Scientific Misconduct and Social Media: Role of Twitter in the Stimulus Triggered Acquisition of Pluripotency Cells Scandal

Background The academic scandal on a study on stimulus‑triggered acquisition of pluripotency (STAP) cells in Japan in 2014 involved suspicions of scientific misconduct by the lead author of the study after the paper had been reviewed on a peer‑review website. This study investigated the discussions on STAP cells on Twitter and content of newspaper articles in an attempt to assess the role of social compared with traditional media in scientific peer review. Objective This study examined Twitter utilization in scientific peer review on STAP cells misconduct. Methods Searches for tweets and newspaper articles containing the term “STAP cells” were carried out through Twitter’s search engine and Nikkei Telecom database, respectively. The search period was from January 1 to July 1, 2014. The nouns appearing in the “top tweets” and newspaper articles were extracted through a morphological analysis, and their frequency of appearance and changes over time were investigated. Results The total numbers of top tweets and newspaper articles containing the term were 134,958 and 1646, respectively. Negative words concerning STAP cells began to appear on Twitter by February 9-15, 2014, or 3 weeks after Obokata presented a paper on STAP cells. The number of negative words in newspaper articles gradually increased beginning in the week of March 12-18, 2014. A total of 1000 tweets were randomly selected, and they were found to contain STAP-related opinions (43.3%, 433/1000), links to news sites and other sources (41.4%, 414/1000), false scientific or medical claims (8.9%, 89/1000), and topics unrelated to STAP (6.4%, 64/1000). Conclusions The discussion on scientific misconduct during the STAP cells scandal took place at an earlier stage on Twitter than in newspapers, a traditional medium.