Organized scientific fraud increases at an alarming rate, the study reveals

From research made to pateries and paid quotes, organized scientific fraud is increasing, according to a new study by the Northwestern University.

By combining the analysis of large -scale data from scientific literature with case studies, researchers have conducted an in -depth investigation into scientific fraud. Although the concerns related to scientific misconduct is generally focused on solitary individuals, the Northwestern Study has rather revealed global sophisticated networks of individuals and entities, who systematically work together to undermine the integrity of academic publishing.

The problem is so widespread that the publication of fraudulent science exceeds the growth rate of legitimate scientific publications. The authors argue that these results should serve as awakening to the scientific community, which must act before the public loses confidence in the scientific process.

The study, “the entities allowing large -scale scientific fraud are important, resilient and growing”, was published in the Proceedings of the National Academy of Sciences.

“Science must better do the police to preserve its integrity,” said Luís an A lookums de Northwestern, the main author of the study.

“If we are not aware of this problem, worse and worse behavior will normalize.

Expert in complex social systems, Amaral is Professor Erastus Otis Haven and Professor of Engineering and Mathematics Applied to McCormick School of Engineering in Northwestern. Reese Richardson, postdoctoral scholarship holder at the Amaral laboratory, is the first author of the newspaper.

In -depth analysis

When people think of scientific fraud, they can remember the reports of withdrawal articles, falsified data or plagiarism. These reports generally concentrate around the isolated actions of an individual, who takes shortened to move forward in an increasingly competitive industry. But Amaral and his team discovered a widespread underground network operating in the shadows and outside the consciousness of the public.

“These networks are essentially criminal organizations, acting together to simulate the science process,” said Amaral. “Millions of dollars are involved in these processes.”

To conduct the study, the researchers have analyzed large sets of data from withdrawn publications, editorial files and image duplication instances.

Most data come from the main aggregators of scientific literature, including Web of Science (WOS), Elsevier SCOPUS, Pubmed / Medline from the National Library of Medicine, and Openalelex, which includes data from the Microsoft academic graphic, Crossref, Orcid, Unpaywall and other institutional standards.

Richardson and his colleagues also collected lists of indexed journals, which are learned journals that have been deleted from databases so as not to meet certain quality or ethical standards.

Researchers have also included data on retracted Watch retraction articles, comments from the article by Pubpeer and Metadata, such as publisher names, submission dates and acceptance dates – articles published in specific journals.

Buy a reputation

After analyzing the data, the team discovered coordinated efforts involving “paper mills”, brokers and infiltrated journals. Operating, like factories, paper mills produce a large number of manuscripts, which they then sell to academics who wish to quickly publish new works.

These manuscripts are mostly of low quality: data manufactured, handled or even stolen images, plagiarized content and sometimes absurd or physically impossible affirmations.

“More and more scientists are taken in paper mills,” said Amaral. “Not only can they buy papers, but they can buy quotes. Then they can appear as scientists well reproduced when they have barely carried out their own research.”

“The paper mills work by a variety of different models,” added Richardson.

“So, we have just scratched the surface of their operation. But they mainly sell everything that can be used to whiten a reputation. They often sell paternity slots for hundreds or even thousands of dollars. A person could pay more money for the first post of author or less money for a fourth survey sharing position.”

To identify more articles from paperwork, the Amaral Group has launched a parallel project that automatically scans published public science and engineering documents. The team specifically looked for authors who have poorly identified the instruments they used in their research. An article with these results was accepted by the newspaper Plos a.

Brokers, diversion and collusion

Amaral, Richardson and their collaborators found that fraudulent networks use several key strategies:

1. Groups of researchers end to publish articles in several journals. When their activities are discovered, the papers are then retracted;
2. Brokers serve as intermediaries to allow mass publication of fraudulent articles in compromised journals;
3. Fraudulent activities are concentrated in specific and vulnerable sub-champs;
4. The organized entities escape quality control measures, such as the newspaper’s deindexation.

“The brokers connect all the different people behind the scenes,” said Amaral. “You have to find someone to write the document. You have to find people ready to pay to be the authors. You have to find a newspaper where you can have everything published. And you need editors in this review that will accept this document.”

Sometimes these organizations completely bypass established magazines, seeking rather that missing journals turn away. When a legitimate newspaper ceases to publish, for example, bad actors can resume its name or website. These actors surreptitiously assume the identity of the newspaper, lending credibility to its fraudulent publications, despite the deceased of real publication.

“It happened to the journal HIV Nursing,” said Richardson. “It was previously the newspaper of a professional nursing organization in the United Kingdom, then he stopped publishing, and his online field expired. An organization bought the domain name and began to publish thousands of articles on subjects unrelated to nursing, all indexed in Scopus.”

Fight for science

To fight against this growing threat to legitimate scientific publishing, Amaral and Richardson underline the need for a multiple approach. This approach includes an in -depth examination of editorial processes, improved methods to detect manufactured research, a better understanding of the networks facilitating this fault and a radical restructuring of the science incentive system.

Amaral and Richardson also underline the importance of solving these problems before artificial intelligence (IA) infiltrates scientific literature more than it has already done.

“If we are not ready to face the fraud that already occurs, then we are certainly not ready to manage what generative AI can do to scientific literature,” said Richardson.

“We have no idea what will end in the literature, what will be considered a scientific fact and what will be used to train future AI models, which will then be used to write more articles.”

“This study is probably the most depressing project with which I was involved in my whole life,” said Amaral. “Since I was a child, I was enthusiastic about science. It is painful to see others engage in fraud and by deceiving others. But if you believe that science is useful and important for humanity, then you have to fight for that.”