Haunschild, R., & Bornmann, L. (2021). Can tweets be used to detect problems early with scientific papers? A case study of three retracted COVID-19/SARS-CoV-2 papers. Scientometrics, 1-19. https://link.springer.com/article/10.1007/s11192-021-03962-7 

University of Michigan - Ann Arbor

- Researchers studied tweets related to the retraction of 3 COVID-19 publications to see if the tweets indicated any form of warning through a qualitative approach to answer the research questions: Do Twitter users mention problems with papers (early) that are later retracted? Are Twitter users who mention retracted publications informed about the paper status? Do they explicitly mention the issues why the publication has been retracted? 

Chosen retracted publications
- Bae, S., Kim, M. C., Kim, J. Y., Cha, H. H., Lim, J. S., Jung, J., ... & Kim, S. H. (2020). Effectiveness of surgical and cotton masks in blocking SARS–CoV-2: a controlled comparison in 4 patients. Annals of internal medicine, 173(1), W22-W23. 
- Mehra, M. R., Desai, S. S., Ruschitzka, F., & Patel, A. N. (2020). RETRACTED: Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis.
- Wang, X., Xu, W., Hu, G., Xia, S., Sun, Z., Liu, Z., ... & Lu, L. (2020). SARS-CoV-2 infects T lymphocytes through its spike protein-mediated membrane fusion. Cellular and Molecular Immunology, 1. 

Can tweets be used to detect problems early with scientific papers? A case study of three retracted COVID-19/SARS-CoV-2 papers

- Selected and compared the altmetrics retrieved from Altmetrics.com of 3 retracted COVID-19 on August 13, 2020
- Used Twitter API to obtain the tweet IDs of any tweet about the 3 retracted publications
- From August 6 to 9 of 2020, they downloaded 42,746 tweets about the Mehra et al. publication and 67,038 tweets about its retraction; 10,875 tweets about the Bae et al. publication and 3,095 tweets for its retraction; 5,428 tweets about the Wang et al. publication and 53 tweets for its retraction
- To classify accounts that tweet about retractions and retracted publications, researchers used an R code provided by Toupin to study 8 different user profile descriptions – faculty and students; communicators and journalists; professionals; political; personal; institutions and organizations; bots; journals and publishers 

Can tweets be used to detect problems early with scientific papers? A case study of three retracted COVID-19/SARS-CoV-2 papers data collection

Most tweets came from personal users (Bae and Mehra) or the faculty and students category (Wang)

Can tweets be used to detect problems early with scientific papers? A case study of three retracted COVID-19/SARS-CoV-2 papers findings

- Tweets do not discuss all the reasons behind a retraction, before and after retraction
- Must also analyze tweet content or compile a list of keywords that notify users of doubtful methodologies and retracted publications
- Run experiments with more retracted publications while also using non-retracted publications as a control group 

Can tweets be used to detect problems early with scientific papers? A case study of three retracted COVID-19/SARS-CoV-2 papers discussion and further research

Learn Before

Related