Are female scientists underrepresented in self-retractions for honest error?

Area of Science:

• Bibliometrics
• Research Integrity
• Scientific Publishing

Background:

• Retractions are crucial for scientific self-correction and literature quality.
• Self-retraction for honest errors is a recognized good research practice, yet challenging for researchers.
• Traditional research assessment models struggle to reward self-correcting behaviors.

Purpose of the Study:

• To investigate the demographics (country, field, gender) of researchers who self-retract publications for honest errors.
• To specifically examine the representation of female scientists in self-retractions, given data scarcity.
• To contribute to understanding gender dynamics in maintaining research integrity.

Main Methods:

• Collected 3,822 retraction records categorized as 'error' from the Retraction Watch Database (2010-2021).
• Filtered for research articles, excluding publisher/editor-initiated or investigation-related retractions.
• Manually analyzed 2,906 research articles to identify author-requested, unintentional errors, categorizing by country, field, and gender (n=281) using Genderize.

Main Results:

• Female scientists constituted 25% of self-retractions for honest errors.
• The highest proportion of female self-retractions was observed among those affiliated with US institutions.
• Analysis focused on sole corresponding authors to ensure clear attribution.

Conclusions:

• Female scientists are significantly represented in the practice of self-correcting the scientific record for honest errors.
• Findings underscore the need to acknowledge and potentially reward the practice of self-retraction as a component of research integrity.
• Further research is needed to fully understand the factors influencing gender representation in scientific self-correction.