A third of scientific papers may be fraudulent

A recent paper has indicated that approximately one-third of studies published in neuroscience journals, as well as around 24% in medical journals, have been found to be either fabricated or plagiarized.

This study, which is currently in preprint and has not undergone peer review, examined a sample of 5,000 published papers, as reported by Science, Semafor reported.

The researchers employed a straightforward automated detection system to identify two key indicators.

Firstly, they examined whether authors were registered with personal email addresses instead of institutional ones.

Secondly, they analyzed whether authors claimed affiliation with a hospital. Papers identified as potentially fraudulent based on these criteria were subsequently scrutinized by human reviewers.

The researchers estimated that approximately 1,500 of the examined papers were likely fraudulent.

Please note that as this study is currently a preprint, it has not undergone the rigorous evaluation process of peer review, which ensures the credibility and accuracy of scientific research.

Written by staff