Academic writing article coverAcademic writing article cover

How Many References Should a Research Paper Have? Journal Rules

There is no single right number. The journal you submit to decides, and in the guidelines below the limit changes with the article type, and some article types have no reference limit at all. JAMA allows 10 references in a Research Letter and up to 125 in a Narrative Review (JAMA, n.d.). Nature says that "as a guideline, articles typically have no more than 50 references" (Nature, n.d.). EPL lists references for a Letter as "no limit" (EPL, n.d.). So find the article-type table in your target journal's guidelines before you count. For a class paper, the number, if there is one, is in your assignment brief or course handbook.

Each rule below was read on the journal's or university's own page on November 15, 2026. Journals revise these pages, so open the current version before you submit.

Reference limits in eight publications' author guidelines

Publication (field)Article typeWhat the guidelines saySource
Nature (multidisciplinary science)Article"As a guideline, articles typically have no more than 50 references." Additional references for Methods or Supplementary Information are not included in this count.Nature, n.d.
JAMA (medicine)Research LetterUp to 800 words of text and 10 referencesJAMA, n.d.
JAMA (medicine)Clinical trial"no more than 50-75 references"JAMA, n.d.
JAMA (medicine)Narrative Review"no more than 125 references"JAMA, n.d.
Emerging Infectious Diseases (infectious disease)Research; Dispatch; Research Letter50; 15; 10Emerging Infectious Diseases, 2026
Preventing Chronic Disease (public health)Original Research; Research Brief; Systematic ReviewUp to 35; up to 12; no limitPreventing Chronic Disease, n.d.
Psychological Medicine (psychiatry and psychology)Commentary; Letter to the EditorMax 20; max 9. No number given for original or review articlesPsychological Medicine, n.d.
IEEE Communications Standards Magazine (engineering magazine)Article"References are limited to fifteen (15)."IEEE Communications Society, n.d.
EPL (physics)LetterReferences "no limit"; seven typeset pages, references includedEPL, n.d.
The World Bank Economic Review (economics)Short Paper; standard submission"No formal limit, but should be concise and relevant"; 40 pages, references includedWorld Bank Economic Review, n.d.

Read the wording as well as the number. Nature calls its 50 "a guideline" and says Articles allow up to 50 references in the main text "if needed and within the average page budget" (Nature, n.d.). IEEE Communications Standards Magazine says "limited to." So Nature's 50 is a guideline, and the magazine's 15 is a limit.

The limit changes with the article type

Three of these journals list a reference limit beside most of their article types, and the range inside one journal is wide. At Emerging Infectious Diseases, a Research article gets 50, a Dispatch 15, and a Research Letter 10 (Emerging Infectious Diseases, 2026). Preventing Chronic Disease allows up to 35 for Original Research and up to 12 for a Research Brief, and sets no limit for a Systematic Review (Preventing Chronic Disease, n.d.). JAMA's range runs from the Research Letter to the Narrative Review in the table above.

A number from one journal doesn't carry over to the next, even for the same kind of paper. JAMA caps a systematic review without meta-analysis at 50-75 references (JAMA, n.d.), while Preventing Chronic Disease sets no limit for a Systematic Review.

Do references count toward the length limit?

Some journals cap length instead of the number of references. Then the question is whether the reference list is inside the length.

  • EPL sets no limit on references for a Letter, but all articles "are limited to a maximum of seven typeset pages, including the title, abstract, references, and everything else" (EPL, n.d.). Every line of references takes a line from the text.
  • The World Bank Economic Review keeps a Short Paper's main text under 7,000 words, references excluded. A standard submission "should not exceed 40 pages," and the 40 pages include references (World Bank Economic Review, n.d.).
  • IEEE Communications Standards Magazine's 4,500-word count leaves out references, and its separate cap of 15 still applies (IEEE Communications Society, n.d.).

So check two things in the guidelines: a cap on the number, and whether references sit inside the word or page count.

When there's no limit, what decides the number?

EPL's Letters and Preventing Chronic Disease's Systematic Reviews have no reference limit, and Psychological Medicine gives no number for original or review articles (EPL, n.d.; Preventing Chronic Disease, n.d.; Psychological Medicine, n.d.). Two of the guidelines say what the list is for. For a Short Paper, The World Bank Economic Review asks for references that are "concise and relevant" (World Bank Economic Review, n.d.). IEEE Communications Standards Magazine says references "should be included to guide readers to more information on the topic of the manuscript," and "the reference list need not include every available source in the subject matter" (IEEE Communications Society, n.d.).

If you want a sense of a journal's norm without a stated limit, count the references in a few recent articles of the same type in that journal. That tells you about those articles only; it isn't a rule.

Over the limit? Where to cut

Start with what the journal lets you move. At Nature, additional references for the Methods or Supplementary Information are not included in the count of 50 (Nature, n.d.).

Next, look for points backed by more sources than the point needs. IEEE Communications Standards Magazine's line that the list "need not include every available source" applies here. The same guidelines ask for "primary literature in favor of reviews," to give credit to the group that first reported a finding (IEEE Communications Society, n.d.). Where a review and the original study support the same sentence, that rule keeps the original.

Then clean up after each cut. A deleted entry must leave the text too, and in a numbered style every later number moves. The steps are in in-text citations don't match the reference list.

Class papers

For coursework, the University of Suffolk's library answers: "There isn't a magic number for how many citations you should include in your reference list." The number depends on "your subject area, the type of work you're doing," the assignment brief, and what your lecturer expects. Its advice is to read the brief and course handbook, check whether you were told to use a specific number of sources, and look at feedback on earlier assignments (University of Suffolk, 2023).

Check the references you keep

A count says nothing about whether each entry is right. Before you submit, run the rest of the reference list checklist, including whether each entry matches its published record.

In PerfectRefs, you paste your reference list; you don't upload the paper. The Existence check looks up each reference in Crossref, OpenAlex, and other scholarly databases and flags any it can't find. It comes with a free daily allowance. The Style check compares authors, title, journal, year, volume, issue, pages, and DOI field by field with the database record, shows the record's values with a link to the source, and lets you accept a fix with one click. It uses premium credits (2 per reference), and new accounts get 10 premium credits. It doesn't know your journal's limit, so counting against the cap is still up to you.

FAQ

How many references should a research paper have?

No number applies everywhere. The journal sets it, and the number can change with the article type: 10 for a JAMA Research Letter, 50 for a research article in Emerging Infectious Diseases, and no limit for a Letter in EPL (JAMA, n.d.; Emerging Infectious Diseases, 2026; EPL, n.d.).

How many references can a review article have?

It depends on the journal. JAMA allows up to 125 for a Narrative Review and 50-75 for a systematic review without meta-analysis; Preventing Chronic Disease sets no limit for a Systematic Review (JAMA, n.d.; Preventing Chronic Disease, n.d.).

Do references count toward the word count?

At some journals. IEEE Communications Standards Magazine leaves them out of its 4,500 words; EPL counts them in its seven pages (IEEE Communications Society, n.d.; EPL, n.d.).

Can PerfectRefs tell me if I have too many references?

No. It checks each entry against database records, not the total against a journal's limit.

References

  1. Emerging Infectious Diseases. (2026, September 15). Article types. Centers for Disease Control and Prevention. https://wwwnc.cdc.gov/eid/article-types
  2. EPL. (n.d.). Author guidelines. Oxford Academic. https://academic.oup.com/epl/pages/author-guidelines
  3. IEEE Communications Society. (n.d.). IEEE Communications Standards Magazine policies and guidelines. https://www.comsoc.org/publications/magazines/ieee-communications-standards-magazine/author-guidelines
  4. JAMA. (n.d.). Instructions for authors. JAMA Network. https://jamanetwork.com/journals/jama/pages/instructions-for-authors
  5. Nature. (n.d.). Formatting guide. https://www.nature.com/nature/for-authors/formatting-guide
  6. Preventing Chronic Disease. (n.d.). Types of articles. Centers for Disease Control and Prevention. https://www.cdc.gov/pcd/for_authors/types_of_articles.htm
  7. Psychological Medicine. (n.d.). Preparing your materials. Cambridge Core. https://www.cambridge.org/core/journals/psychological-medicine/information/author-instructions/preparing-your-materials
  8. University of Suffolk. (2023, July 14). How many citations should I have in my reference list? Library and Learning Services. https://uos.libanswers.com/library-learning-services/faq/268799
  9. World Bank Economic Review. (n.d.). Author guidelines. Oxford Academic. https://academic.oup.com/wber/pages/Author_Guidelines

Original article by PerfectRefs. To republish it, credit PerfectRefs and link to this page. Do not edit it or use it commercially.