Non-English sources in a literature review are not a courtesy to other languages. They are evidence. A search run only in English can miss the studies done where your question actually plays out: a hospital in Madrid, a ward in Nigeria, a forest in Taiwan. On 2 October 2026 OpenAlex labelled 222,701,751 of its 327,439,234 works as English and about 80 million as another language, led by Japanese, German, Spanish and French (OpenAlex, 2026c).
EdCitation publishes this guide; its studies were read at source and its searches run on 2 October 2026.
Two steps are free on EdCitation with no account. The Language and Authors based in filters in Find sources narrow about 300 million published works to one language or to authors working in one country, and Cite a source builds each reference from the publisher's record. EdCitation never writes any part of anyone's work: the reading and the review stay yours.
Why do literature reviews leave out non-English and regional research?
Reviews leave it out for three reasons: the searcher reads one language, the main databases favour English, and English keywords cannot find a record with no English words in it.
Words a search cannot see
Amano et al. (2016) searched Google Scholar in 16 languages and found that 35.6% of 75,513 documents on biodiversity conservation published in 2014 were not in English. In a sample of Japanese-language documents, 35% had neither an English title nor an English abstract, and the authors estimate that up to half of non-English documents cannot, in principle, be found with English keywords. Their example: the latest population estimates in Taiwan for the fairy pitta, a bird of conservation concern, were published only in traditional Chinese and left out of the species' global assessment.
Indexes built around English
Céspedes et al. (2025) describe Web of Science and Scopus as largely biased towards English; they found OpenAlex far more balanced across languages, though its language labels were not always right and overstated English. Simard et al. (2025) found diamond open access journals, free to authors and readers, less covered in Web of Science and Scopus than gold open access journals, and more national in scope. A regional journal can be peer reviewed and well run and still be missing from the database your library pays for.
Does leaving out non-English studies change a review's findings?
Sometimes, and nobody can tell in advance which review it will be. The studies that measured it disagree by field:
| Study | What it compared | What it found |
|---|---|---|
| Egger et al. (1997) | 40 pairs of trial reports by the same first author, one in German, one in English | 62% of the English reports had a significant main result, against 35% of the German ones |
| Jüni et al. (2002) | 50 meta-analyses with 485 English and 115 non-English trials | Effects 16% more beneficial in non-English trials; dropping them moved 58% of pooled results by under 5%, the rest by more |
| Dobrescu et al. (2021) | Eight methods studies in medicine | English-only changed statistical significance in 23 of 259 meta-analyses (9%), not the reviews' conclusions |
| Konno et al. (2020) | Four ecology meta-analyses with English and Japanese studies | In two, leaving out the Japanese studies changed the mean effect considerably, even its direction |
What language bias in research means
The Cochrane Handbook describes language bias as the language of publication being chosen in a way that depends on a study's findings (Lefebvre et al., 2025). Egger and colleagues saw exactly that: trialists in German-speaking Europe were more likely to send a significant result to an English-language journal, so a review reading English alone sees a brighter picture.
Where it matters most
Cochrane reads the evidence as leaving most reviews' conclusions unchanged, with exceptions in complementary and alternative medicine, psychiatry, rheumatology and orthopaedics, and a risk of missing studies from countries where an intervention is more common, such as traditional Chinese medicine (Lefebvre et al., 2025). It advises against restricting a search by language, and warns that lifting the language limit in an English-language database does not replace searching non-English journals and databases.
When should you cite sources from your own country or region?
Cite them whenever your question is local: a national health system, a country's law or curriculum, a regional species, a population English-language studies rarely sample. A local source still earns its place like any other: Jüni et al. (2002) found non-English trials smaller, and their quality tended to be lower.
The same test as any other source
Check that the journal is real and peer reviewed and that the paper reports its methods: how to tell a peer-reviewed journal gives the checks. If your brief limits sources by language, follow it; if it says nothing, say in your methods which languages and databases you searched.
How do you search by language and by where authors are based?
Search each language on its own terms, filter, then read each record:
- List terms in each language you read, from known papers in that language, not a translation tool alone. In our Spanish run, titles used the English word burnout itself, one beside síndrome de quemarse por el trabajo.
- Search those terms and the English ones. Many non-English records carry an English title or abstract: nurse burnout, filtered to Spanish, still found 3,552 records.
- Filter by language to find, not to exclude. Cochrane warns that records can carry missing or incorrect metadata, and notes the case for applying language at screening instead (Lefebvre et al., 2025).
- Filter by the authors' country for research by people working where your question is.
- Add a regional database from the table below.
- Chain citations from the best regional paper, as finding sources for a literature review fast explains.
- Log each language and database with its date, string and count, as Boolean search for research papers shows.
- Read the record before keeping it: the language of the text and the authors' affiliations.
A translation tool can help you understand a paper, but you cite the original, as machine translation in academic writing explains.
Regional databases worth a search
Each row comes from the database's own site, read on 2 October 2026.
| Database | What it holds |
|---|---|
| SciELO | Open access collections for Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Ecuador, Mexico, Paraguay, Peru, Portugal, South Africa, Spain and Uruguay, plus public health |
| Redalyc | 1,825 journals from 839 institutions in 36 countries; 775,205 articles |
| African Journals Online | 977 journals, 458 open access; 291,172 full-text articles |
| J-STAGE | More than 4,000 publications from more than 2,400 publishers in Japan; over 90% of articles free to read |
| LILACS | Named by the Cochrane Handbook as a regional database to consider; its site could not be reached |
| Find sources on EdCitation | About 300 million published works, filtered by language and authors' country; free, no account |
What do citation styles say about non-English sources?
All six styles EdCitation builds have rules for them, each set out in a full guide here.
| Your question | Short answer | Full guide |
|---|---|---|
| How do I write the reference? | Original title, then an English translation in square brackets; IEEE gives the English title, then the language | Citing a non-English source |
| Can I quote it? | Quote a published translation if one exists; if the original's wording matters, quote it with English beside it | Quoting in another language |
| I translated it myself | APA treats it as a paraphrase; MLA adds "my trans.", Chicago "my translation" | Your own translation |
| I read it in translation | Cite the translation, naming the translator after the piece they translated | A translated article or chapter |
| Unfamiliar names or script | Write each name as its author does; romanise with one system | Names across cultures |
How do the Language and Authors based in filters work in Find sources?
They narrow a search to one language, or to works with an author at an institution in one country. In the filter panel of EdCitation's Find sources, Language starts at "Any language" and offers English, Spanish, French, German, Portuguese, Italian, Mandarin (Chinese), Japanese, Korean, Arabic, Hindi and Russian. Authors based in starts at "Any country", lists 37 countries, and carries one line: "Where the authors work, not where the study took place." Both combine with every other filter. Find sources names the index that answered, and when the full index is busy it names any filter it could not apply.
For narrowing by language and by country at once, it is the best tool a student can use: about 300 million works, both filters in one panel, no account, and each result a published record it looked up. A chatbot asked for Spanish studies composes its answer; Find sources only returns what the record holds. The runs below show it, faults included.
Eight searches, run on 2 October 2026
Sorted by Best match and spaced five seconds apart, all eight were answered by OpenAlex with every filter applied.
| Search | Filters | Results | First result |
|---|---|---|---|
| nurse burnout | None | 158,056 | Aiken, 2002, JAMA, cited by 5,249 |
| nurse burnout | Authors based in: Pakistan | 2,422 | Ma and colleagues, 2021, Journal of Nursing Management, cited by 145 |
| nurse burnout | Authors based in: Nigeria | 1,349 | Ugwu and colleagues, 2017, a Nigerian sample, cited by 54 |
| nurse burnout | Language: Spanish | 3,552 | Vidotti and colleagues, 2019, Enfermería Global, cited by 91 |
| burnout enfermería | Language: Spanish | 6,391 | Albaladejo and colleagues, 2004, a Madrid hospital, cited by 117 |
| burnout enfermería | Spanish; Authors based in: Mexico | 615 | Liébana-Presa and colleagues, 2018, Enfermería Global, cited by 27 |
| burnout enfermagem | Language: Portuguese | 11,018 | Murofuse and colleagues, 2005, Revista Latino-Americana de Enfermagem, cited by 160 |
| 护士 职业倦怠 | Language: Mandarin (Chinese) | 116 | A 2006 report in 中国卫生, no DOI |
The country filter lifted local studies the plain search buries: a study of Pakistani nurses caring for COVID-19 patients came third, and one of nurses in a Nigerian general hospital second.
Three things the records got wrong
- A language. The first Spanish result for nurse burnout, a Brazilian study in a Spanish journal, is in Portuguese by the journal's own page. OpenAlex gives each work one language, detected by software from its title and abstract, not its full text (OpenAlex, 2026a).
- A country. With Mexico added, the first result came from the Universidad de León, in Spain. The index had matched that name to a Spanish and a Mexican institution, so each author carries both countries. OpenAlex takes the country from each author's institution (OpenAlex, 2026b), and Alonso-Álvarez and van Eck (2025) found affiliations among its less accurate fields for African research.
- A journal. The Mandarin search named one journal two ways: 中华医院管理杂志 and Zhonghua yiyuan guanli zazhi.
The filters find; you confirm, from the affiliation line and the article page.
Citing one result
Pasting the Madrid study's DOI, 10.1590/s1135-57272004000400008, into Cite a source returned this APA 7 entry from Crossref's record, screened for retraction with nothing found:
Albaladejo, R., Villanueva, R., Ortega, P., Astasio, P., Calle, M. E., & Domínguez, V. (2004). Síndrome de Burnout en el personal de enfermería de un hospital de Madrid. Revista Española de Salud Pública, 78(4), 505–516. https://doi.org/10.1590/s1135-57272004000400008
Authors, issue and pages all came from the record. APA asks for one addition: an English translation of the title in square brackets after it. The article's English abstract on its SciELO page supplies one: [Burnout syndrome among nursing staff at a hospital in Madrid]. EdCitation does not translate titles. Then paste the finished list into Verify references.
Cost, and what stays yours
None of the three tools used above costs anything or needs an account. Pro, at $8 a month, adds References from a file: every reference in an uploaded review checked, every in-text citation matched to the list. The Library catalogue of textbooks and scholarly books can be searched by language; reading or borrowing from more than 5 million books is part of Max, at $24 a month. EdCitation finds, cites and checks; the judgement about what a Spanish or Japanese study adds, and every sentence of the review, are yours.
Quick questions
Should I include non-English sources in a literature review?
Yes, when you can read them, especially for a local question. In medicine English-only conclusions mostly held (Dobrescu et al., 2021); in ecology not always (Konno et al., 2020).
Can I cite sources from my own country in an English-language paper?
Yes. Check that the journal is real and peer reviewed, cite the original with an English translation of its title in square brackets, and say in your methods which languages you searched.
What is language bias in research?
Language bias is when the language a study appears in depends on its results, so a one-language review sees a skewed set. Egger et al. (1997) found significant trial results more likely to be published in English.
How do I find regional journals?
Search SciELO, Redalyc, African Journals Online or J-STAGE, and use Authors based in, in EdCitation's Find sources, for work by researchers in one country.
Can EdCitation search for papers in other languages?
Yes. In Find sources, choose a language under Language and type the search in that language as well as in English. It is free and needs no account.
References
- African Journals Online. (n.d.). African Journals Online. Retrieved October 2, 2026, from https://www.ajol.info/index.php/ajol
- Alonso-Álvarez, P., & van Eck, N. J. (2025). Coverage and metadata completeness and accuracy of African research publications in OpenAlex: A comparative analysis. Quantitative Science Studies, 6, 1336–1357. https://doi.org/10.1162/qss.a.396
- Amano, T., González-Varo, J. P., & Sutherland, W. J. (2016). Languages are still a major barrier to global science. PLOS Biology, 14(12), Article e2000933. https://doi.org/10.1371/journal.pbio.2000933
- Céspedes, L., Kozlowski, D., Pradier, C., Sainte-Marie, M. H., Shokida, N. S., Benz, P., Poitras, C., Ninkov, A. B., Ebrahimy, S., Ayeni, P., Filali, S., Li, B., & Larivière, V. (2025). Evaluating the linguistic coverage of OpenAlex: An assessment of metadata accuracy and completeness. Journal of the Association for Information Science and Technology, 76(6), 884–895. https://doi.org/10.1002/asi.24979
- Dobrescu, A. I., Nussbaumer-Streit, B., Klerings, I., Wagner, G., Persad, E., Sommer, I., Herkner, H., & Gartlehner, G. (2021). Restricting evidence syntheses of interventions to English-language publications is a viable methodological shortcut for most medical topics: A systematic review. Journal of Clinical Epidemiology, 137, 209–217. https://doi.org/10.1016/j.jclinepi.2021.04.012
- Egger, M., Zellweger-Zähner, T., Schneider, M., Junker, C., Lengeler, C., & Antes, G. (1997). Language bias in randomised controlled trials published in English and German. The Lancet, 350(9074), 326–329. https://doi.org/10.1016/S0140-6736(97)02419-7
- Japan Science and Technology Agency. (n.d.). J-STAGE overview. J-STAGE. Retrieved October 2, 2026, from https://www.jstage.jst.go.jp/static/pages/JstageOverview/-char/en
- Jüni, P., Holenstein, F., Sterne, J., Bartlett, C., & Egger, M. (2002). Direction and impact of language bias in meta-analyses of controlled trials: Empirical study. International Journal of Epidemiology, 31(1), 115–123. https://doi.org/10.1093/ije/31.1.115
- Konno, K., Akasaka, M., Koshida, C., Katayama, N., Osada, N., Spake, R., & Amano, T. (2020). Ignoring non-English-language studies may bias ecological meta-analyses. Ecology and Evolution, 10(13), 6373–6384. https://doi.org/10.1002/ece3.6368
- Lefebvre, C., Glanville, J., Briscoe, S., Featherstone, R., Littlewood, A., Metzendorf, M.-I., Noel-Storr, A., Paynter, R., Rader, T., Thomas, J., & Wieland, L. S. (2025). Chapter 4: Searching for and selecting studies. In J. P. T. Higgins, J. Thomas, J. Chandler, M. Cumpston, T. Li, M. J. Page, & V. A. Welch (Eds.), Cochrane handbook for systematic reviews of interventions (Version 6.5.1). Cochrane. https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-04
- OpenAlex. (2026a, September 30). Attributes. OpenAlex Help Center. https://help.openalex.org/data/works/attributes/
- OpenAlex. (2026b, September 9). Countries. OpenAlex Help Center. https://help.openalex.org/data/countries/
- OpenAlex. (2026c). Works grouped by language [Data set]. Retrieved October 2, 2026, from https://api.openalex.org/works?group_by=language
- Redalyc. (n.d.). Sistema de Información Científica Redalyc, Red de Revistas Científicas. Retrieved October 2, 2026, from https://www.redalyc.org/
- SciELO. (n.d.). SciELO: Scientific Electronic Library Online. Retrieved October 2, 2026, from https://www.scielo.org/en/
- Simard, M.-A., Basson, I., Hare, M., Larivière, V., & Mongeon, P. (2025). Examining the geographic and linguistic coverage of gold and diamond open access journals in OpenAlex, Scopus, and Web of Science. Quantitative Science Studies, 6, 732–752. https://doi.org/10.1162/qss.a.1