News Release

Would you share your scientific results before publication?

Prepublication disclosure of scientific results: Norms, competition, and commercial orientation

Peer-Reviewed Publication

American Association for the Advancement of Science (AAAS)

Scientists who surveyed more than 7,000 active faculty researchers in the U.S. and abroad report that more than half of them had disclosed their results before publication, largely to receive feedback. With regards to other motivations for pre-publication disclosure, as well as trade-offs, there is great variation by field, say the authors - with social scientists and mathematicians, for example, being the most likely to disclose.

The results of this study offer novel insights into why, and at what stage, researchers share their work before they publish, as well as how norms, competition and commercial orientation (factors the authors collectively call the "NCC variables") influence their decision.

To date, whether or not scientists should share information with colleagues before publication is widely debated, especially given the possibility of being scooped. To illuminate factors that affect scientists' willingness to do so, Jerry G. Thursby and colleagues analyzed publication data and survey responses from 7,103 active faculty researchers from the U.S., Germany and Switzerland across nine scientific fields. They explored whether these individuals had indeed disclosed before publication, and if so, the motivation and stage, as well as information on how often respondents presented unpublished results to general audiences versus withholding such information when presenting.

Overall, 67.2% of respondents reported disclosing before publication and the dominant motive was to receive feedback; engineers and computer scientists were less likely to disclose. More formulaic fields, such as computer science and mathematics, often present their research before publication to attract competitors to work separately on a problem in their field, but not their specific problem, the authors note. Disclosing behavior generally relates to differences in the 12 NCC variables, and differences in respondents' beliefs account for 70% of the variation across fields, Thursby et al. say.

###


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.