{{ site.title }}

Realities of Academic Data Sharing (RADS) Research Purpose

Background Information
Research Streams

Institutional Infrastructure Survey & Campus Administrator Interviews
Funded Researcher Survey & Interviews
Inclusion/Exclusion Criteria for Study Participants
Data Collection Instruments
Informed Consent
Confidentiality, PII, and Data Deidentification

Institutional Review
Plans to Make Results Available
Contact Information

Background Information

This study, funded by the National Science Foundation (NSF #2135874), is an exploratory examination of the costs, activities, services, and infrastructure of making research data publicly available from funded research across six academic institutions.

Federally mandated policies requiring public access to funded research data have impacted, and continue to impact, how institutions and researchers support the entire research data life cycle. The 2013 White House Office of Science and Technology Policy (OSTP) Holdren Memo “Increasing Access to the Results of Federally Funded Scientific Research,” as well as others, transformed how institutions and researchers manage their research data. This trend is only increasing, as the 2022 OSTP Nelson Memo, “Ensuring Free, Immediate, and Equitable Access to Federally Funded Research,” requires all federal agencies with research and development expenditures to update and implement their public access policies, no later than 2026, to make publications and their supporting data publicly accessible without an embargo period.

In response to these growing federal requirements to share publicly funded research data, many academic institutions have developed and launched a variety of support services to reduce the faculty burden in meeting these requirements. These services are often spread across the institution and housed in various administrative units, such as campus IT, the university libraries, and the research office, among others. Similarly, a multitude of platforms exist for sharing research data. Researchers store their data in a variety of platforms including institutional repositories, generalist repositories, discipline-specific repositories, or in other locations (such as personal websites) to comply with these mandates. The extent of where funded research data are shared, as well as the costs to support this sharing, are not fully understood.

RADS considers these questions and examines where research data from federally funded projects are shared, and the costs to support this sharing, since the issue of the 2013 Holdren Memo up until 2022. The six institutions participating in this study are Cornell University, Duke University, University of Michigan, University of Minnesota, Washington University in St. Louis, and Virginia Tech, all of which are members of the Data Curation Network (DCN). The Association of Research Libraries (ARL) is the awarded organization, but RADS is a joint project between ARL and the DCN.

By examining researcher and institutional costs, activities, services, and infrastructure, the goal of the RADS study is to provide an exploratory, yet comprehensive, analysis of the true costs of  data sharing of funded research. Results of our study will inform funders, institutions, and other stakeholders.

Research Streams

Two investigative research streams occurred at each of the six institutions in 2022 and 2023.

Institutional Infrastructure Survey & Campus Administrator Interviews

Campus administrators were surveyed in 2022 and asked to determine what costs, activities, services, or infrastructure they are developing, or have developed, within their office, department or unit, to enable research data sharing from 2013 to 2022.

Follow-up administrator interviews were conducted in early 2023, with the purpose to clarify survey responses and gather additional detail around research  data management activities and expenditures. Although RADS did not originally intend to take-up questions around the new NIH Policy for Data Management and Sharing specifically, the interviews were timely, as many units in the participating institutions had made recent investments to services and infrastructure to implement this new policy. Thus, these interviews served the dual-purpose role of eliciting responses for activities and expenses retrospectively, as well as providing an opportunity to inquire into future investments due to policies resulting from the 2022 OSTP Nelson Memo.

Funded Researcher Survey & Interviews

Funded researchers were surveyed in 2022 and asked to identify the activities and infrastructure used to share their research data, and their subsequent costs, of specific grants identified from the Department of Energy (DOE), the National Institutes of Health (NIH), and the National Science Foundation (NSF) award databases. 

The survey asked these project PIs if they would be interested in a potential follow-up group interview, which could potentially include members of their project research team. During these interviews, which were conducted in early 2023, clarifying details from survey responses were solicited, specifically around labor (staffing), service, and infrastructure costs, as well as other decisions related to where and how data were shared. NIH funded researchers were asked specifically if and how they were preparing for the new NIH Policy for Data Management and Sharing.

Inclusion/Exclusion Criteria for Study Participants

To identify possible participants of the institutional infrastructure survey, RADS PIs each did a scan of their institution to identify departments and units/offices that support funded researchers with data sharing. After this scan, administrators of these departments and units/offices were identified, and then invited to participate in the survey. Administrators were asked to report on activities and expenditures for their unit only (when administrators from multiple units under one department participated). In addition to this identification, other criteria for survey participation included:

  • Knowledge of department/unit infrastructure expenditures
  • Knowledge of personnel activities to support data sharing
  • Knowledge of personnel salaries

Criteria for the inclusion of administrator interview participants included:

  • An affirmative response from the administrator in the 2022 survey that they were interested in a follow-up group interview.
  • Broad knowledge of past, and potential future, investments into infrastructure and personnel support of data sharing.

Note: Specific interview participants were also identified based on their service area (e.g., research office, IT support), ensuring a range of service providers representation across all six RADS institutions.

Participants (project PIs) for the researcher survey were identified using several inclusion and exclusion criteria:

  • Researchers must have been externally funded and their project listed in one of three funder award databases (DOE, NIH, and NSF).
  • Researchers must have been currently employed at the same institution as when the award was granted.
  • Awarded projects must have been considered one of the five disciplinary areas of consideration in this study: environmental science, materials science, psychology, biomedical sciences, physics, or a cross-disciplinary study including one of these six areas.
  • Projects must have been completed between 2013 to present, excluding no-cost grant extensions. Note: The year 2013 was selected as the starting point for analysis due to the release of the OSTP issued Holdren Memo.

Criteria for the inclusion of researcher interview participants included:

  • An affirmative response from the project PI in the 2022 survey that they were interested in a follow-up interview, possibly with their research team
  • When applicable, names and contact information of the research team members provided by the project PI
  • Consent of the identified participants

Note: Of the researcher survey responses indicating interest in a follow-up interview, the RADS team identified participants who represented projects in each of the five discipline areas.

Data Collection Instruments

Informed Consent

Informed consent was required for survey and interview participation. Informed consent for the surveys were sought in “click to consent” format in Alchemer, the survey platform, immediately preceding the start of the survey. Verbal consent was required immediately preceding interviews. If data is reused, additional consent will not be requested from participants.

Cornell University
Duke University
University of Michigan
University of Minnesota
Virginia Tech
Washington University in St. Louis

Confidentiality, PII, and Data Deidentification

Selection of participants, their consent to participate in this study, and their responses will be de-identified and remain confidential when reports of this study are released publicly. However, potential identifiers, including institution names, discipline areas, or office or department descriptors will be included in data analysis. Raw data will not be deidentified, as responses will be attached to participant identifiers, including name, title, department, and email, in both surveys.

For all data collection methods, a coding scheme will be implemented to replace the direct identifiers with numeric codes. Personally identifiable information (PII) collected in the surveys and interviews will be recoded using numeric identifiers and researchers of this study will maintain a linking document with code descriptors. The final analysis dataset will not have direct identifiers (names or emails), but will include office or department and institution. PII collected in this survey will be maintained until grant end and final reports are released.

Raw identifiable data will be stored with ARL on secure servers and in Alchemer during the course of the project. Assets with identifiable information will be destroyed at the end of the project. The digital assets from this project, excluding raw identifiable data, will be retained and curated for a minimum of 10 years.

Institutional Review

ARL does not have its own institutional review process and, due to this, each institution in the RADS study submitted an application to their Institutional Review Board (IRB) separately. Each institutional IRB issued an exempt status for the project on the following dates:

  • Cornell University: June 30, 2022
    Note: The Cornell IRB did not issue an exempt status, as it determined this study did not qualify as human subjects research; no modifications for the study were recommended.
  • Duke University: July 27, 2022; amended protocol approval September 23, 2022
  • University of Michigan: June 10, 2022
  • University of Minnesota: June 7, 2022
  • Virginia Tech: June 7, 2022
  • Washington University in St. Louis: June 30, 2022

Plans to Make Results Available

Research data from this award, NSF #2135874, are under obligation to be shared and made publicly available.

Deidentified research data will be made available for reuse through each institution’s institutional repository and through ARL’s website. Please contact each institution’s co-PI (see contact information below) for more information on how data will be stored locally.

Results from this study have been, and will continue to be, broadly disseminated through conference presentations and materials, scholarly articles, organizational blog posts, and as part of publicly available reports published on ARL’s website and in institutional repositories.

A list of up-to-date research outputs are listed on the main RADS web page. All assets resulting from this award, NSF #2135874,is or will be licensed as Creative Commons-Attribution (CC-BY 4.0), to allow for the broad sharing and adaptation of the materials.

Contact Information

Association of Research Libraries

  • Cynthia Hudson Vitale, Director, Science Policy and Scholarship, project PI
    Email: cvitale@arl.org

Cornell University

  • Wendy Kozlowski, Library Lead for Research Data Services, project co-PI
    Email: wak57@cornell.edu

Duke University

  • Joel Herndon, Director of the Center for Data and Visualization Sciences, project co-PI
    Email: joel.herndon@duke.edu

University of Michigan

  • Jake Carlson, formerly Director of Deep Blue Repository and Research Data Services, University of Michigan; currently Associate University Librarian for Research, Collections and Outreach, University of Buffalo, project co-PI
    Email: jakecarl@buffalo.edu

University of Minnesota

  • Alicia Hofelich Mohr, Research Support Coordinator, Liberal Arts Technologies and Innovation Services (LATIS), project co-PI
    Email: hofelich@umn.edu 

Virginia Tech

  • Jonathan Petters, Assistant Director, Data Management & Curation Services, Data Services, project co-PI
    Email: jpetters@vt.edu

Washington University in St. Louis

Affiliates