What Is a Secondary Source? (And When Can You Cite One?)

The definition, and why the label moves

Primary sources are the original, first-hand record — the study, the letter, the dataset, the artifact. Secondary sources stand one step away, interpreting or summarizing that record. The label is relational: the same object changes roles with the question. Write about 1936 election coverage and the newspapers are primary; write about how the election happened and the same newspapers are secondary, because historians' monographs are doing the interpreting for you.

Tertiary sources — encyclopedias, Wikipedia, most study guides — sit another step out: compiled from secondaries. They are for orienting, never for evidence, and citing them in a paper is the fastest way to tell a grader the research never started. Why any of this matters: a paper built mostly on secondaries is conclusions about conclusions. Course assignments want you near the data, because working with primary evidence is the skill being graded.

The numbers at a glance

Primary versus secondary, field by field — 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.

When citing secondary work is acceptable — even expected

Legitimate uses are broad. Background and context paragraphs lean secondary by design. Literature reviews are largely secondary work — that is their job. Theory frameworks cite the exposition of the theory, not its every experiment. And secondary research is real scholarship: review articles and meta-analyses synthesize primary studies, and in evidence-based medicine a systematic review outranks any single trial it aggregates.

The failure pattern is narrower and familiar: citing a textbook for a study's finding instead of citing the study. Textbook summaries compress — a sample of 14 becomes "studies show" — and the compression is invisible until you need the detail. The working rule: chase the original, fifteen minutes via Google Scholar and your library's link resolver, most of the time. When the original is truly unreachable, the honest fallback is APA's (Smith, 2020, as cited in Jones, 2024), capped at two or three per paper.

The discipline flips that trip students

Psychology students cite a Psychology Today write-up of a study that sits one database click away — secondary where primary was cheap. History students quote a soldier's diary through a documentary — a secondary quotation of a primary source, when the archive transcription might be findable. Both flip the moment the student asks what the paper is using the source to prove. Both mistakes also cost the same five minutes of database time to fix.

The genuinely tricky cases are the double roles. A meta-analysis is secondary research — it analyzes studies — yet in a literature review about meta-analytic method, it becomes the primary object of analysis. A biography is secondary for its subject's life and primary for its author's era. The question that resolves every case: what am I using this source to demonstrate? Answer that first, and the label assigns itself.

Finding primary sources through library databases

The mechanics are unglamorous and effective. Filter by document type — in PsycINFO, the Methodology filter set to "Empirical Study" separates original research from commentary in one click. Original research articles have Method and Results sections; that structure is the fifteen-second tell when a filter is unavailable. Putting "review" in your search terms deliberately locates the secondaries when you want them, instead of by accident.

From secondary to primary: every review article's bibliography is a pre-screened list of primary studies, and mining one strong review's references usually surfaces ten to twenty originals in an afternoon — snowballing, in the trade. Then cite the originals you actually used, formatted individually; the citation generator handles the per-entry formatting. The review that led you there gets cited once, in the background paragraph where it belongs.

A five-minute audit of your own source list

Before submitting, run the one-sentence test on every reference: can you state, without opening the PDF, what this source argues and where it supports your paper? Sources that fail the test are collected, not used — cut them or promote them by actually integrating their evidence. Count what remains; that number, not the folder in your reference manager, is what satisfies the requirement.

Two composition checks finish the audit. First, no more than a third of your list should be non-scholarly — news pages, organizational posts, encyclopedias. Second, your three or four load-bearing sources should appear in the text two or more times each; if every source appears exactly once, your paper is a relay of summaries, and the fastest fix is deepening the analysis around your best two sources. The citation generator below formats whatever survives the audit in APA, MLA and Chicago.