National Network of Libraries of Medicine
English Arabic Chinese (Simplified) French Hindi Japanese Korean Persian Portuguese Russian Spanish

Research Data Management FAQs

 
Working with Researchers

What reasons do researchers give when they don't want to share datasets, and how can I respond to these barriers?

Here are some reasons why researchers don’t want to share their datasets:

  •     Misuse or abuse of data
  •     Cost of preparing data to be shared
  •     No utility or value beyond current use
  •     Lack of knowledge, technological ability, or time
  •     Re-analysis that disputes findings based on previous work
  •     Relinquishing publication opportunities based on future data mining
  •     Ethical concerns
  •     Intellectual property or commercialization

Here are some ways you can respond to these concerns:

  •     Funders require sharing
  •     Open exchange is becoming more of a norm/critical for collaborative research
  •     Gives an opportunity to be first to publish
  •     Get a DOI and make it citable

Within engineering and chemistry (e.g., pharmaceuticals) in particular, commercial and proprietary considerations for the data often limit what researchers are willing to share.

Researchers have a strong sense of ownership over their data and don’t like the idea of others using it for their own purposes and getting publications that could have been “theirs.” Tell them that it’s acceptable to keep it private or by request until they are done publishing on it, but that getting it ready to share early will save a lot of headaches.
 
Many researchers acknowledge the amount of work and lack of incentive for preparing data to share well. Try to encourage them to integrate curation into their processes which will both improve data quality and make sharing easier when they are compelled to.

Sources

Rinehart, A. (2018, August 3). NNLM Research Data Management Series: Approaching Resistance to Change in Research Data Management [Video File]. Retrieved from https://www.youtube.com/watch?v=Z_OQ3sQFPMQ&list=PLUlRqrjIldD4igiacv1E0Y...
 

How should librarians approach speaking with researchers about their data management practices?

There are several options for speaking with researchers about their data management practices - build relationships with researchers/ departments and offer to help with a data management plans:

  • Use  online resources such as https://dmptool.org/promote
  • Use the information and resources in a libguide or webpage
  • Have a class at the department or library with food if possible
  • Have researchers request a one-on-one consultation with a data management specialist
  • Offer DMP draft review sessions or have researchers send you their data management plan draft for review.

It’s important to acknowledge the skills and time involved and be empathetic about the fact that these processes and requirements are new and something that researchers aren’t typically taught during their training. Centering the library as a resource for training and making compliance with new mandates easier is effective.

Sources

Data Management Plan Tool
University of Oregon Libraries Data Management Plan
 

Are there core data management needs regardless of discipline/subject area?

Here is a summary document about data management needs:
http://learn-rdm.eu/wp-content/uploads/red_LEARN_Elements_of_the_Content...

Sources

http://learn-rdm.eu/wp-content/uploads/red_LEARN_Elements_of_the_Content...
 

What are effective methods for advocating that librarians have the capabilities and knowledge to support research data management needs?

All the same services and tasks that librarians have been providing their users with for books, articles, audiovisual materials, and so forth, are applicable to data.

Sources

Henderson, Margaret E. Data Management: A Practical Guide for Librarians. Rowman & Littlefield Pub Inc, 2017.
 

What are the tangible benefits to research communities when involving librarians in research data management projects?

Here is a summary of the benefits of data management planning:

  • Data can be found and understood when it is needed
  • Data or notes are less likely to be missing when it comes time to process and analyze results
  • All project staff are aware of what they need to do when doing research and collecting data
  • There is continuity if project staff leave or new researchers join
  • Unnecessary duplication (e.g., recollecting or reworking data) is avoided
  • Permissions and ownership are understood so there should be no impediments to publication
  • Data underlying publications are maintained, allowing for validation of results
  • Data can be found if it is needed for sharing
  • Data is available if needed for Freedom of Information Act (FOIA) requests or to resolve intellectual property issues
Sources

(Based on Jones, 2011; Krier and Strasser, 2014; Briney, 2015)
 

What are common research data management IRB requirements researchers must include when submitting for approval?

The Institutional Review Board (IRB) is the body which assesses research carried out on human subjects, so it is less interested in RDM than the content of the research. The better question is, what are RDM requirements when submitting applications to the university’s body that assesses grant applications for compliance with regulations and mandates, before they are submitted? This would be the university’s (or organization’s!) Office of Sponsored Programs, Office of Research Programs, or similar. Answering that questions, some common requirements might be:

  • A data management plan (required by some funders). Researchers may want to use DMPTool or online templates (discipline-specific is good) for this
  • Information about data storage and security (to be sure that privacy/confidentiality concerns are met, especially involving human subjects, and related to standards/regulations such as HIPAA if health information is involved)
  • Information about retention of data (many universities have records retention policies, which may or may not apply to research data); this issue also relates to whether a data repository will eventually be used for data storage, and if so which one (institutional or otherwise)
  • Information about plans to share data (also required by some funders)—placement in an appropriate repository may address this question
  • Information about data ownership, especially if there is proprietary information with commercial potential
  • Information about data portability (which members of the research team will retain which rights if they leave the university?)
  • Information about any budgetary issues related to the data specifically
  • In some cases, information about intent to publish so that access to research products is open (this can have budgetary implications if a researcher has to pay a publisher for open access status)
Sources

Columbia Data Management Plan Templates

NIH Data Sharing Plan information

Data Society Supporting Ethics Data Research
 

Where can I learn more about metadata standards?

Metadata is simply defined as “data about data” and is the information we create, store, and share to describe things. The National Information Standards Organization’s Understanding Metadata (https://groups.niso.org/apps/group_public/download.php/17446/Understandi...) is an excellent introductory primer. In the context of research data management, metadata is often referred to as “data documentation.” The metadata provides the details about what, where, when, why, and how the data were collected, processed, and analyzed, in order to help you document your processes and to help other researchers find, use, and cite your data.
 
Metadata can be conveyed in simple text files (“readme” files) or can be structured in a standardized, machine-readable form, often referred to as a “schema.” Some examples of metadata standards are Dublin Core, DataCite, DDI (Data Documentation Initiative), EML (Ecological Metadata Language), and MIBBI (Minimum Information for Biological and Biomedical Investigations). Many data repositories and archives recommend or require that your metadata follow a specific standard. To assist in finding metadata standards, the Digital Curation Centre maintains a directory of metadata standards (http://www.dcc.ac.uk/resources/metadata-standards/list) organized by subject discipline, including tools to implement the standards and repository use cases. There is also a community-maintained version of this directory (http://rd-alliance.github.io/metadata-directory/) supported by the Research Data Alliance. Another great resource for locating relevant data standards, databases, and policies in the life sciences, biomedical sciences, and environmental sciences is FAIRsharing.org.

Sources

Understanding Metadata

Metadata Standards

Metadata Directory

FAIRsharing
 

What are acceptable locations for storage and preservation of datasets?

Your data storage strategy should include planning for both active data during your research project, and archival data after your research project ends. Consult the information technology (IT) department at your institution to determine what resources are available with regard to storage, backup, and security for your data. Security is especially important if you are storing confidential, sensitive, or personal data.
 
For active data, data storage options include personal computers (PCs) and laptops, external storage devices, flash drives and CD/DVD, networked drives, and cloud storage. PCs and laptops are convenient, but they are not recommended to store master copies of data or for long term storage; it is crucial to back files up regularly. External hard drives are a good option for backup of data and also provide security, both through encryption and the simple detachment of the drive. Flash drives are a security risk and should not be used to store master copies of data or for long term storage.  Networked drives managed by your IT department are backed up regularly, replicated, and secure, so are highly recommended for storage of master copies of data. Cloud storage options such as Dropbox, Box, Google, or Amazon can be considered for master copies of data but are not recommended for in-process data or large files. There may be size, cost, or privacy limitations that could pose a risk to your data, so it is important to read and understand the terms of service. Any storage strategy should include a backup plan and involve “the rule of 3”: store at least two copies of your original data locally and keep one off site.
 
Your data storage strategy should take into account how and where to store data for archiving and future use after your research project ends.  Considerations include: departmental, institutional, publisher, or funder policies on data retention; how you will adhere to the policies; and, requirements for public data sharing and re-use. The most effective way to share data publicly and preserve it is to deposit it to an established data repository, such as a discipline-specific repository or an institutional repository. This helps to ensure that data are consistently available and accessible, are assigned permanent and unique identifiers for ease of citation, and are preserved for the long term. Discipline-specific repositories are recommended because they are recognized by the subject community and your data is collocated with similar datasets. re3data.org maintains a curated registry of research data repositories and includes a Repository Finder tool, hosted by DataCite, to help find the right repository for your data.
 
To prepare your data for long-term storage, data cleaning and documentation are critical to ensure that others can re-use the data and that any human subject data has been de-identified.  Files should be stored in non-proprietary formats if at all possible.
 
The UK Data Service provides extensive resources on data storage (https://www.ukdataservice.ac.uk/manage-data/store), recommended file formats (https://www.ukdataservice.ac.uk/manage-data/format/recommended-formats), and depositing data (https://www.ukdataservice.ac.uk/deposit-data).

Sources

https://www.re3data.org/

https://www.ukdataservice.ac.uk/manage-data/store

https://www.ukdataservice.ac.uk/manage-data/format/recommended-formats

https://www.ukdataservice.ac.uk/deposit-data
 

Assessment

How do I conduct a SWOT analysis?

You can learn about conducting a SWOT analysis here: https://www.health.state.mn.us/communities/practice/resources/phqitoolbo...

Sources

State of Minnesota Department of Health SWOT Analysis
 

How do I conduct an environmental scan?

A general overview of environmental scans can be found here: https://www.fordham.edu/info/26625/conducting_an_environmental_scan

For insight on how one library implemented an environmental scan, you can read Boise State’s 2017 Research Data Environmental Scan Report.

 
Where can I learn how to develop a survey?

There are a number of different places you can visit to learn how to develop a survey. Here are a few examples:
 

Are there best practices I should follow?

Yes, there are some general guidelines you should follow when designing surveys. Here are some best practices:
  • Before starting your survey, identify your purpose, target population, data management plan, privacy issues (will the responses be confidential? anonymous?), and any other purpose-related and practical questions that will come up through the process
  • Keep your survey as short and to the point as possible
  • Avoid vague questions or questions that have multiple parts
  • Avoid leading questions
  • Consider biases in the construction of the survey and research project as a whole and mitigate them
  • Avoid confusing matrices
  • Phrase questions in a way that allow survey respondents to elaborate if necessary
  • Here are some other resources:
  • When to Use What Research Design, W. Paul Vogt, Dianne C. Gardner, and Lynne M. Haeffele, Guilford Publications, 2012
  • “A Step-By-Step Guide to Developing Effective Questionnaires and Survey Procedures for Program Evaluation & Research,” Keith Diem, Program Leader in Educational Design, Rutgers University, https://njaes.rutgers.edu/fs995/
  • Qualtrics - “10 Tips for Building Effective Surveys”: https://www.qualtrics.com/blog/10-tips-for-building-effective-surveys/
 

Where can I find results of surveys others have conducted before me?

Where you look for survey data will largely depend on your purpose. For example, if you’re you seeking government data, you might consider going to data.gov. If you’re seeking data on a specific library, you would check their assessment page and/or reach out to their data librarian. Some other sources of data include the US Census Bureau, healthdata.gov, the World Health Organization, and UNICEF. NNLM data will eventually be made available visually through the NEO Dashboard project.
 

How do I create a needs assessment using an electronic survey tool such as Qualtrics or REDCap?

A needs assessment through an electronic survey would depend on the size/departments of the institution being assessed. The needs assessment could also focus on specific topic, such as spaces, technology, effectiveness, etc. From there, focus on asking users about their usage and having them scale different services (perhaps using a Likert scale or true/false questions). Here are some examples of needs assessments:

How do I get participation of clinical services staff/how do I promote the survey?

  • Send reminder emails at certain intervals (for example, six weeks, three weeks, one week, three days before the close of the survey). Be careful not to send too many reminders.
  • Offer an incentive (drawing, credits for a service offered, etc.)
  • Reach out to the main listservs for that community
  • Attend events that the staff attend and do outreach there if possible
  • Ask any contacts in the target population to promote the survey within their coworkers
 
Sources

Developing RDM Services

Who needs to be at the table when developing RDM services?

Potential partners in developing RDM services could include:

  • IT Department
  • VP Research
  • Office of Sponsored Programs
  • Other Librarians
  • Research Administrators
  • Scholarly Communications
  • PIs and other Researchers
  • Library Administration

How do I market the new resource to users?

Some ways to market new resources include:

  • Library newsletter
  • Campus email lists
  • Marketing around library (screens, banners, etc)
  • Embed in traditional library instruction
  • Research administrators and other campus partners
  • Contacts within academic departments

Who should I involve when developing a LibGuide?

Potential partners in developing LibGuides could include:

  • IT Department
  • VP Research
  • Office of Sponsored Programs
  • Other Librarians
  • Research Administrators
  • Communications Department

What resources should I include in a LibGuide?

You may want to customize your own library’s LibGuide, but you can view examples from the following published LibGuides:

Sources

Research Data Management at Princeton
Washington State University Health Sciences Research Data Management
NYU Health Sciences Library Data Management Subject Guide
 

How do I overcome “culture” issues; and/ or overcoming traditional views of libraries?

See also “How should librarians approach speaking with researchers about their data management practices?”

Some ideas from articles include:

  • Thinking outside the box
  • Building relationships
  • Hosting classes
  • Drop-in sessions
  • Consultations
  • Grass roots work -- start with undergraduate and graduate students
Sources

Bardyn TP, Patridge EF, Moore MT, Koh JJ. Health Sciences Libraries Advancing Collaborative Clinical Research Data Management in Universities. Journal of escience librarianship. 2018;7(2):e1130. doi:10.7191/jeslib.2018.1130. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6124496/

Faniel, Ixchel, & Lynn Silipigni Connaway. "Librarians' Perspectives on the Factors Influencing Research Data Management Programs." College & Research Libraries[Online], 79.1 (2018): 100. Web. 16 Oct. 2018. https://crl.acrl.org/index.php/crl/article/view/16610/18464
 

How do I transition from single informationist to team-based RDM?

The Data Curation Network project brings together the perspectives of research data librarians, academic library administration, and data curation subject experts from nine major academic institutions and general-purpose or disciplinary data repositories to launch a Data Curation Network
There are several papers about RDM collaborations at an institutional level such as Building a Research Data Management Service at the University of California, Berkeley  

Sources

Data Curation Network

Building a Research Data Management Service at the University of California, Berkeley
 

What are existing RDM collaborations lead by health sciences librarians?

In New England there is a group that meets a few times a year called the Research Data Management Roundtables. Originally a sponsored initiative from UMass Medical School, this community of practice discusses trending topics and other initiatives.

Sources

Atwood, Thea P., Condon, Patricia B., & Goldman, Julie. "Grassroots Professional Development via the New England Research Data Management Roundtables." Journal of eScience Librarianship 6(20): https://escholarship.umassmed.edu/cgi/viewcontent.cgi?article=1111&conte...
 

Learning and Teaching Opportunities

What credentials do health sciences librarians need to demonstrate capabilities (is an LIS degree enough or do they need to have another degree?)?

The MLA has identified profession competencies for health information professionals these are essential professional skills and abilities that can be observed, measured, and taught. Available at https://www.mlanet.org/p/cm/ld/fid=1217
“Librarianship is a people profession; a librarian’s job is to connect people with the information they are seeking, whatever format that may take” (Cragg and Birkwood, 2011).
The LIS includes a broad range of skills, which are all needed to help with data management. The framework of LIS fits with the role of data management (Bates, 1999).

Sources

MLA Professional Competencies

Federer, L. (2018). Defining data librarianship: a survey of competencies, skills, and training. (new window) Journal of the Medical Library Association: JMLA, 106(3), 294. https://doi.org/10.5195/jmla.2018.306

Khan, H. R., & Du, Y. (2017). What is a Data Librarian?: A Content Analysis of Job Advertisements for Data Librarians in the United States Academic Libraries. http://library.ifla.org/2255/1/139-khan-en.pdf
 

What learning opportunities are there for librarians who want to learn more about research data management?

For information on learning opportunities in research data management, visit:
NNLM Data Courses and Workshops
NNLM  RDM Webinar Series
 

How do I create a create introductory RDM class (data sharing and preservation)?

For information and ideas on how to create an introductory research data management class, visit:
Research Data Management Toolkit