An Analysis of Datasets within Illinois Digital Environment for Access to Learning and Scholarship (IDEALS), the University of Illinois Urbana-Champaign Repository

Abstract

Objectives: The objective of this study is to identify: (1) how many datasets are within Illinois Digital Environment for Access to Learning and Scholarship (IDEALS); (2) which types of files are deposited in the repository; (3) which research methodologies are associated with these datasets; and (4) which research discipline or research communities are associated with these datasets within IDEALS. Methods: Datasets collected in this study were found using the University of Illinois repository IDEALS website link https://www.ideals.illinois.edu. The keywords used were data or dataset. In order to facilitate analysis, datasets were analyzed using MS-Excel spreadsheets. They were coded by title, issue date, research methodology, research discipline, and community to explore patterns of use and the relationship to data management and research data services. Results: There are 507 datasets in IDEALS dating from 1905-2015. Text files are the most frequently deposited file type; bibliographies represent 34% of the datasets; and, farming inventory lists are 26% of the datasets. Various research disciplines represent 18% of the datasets and research communities are associated with 78% of the datasets. 7% of the datasets are sponsored by NSF, NIH, IMLS and DOE funding agencies. Conclusion: Understanding the file types, research methodologies, research disciplines and research communities within a university's current infrastructure, will provide a representation of the datasets and research supported within the university repository. It will enhance academic librarians and repository managers' data management conversations with researchers and provide information needed to needed to improve workflow deposit and batch loading. It will enhance research data services, meet researcher's needs, assess short-term preservation, and determine long-term preservation needs.

What’s in the Box? Assessing the potential usability of four decades of thesis and dissertation supplementary files

“…While these are common data types in this and other studies (e.g. Van Tuyl and Michalek 2015, Whitmire et al 2015, Wiley 2015, Parham et al 2012 it is apparent that research data curators need to consider creating best practices for documenting and making actionable non spreadsheet/ tabular data. Certainly, the resources provided by national and international organizations (e.g.…”

“…While some researchers have raised concerns about the quality of data (Merson et al 2016) and its retrievability (Vines et al 2014), relatively little has been written to evaluate data quality in data repositories or even what types of data are being deposited. Some notable examples of projects to evaluate quality data sharing have focused primarily on retrievability of the data (Savage and Vickers 2009) or repository contents (Wiley 2015), but relatively little has been written about the quality of the shared data. A notable exception to the dearth of investigation in this area is the work of Naudet et al (2018), who examined the retrievability of datasets and the reproducibility of the results for two journals with strong data sharing policies, finding that only approximately45% of authors were able to produce their datasets with sufficient information to reproduce their results.…”