The "Xiaobai" and Nobel Prize winners submitted it, and the reviewer's approach was unexpected!

Text ｜ Li Muzi

A study confirms that it will help the publication in the scientific community.

When the only author of a paper was unknown, only 10%of the reviewers suggested to accept this paper, but when the signature of the same papers is the Nobel Prize winner, 59%of the reviewers agreed.

The study recruited hundreds of researchers to review a manuscript. "Unbelievable." Mario Malicki, editor -in -chief of Stanford University, who did not participate in this study, said: "This is the biggest random control test we have seen about bias."

Picture source: getty

Over the years, scientists have been complaining about the "Matthew effect". Matthew (a wealthy fable in the Bible Matthew) is a terminology created by sociologists Robert Merton and Harriet Zuckerman in 1968. Researchers used to describe high status often obtain more proportional to get more. Research results.

However, there are often weaknesses recording this prejudice, such as small samples or lack of randomness. To avoid these issues, a team led by the University of Swtruk University Jürgen Huber sent an email to about 3,300 researchers to ask if they are willing to review the economic research results prepared for the journal.

There are two authors from Chapman University in the United States. They are Vernon Smith, the winner of the Nobel Prize in Economics in 2002. Last year, he had more than 54,000 citations on Google scholars' website; Sabiou Inoua, only 42 citations last year.

The potential reviewers received one of the three descriptions of the paper: first, only named Smith, listed him as a communication author; second, only Inoua; third, no authors.

The team recently reported at the International Press Comment and Scientific Publication Conference held by Chicago that 821 researchers agreed to review (the results of the research also appeared in the pre -print book of the SSRN server last month).

Smith's outstanding performance seems to have affected people's reactions: 38.5%of the researchers who only get his name have been invited by the review; among people who have not received their names, this ratio is 30.7%; only the people of the Inoua people It is 28.5%.

Then, the research team conducted the next study to avoid prejudice in their own research. They focus on 313 voluntary reviewers who did not receive the author's name at first, and randomly allocated to them one of the three manuscripts for review, one of which only listed Smith, and the other only listed only listings. Inoua, the third copy has no authors.

The team also informed the reviewer that their assessment will be part of an experiment that involves multiple invited peers for review, rather than the usual two to three.

Smith's manuscript has won the highest evaluation of the reviewer, and they praise them to include new information and data support conclusions.

24%of the people reviewed the version without the author, and it was recommended to receive the version (directly or slightly modified).

At the same time, Smith and Inoua are also modifying this paper, and they subsequently published it as a pre -printed book in the journal.

This obvious difference may not surprise many people, but one of the authors of the new research is worried. CHRISTIAN KONIG KERSTING, a behavioral economist at the University of Instalbrook, said: "The same work should not be evaluated by the author's different. This makes it difficult for young and unknown researchers to take the first step in the academic process. ","

Researchers who study and publish bias suggest that the double -blind reviews that the authors and reviewers are covered up may reduce the Matthew effect.

But Kerting said that this strategy may not work because the reviewer can usually identify the author from the pre -print or conference report.

not

"China Science News" (2022-09-15 The 2nd Edition of the International Title of the Origin "The Essay Author's Famous Impact Reviewer Review"))

not

Edit | Zhao Lu

Capture | Guo Gang

- END -