For PLOS, increasing data-sharing rates—and especially increasing the amount of data shared in a repository—is a high priority. Research data is a…
Written by Lauren Cadwallader, Lindsay Morton, and Iain Hrynaszkiewicz
The latest Open Science Indicators (OSI) dataset has arrived! And with it, a brand new indicator. Read on for details, and be sure to explore the data for yourself.
About Open Science Indicators
Open Science Indicators identify and quantify instances of specific Open Science practices in published scientific literature using a combination of Natural Language Processing (NLP) and Artificial Intelligence (AI) techniques applied and developed in partnership with DataSeer. At PLOS, we use indicators as a tool to support our mission to advance Open Science. Other members of the scholarly community use them to understand trends in their institution; to explore markers of research integrity, and to benchmark new tools for monitoring Open Science.
The latest Open Science Indicators dataset
This is the fourth quarterly update to Open Science Indicators since we introduced the dataset in December 2022. It covers January 1 2019 to June 30 2023 (four full years, plus the first half of 2023). Like the previous installments, this update measures data-generation and -sharing, code-generation and -sharing, and preprint posting. It also includes a preliminary version of our first new indicator: protocol sharing. Watch this space for more on how we developed the protocol sharing indicator in a future post.
OSI Version 4 (Sept 23) also includes an expanded comparator corpus, which has doubled the number of articles from other publishers included in the dataset from 8,176 in version 3 (June 23) to 16,023 today.
See the README for a full list of updates related to version 4.
For the first half of 2023, all four indicators remain stable overall, consistent with patterns observed in 2022. In keeping with previous results, PLOS articles were more likely to link to public research data or code than non-PLOS articles, and equally likely to link to a preprint.