoted statistic is worse than no statistic: the reader who follows your citation must find the exact number you used, with the exact wording of its conditions, or the citation has failed at its only job. That is why the fix for a shaky statistic is rarely a formatting correction and usually a better source: find the document where the number was computed, and the citation formats itself.

The numbers at a glance

Statistic citation cases and their fixes — the range, and where each row actually shows up in coursework. Skim it first; the sections below explain why the range exists and which row applies to you.

Datasets, self-computed figures and the personal-communication ceiling

Two statistic cases come from you rather than from a published page. First, figures you computed yourself from a dataset: cite the dataset as the source, follow the how-to-cite-a-dataset page for the entry anatomy, and describe the computation in text, in the form author's calculation from the 2023 wave. The description is what makes the number checkable, and graders in methods-heavy courses look for it. Second, a figure an expert gives you in conversation, lecture or email: APA 7 allows it as an in-text-only citation marked personal communication with the date, and it never enters the reference list because no reader can retrieve it. Both restrictions exist for the same reason: a citation the reader cannot follow is a dead end, and dead ends cannot carry claims in graded work.

That non-retrievable ceiling is the reason personal-communication statistics are weak evidence in graded work. A number no reader can verify cannot carry a claim alone, so treat it as a lead: find the published figure that supports the same point, or mark the claim as preliminary in your own prose. Lecture slides sit in the same territory, and the how-to-cite-a-lecture page covers them; the pattern is consistent across the whole citation system, that the weaker the retrievability, the less weight the source can bear. Statistics earn their reputation as a paper's strongest evidence precisely when they resolve to a document the reader can open. The lecture case also explains the exception's exact wording: personal communication citations carry the date because the only verification possible is asking the source again, and dates make that possible.

The secondhand trap and the as-cited-in chain

The most common statistic miscitation in undergraduate writing is citing the number to the news article that mentioned it, or worse, to another student's paper. The honest options, in order of preference: chase the original agency report the article was covering, which takes one search of the agency's site and upgrades the citation from journalism to primary source; or, when the original is genuinely unreachable, use the secondary-citation form, APA 7's as cited in or MLA 9's qtd. in. The what-is-a-secondary-citation page covers the mechanics, and the cost is deliberate: the form tells the grader you never read the original. That admission is the form's design: honest about the shortcut, which is why sparing use reads as judgment and constant use reads as research that never left the news cycle.

Secondary form is a tool, not a habit. A reference list with three or more as-cited-in entries reads as a research-depth problem before a reader checks a single source, because statistics are exactly the category where the original is almost always reachable: agencies publish their tables and journals archive their abstracts, and the how-to-cite-a-government-website page covers the agency-site anatomy. Budget ten minutes per statistic for the chase and the list stays primary. The exception that proves the rule is the archived or paywalled original the course cannot access; that is what the secondary form was built for, and using it once, correctly, signals judgment rather than laziness. Everything else in the list should resolve to a document, because at grading time the difference between the two lists is visible without reading a single source.

Reproducing statistics: tables, figures and the honesty notes

Copying a statistic into prose needs only the in-text citation, but reproducing the table or chart it came from triggers the figure machinery: a number, a title, and a note crediting the source. APA 7's form is a note reading Adapted from the source's title, with author, year, publisher and the table or page locator. The note is where honesty lives: adapted means you changed something, from means you copied it, and the reader can judge the transformation. A chart rebuilt in a spreadsheet from someone else's numbers is adapted, and saying so costs nothing, while omitting the note is the accidental plagiarism the how-to-avoid-plagiarism guide warns about. The same machinery applies to charts you redraw: changing the colors is not authorship, and the note costs one line while the integrity risk costs a hearing.

Two audits close the loop. First, traceability: every statistic in the draft resolves to a reference entry, and every agency name appears the same way each time, with the APA rule of spelling the name out on first use followed by its bracketed abbreviation. Second, dating: statistics go stale faster than any other evidence, so a source with no date takes the n.d. form covered by the how-to-cite-a-source-with-no-date page, and fast-moving series carry a retrieval date when the content is designed to change. The source-credibility-checker scores the publishing agency in a minute, which is the last check before the citation earns its place. Ten minutes per statistic is the realistic audit budget at essay length, and it is the same ten minutes the original-source chase costs, which is why the two audits share a morning.

Three minutes to a clean citation

Edge-case citations punish hesitation, not ignorance: the student who spends twenty minutes googling a no-author format loses more time to doubt than the format ever costs. Decide the pattern once — title moves forward, organization counts as author, n.d. when no date exists anywhere — and apply it mechanically. Consistency across your list matters more to graders than perfect recall of manual arcana.

Automate what's mechanical. The citation generator below formats journal articles, books, chapters and websites in all three major styles, including the awkward cases on this page. Paste the metadata, copy the output, done — the minutes you save belong to analysis, which is where rubric points actually live. Cross-check in-text forms against the table above and the reference list is submission-ready.