Data for: Institutional repositories measurably increase the FAIRness of research outputs: Actionable data for librarians

These data and code support work presented at the ER&amp;L 2023 conference on March 6, 2023 in Austin, TX, USA.<b>Presentation description:</b><br>Datasets and non-traditional research outputs are a growing use case for institutional repositories. We quantify how much FAIRer institutional repository outputs are and examine other benefits like reuse metrics and Altmetrics. Librarians can use this to quantify the value of their role and their institution's investment in the repository.<b>Methods and file description:</b><br>Metadata was harvested through Figshare's API (https://docs.figshare.com). Records publicly shared on figshare.com by researchers affiliated with a US based institution were found through figshare.com profiles registered using an email using an institution domain. This information is not public and the affiliation information was removed from the ANALYSIS-DATASET. All other information is publicly available through the API. The compressed file called 'additional-datasets.zip' contains the author, categories, funder, and file information for the researcher records and institution records. The Jupyter Notebook harvests metadata, creates datasets, produces figures used in the presentation, and creates the dataset used for the negative binomial analysis in R. The data-for-R.R file performs the negative binomial analysis which shows relative differences in several metrics for the groups of records.Altmetric data presented at the conference are not public and not included in this dataset.