Language bias in global university rankings: the English-language advantage

The English-language dominance in bibliometric databases

The two major bibliometric databases that underpin global university rankings—Clarivate's Web of Science and Elsevier's Scopus—are overwhelmingly English-language in their journal coverage. While both have expanded the number of non-English journals they index in recent years, the vast majority of indexed publications remain in English. This structural feature of the data sources means that rankings derived from bibliometric indicators inherently favor institutions and countries where English is the primary language of academic publication.

The implications for non-Anglophone universities are significant. A German university whose scholars publish extensively in German-language journals, a French institution with a strong tradition of French-language social science, or a Chinese university producing important research published in Chinese may see their research output undervalued or entirely invisible to ranking systems. The pressure to publish in English-language journals distorts research agendas, incentivizing scholars to frame their work for international audiences rather than addressing local or regional questions in their native languages.

Disciplinary dimensions of language bias

Language bias does not affect all disciplines equally. In the natural sciences, medicine, and engineering, English has long been the dominant language of international publication, and non-English journals in these fields are relatively rare. The bias is less severe here because scholars in these disciplines are already communicating primarily in English. In the social sciences, law, and especially the humanities, the situation is different. These fields have strong national and regional traditions, and important scholarship regularly appears in French, German, Spanish, Portuguese, Arabic, Chinese, Japanese, Russian, and many other languages.

The humanities face a particularly acute problem. Books, edited volumes, and national-language journals are the primary publication venues in many humanistic disciplines, yet these formats are poorly covered by bibliometric databases. A historian publishing a landmark monograph in Italian, or a philosopher writing in Japanese, may make contributions that are highly regarded within their scholarly community but are invisible to ranking systems. This systematically depresses the ranking positions of universities with strong humanities faculties, especially in non-Anglophone countries. Universities in France, Italy, Spain, and Latin America, which have distinguished humanities traditions, are especially affected.

Reputation survey effects and policy responses

Language bias extends beyond bibliometrics into reputation surveys. Academic respondents to reputation surveys are more likely to recognize and nominate institutions whose work they read, and language barriers reduce the likelihood that scholars in one linguistic community will be familiar with the work of institutions in another. An excellent university in Brazil or Turkey may struggle to build global reputation partly because its scholars' output, however excellent, is published in Portuguese or Turkish and therefore less accessible to the global academic community that dominates survey responses.

Some countries have responded proactively. The Chinese government's investment in English-language journals and international publication has contributed to the rapid rise of Chinese universities in global rankings. South Korea and Japan have similarly incentivized English-language publication. These strategies can improve ranking positions, but they also raise concerns about the marginalization of national-language scholarship and the narrowing of research agendas toward topics that succeed in international journals. Several European countries have developed alternative evaluation systems that value national-language publications and diverse output formats, seeking to balance international visibility with domestic scholarly contributions.

The path forward for more linguistically inclusive rankings

Ranking organizations have acknowledged language bias and taken some corrective steps. Scopus has expanded its coverage of non-English journals, particularly in Spanish, Portuguese, and Chinese. Times Higher Education has explored incorporating book citations and alternative impact measures. Webometrics, by measuring web presence rather than journal publications, offers a partial bypass of the language problem, though it introduces its own biases related to web infrastructure and digital engagement.

The fundamental challenge remains: how to compare research output across languages without imposing a single linguistic standard. Field-normalization can address some disciplinary differences in citation rates, but it cannot account for the fact that identical work may receive different levels of international attention depending on the language of publication. For universities and policymakers, the practical response is to use rankings with awareness of their linguistic limitations, supplement global rankings with regional and national assessments that value local-language scholarship, and support the expansion of multilingual coverage in bibliometric databases. Language diversity in scholarship is an intellectual asset, not a defect to be eliminated for the sake of ranking performance.