‘I didn’t know you had a new job at Imperial College,’ reads a message that Swiss epidemiologist Marcel Salathé received from a colleague on December 21. Salathé is surprised: he is not going to work for Imperial College in London at all. How does his colleague get there? The colleague shows Salathé a notification from Google Scholar, a search engine that collects scientific publications and provides an overview. Google Scholar shows a new publication in the Journal of Economics and Economic Education Research in the name of Marcel Salathé, allegedly employed at Imperial College.
The Swiss epidemiologist shares his surprise Twitter, because he did not write this research at all, he also confirms by telephone. “I thought there had been a mistake in the editing of the magazine. Maybe someone from Imperial College wrote this study, and my name got on it by accident.” But after looking at the investigation, the Swiss saw that there was more to it. The article is so bad and nonsensical that it has to be a fake paper. “It’s a lot worse than a computer-generated study could be these days,” Salathé said. The sentence structure is strange, sometimes words are missing, it is ‘almost random chatter’. Google Scholar has since removed the link to the article. It is still available on ProQuest, a similar online science search engine. ProQuest says it will investigate the matter by e-mail if requested.
For scientists, having a reliable reputation is crucial. Also for Salathe. He is a data epidemiologist, and has become a public figure in Switzerland during the corona crisis. In his work he combines computer science, artificial intelligence and epidemiology. He worked on the technology behind digital source and contact research in Apple and Google smartphones and chairs the steering committee of the Swiss Covid-19 research program.
If something like this goes viral, it could damage your reputation
Marcel Salathé victim
Yet Salathé is not afraid of reputational damage from the fake paper. “It’s so fake that everyone can see it’s not mine.” He does think that this kind of misinformation could cause bigger problems in the future. “Not so much for myself, because normally I work on non-political topics. That is now a bit different in the pandemic.”
He is therefore more concerned in the long run about climate scientists, for example. “If something like this goes viral, it can damage your reputation. Once false information has come out, people often pay little attention to corrections.”
The fake article looks very much like a real investigation, not written by Salathé, that was launched in August 2020. Nature popped up, with some words exchanged for synonyms. For example, the original introduction states: “A range of digital data sources are being used to enhance and interpret key epidemiological data gathered by public-health authorities for COVID-19.This was plagiarized in the fake article by:A variety of digital information sources are being employed to boost and interpret key medicine information gathered by public-health authorities for COVID-19.‘Data’ has been replaced by ‘information’, ‘enhance’ for ‘boost’ and ‘epidemiological data’ for ‘medicine information’.
Elisabeth Bik is an expert in the field of scientific fraud and analyzes suspicious articles for indications of plagiarism or image manipulation. In this case, she recognizes a technique that stands out by so-called tortured phrases. A computer takes the text of an existing scientific publication and replaces a number of terms with their synonyms, so that an article does not stand out in automatic plagiarism detections. Yet it often makes a text unreadable, because conventional concepts are replaced by bizarre alternatives. Think of replacing ‘big data’ with ‘enormous information’.
High publication pressure
The journal in which the fake research was published does not have a good reputation. The Journal of Economics and Economic Education Research is affiliated with Allied Business Academies (formerly known as Allied Academies), a collective of academic journals on business and economics. The organization stands Beall’s List, a review of possible predatory magazines or publishers made by Jeffrey Beall, a University of Colorado librarian.
Robbery magazines make money by publishing bad research for a fee. As a scientist, it is standard – even for honest journals – to pay for the open access publication of your research. The article is then available for free, because the researchers themselves have paid for the editorial costs of the journal. Robbery magazines abuse this practice by publishing every research that is offered for a fee. The publication pressure for scientists is particularly high in countries such as China. It’s more about that you publish than where you publish. Robbery magazines do not have peer review – assessment of the research by fellow scientists. As an outsider it is often difficult to see whether a magazine is a predatory magazine or not.
Pure predatory magazines are rare: many questionable magazines are somewhere in between integrity or predatory magazine, explains Bik. Then some of the research that a journal publishes is of crappy quality or plagiarized, but there are also some original studies in between. According to Bik, that also seems to be the case with the Journal of Economics and Economic Education to be the case. It is possible that the honest scientists who publish in these journals do not realize that this is a potentially debatable party.
I don’t see how anyone could benefit from this
Elisabeth Bik fraud-expert
The special thing about the fake publication in Salathé’s name is that there is no scientist who can show off this article on his resume. That’s why it’s guesswork as to a motive. “I don’t see how anyone can benefit from this, except maybe the magazine itself,” says Bik. The journals may want to show off well-known Western names as authors, so that the journal’s status will increase and more scientists will want to publish there for a fee. “It sometimes happens that someone from Russia or China, for example, writes a paper and then adds a Western author as a co-author, while that person knows nothing about it. That is intended to give the paper a little more status,” explains Bik. But in this case, there’s no co-author taking advantage of it, so it can only benefit the status of the journal itself, it seems.
Salathé is not the first to experience this. In June 2021 wrote integrity blog Retraction Watch about Jamie Burr, a Canadian exercise physiologist who went through something similar. Salathé expects even more cases: “There are much bigger names in my field that you would rather use if you want to get a lot of attention. I think there are thousands of such cases. I happened to notice it because Google Scholar picked it up and recommended the algorithm to my colleagues.”