Why are non-Anglo scientists not being named in US news?

News media plays a key role in disseminating research findings to the public. It also adds to scholars’ academic prestige and shapes the public’s perception of who is doing good science. Yet, we know little about how journalists choose to present researchers in their news stories and the potential downstream effects these choices have on the career and leadership representation of marginalised groups in academia.

To begin to analyse how scientists’ demographics are related to science reporting, it is useful to separate media coverage into two aspects: (1) the likelihood of coverage: whose paper gets reported on and what we might call (2) the quality of coverage: how thorough and accurate the reporting is when it is published. For example, media coverage may give much or little space to describing the researchers, and potentially include their quotes.

In our recent paper, we focused on coverage quality, measuring the variation in how journalists choose to describe the researchers behind a paper. Specifically, we assess whether the author is mentioned by name.

Focusing on the subset of papers receiving coverage, we sidestep the question of whose research is covered in the news in the rst place, which may itself be associated with author demographics and other confounding factors, such as a paper’s newsworthiness and the author’s self-promotion on social media.

Perceived ethnicity

We focused on the “ethnicity” of authors rather than their nationality for two reasons: (1) an author’s nationality may be very uid, especially in the US; (2) journalists generally see only author names’ and affiliations on reading the paper, and the names often reliably signal race and ethnicity. We decided to use ethnicity because it contains richer information than race.

To match the information available to journalists, we based our study on the perceived ethnicity inferred from names. This choice entails substantial trade-offs. On the one hand, self-described identities may occasionally differ from perceived ones, and some authors may identify with multiple ethnicities.

Nevertheless, in most of these mismatching cases, journalists will not know how authors self-identify themselves and instead infer identity from names. In these cases, using authors’ self-identities would be problematic, as it would misrepresent the actual perceptions journalists form and possibly use when they write their stories.

Our corpus of news-reported papers was sourced from Altmetric.com. It consists of 223,587 news stories from 288 US-based outlets reporting on 100,486 scientic papers. We additionally obtained papers’ and authors’ metadata from the Microsoft Academic Graph and the Web of Science databases.

For each paper, we focused on authors at the highest “risk” of being mentioned (the rst author, last author and all other authors designated as the corresponding author) and treated each (story, paper, author) triplet as an observation in a logistic regression that predicts author mention with author demographics and controls. We developed a computational method to detect three types of attributions, including name mentions, quotations and institution mentions.

We found substantial disparities in author mentions across name-inferred ethnicities. These disparities are robust to important factors related to the paper, story and author such as research topics, journal impact, author prestige, corresponding author status, afliation location, name complexity, article length, etc.

Specically, we found that most authors with minority-ethnicity names are signicantly less likely to be mentioned than those with Anglo names. Of these minority groups, authors with European names were disadvantaged the least, while East Asian and African names disadvantaged the most. There is an up to six percentage points decrease in the mention probability for East Asian and African names, which equals to a 15% decrease relative to Anglo-named authors in media representation.

Media bias?

Second, the disparities are similarly large for authors with African and East Asian names who are afliated with US institutions. In particular, the US-based Chinese, non-Chinese East Asian and African-named authors experience a 4.8, 3.8, and 4.6 percentage point drop in mention rate compared to their Anglo-named US counterparts. This suggests that being affiliated with US institutions in the same geographic region does not eliminate the observed disparities.

These findings could be due to media bias. For example, journalists may consider certain names as less authoritative than Anglo names. However, our result so far does not directly imply media bias because all US-based authors may still differ in other factors, such as the perceived or actual English proficiency that could impact journalists’ decision on reaching out to them.

To directly test rhetorical bias on the part of the media, we examined “institution-substitution” where the author is mentioned by their institution but not by name, for example, being named as “researchers at the University of Michigan”. We find that journalists are more likely to replace African and East Asian named authors with their institutions. Among US-based authors, this mention type should not depend on pragmatic factors such as English uency. Thus, this substitution effect likely reveals that journalists place less rhetorical value on authors with minority names.

Types of outlets

Third, we also find consistent disparities across three types of outlets, including press releases, general news and stories published in outlets focusing on science and technology.

The disparity in press releases outlets is particularly notable, as stories in these outlets typically reuse content from university press releases, suggesting that universities’ press ofces themselves, while having less disparity than other outlet types, still prefer to mention scholars with Anglo names.

This result is unexpected because local press ofces are expected to have greater direct familiarity with their researchers, reduce the misuse of stereotypes and to be more responsible for representing minority researchers equitably.

The largest disparities are seen in general news outlets, for example, The New York Times and The Washington Post, where again scholars with Chinese- and African-associated names have a 6.0-8.0 percentage point drop in mention rates.

This signicant drop reduces nearly one third of the deserved media representation of a large community of scientists (general news outlets mention authors with a 24.2% chance on average).

An alarming result

As general news outlets often have well trained editorial staff and science journalists dedicated to accurately reporting science and tend to publish longer stories that have room to mention and engage with authors, this result is alarming. Historically, these ethnic minorities have been stereotyped and underrepresented in US media and leadership roles, which has continued in objective science reporting across all outlet types. The mechanisms of this variation deserve further investigation.

Our work shows that science journalism is rife with disparities in terms of which authors receive name attributions when their research papers are reported in US news. Mention rates are especially low for East Asian and African names, less pronounced for European names and even less pronounced for Indian and Middle Eastern names.

As science continues to globalise and is increasingly produced by authors from non-Western ountries, the way English-language media responds to non-Anglo-named scholars will only grow in importance.

While our study only focuses on the “second stage” of science media coverage and its quality, it is likely that such ethnic disparities are even larger in the rst stage of coverage, where media outlets choose whose papers to report on in the rst place.

Our work thus suggests that disparities in science media are likely to compound across different aspects of coverage, yielding differences in outcomes much larger than those shown by studies of any one stage.

Hao Peng is a postdoc at the Kellogg School of Management of Northwestern University. He is affiliated with the Northwestern Institute on Complex Systems. He studies human-centred topics at the intersection of computational social science, science of innovation, and social networks. His research aims to generate novel insights that can help organisations leverage the full potential of human capital to accelerate discoveries and breakthroughs. Misha Teplitskiy is an assistant professor at the University of Michigan School of Information. He is a sociologist of innovation, studying how institutions and technology can help accelerate scientific discovery. His recent work focuses on understanding and mitigating biases in research assessment. David Jurgens is an associate professor at the University of Michigan jointly in the School of Information, and the Department of Computer Science and Engineering. His research develops methods in natural language processing and computational social science to study human behaviour and language.

This post draws on the authors’ article, Author mentions in science news reveal widespread disparities across name-inferred ethnicities, published in Quantitative Science Studies. It was first published on the LSE’s Impact of Social Science blog here. It gives the views and opinions of the authors and does not reflect the views and opinions of the Impact of Social Science blog (the blog), nor of the London School of Economics and Political Science.