Research Data Access and Preservation Summit, 2015 Minneapolis, MN April 22-23, 2015 Leslie Barnes, Dylanne Dearborn and Andrew Nicholson

1. Co-curricular RDM: a Pilot Service for Graduate Students at the University of Toronto
Leslie Barnes, Digital Scholarship Librarian, Dylanne Dearborn, Research Data Librarian, Andrew Nicholson, GIS/Data Librarian
University of Toronto Libraries
ald
Background
In response to the lack of research data management instruction at University of
Toronto (UofT), members of the Research Data Management Working Group at
University of Toronto Libraries (UTL) embarked on developing a curriculum in
RDM that would focus on graduate students, a group poised to have a broad im-
pact on RDM practice throughout the university.
Before actively developing the curriculum, we held an open mic event marketed to
graduate students, “Speed Data-ing: an Open Mic Event Without the Romance.”
Participants were invited to share their research data stories to facilitate discussion
among peers and enable librarians to discover discipline-specific research needs,
issues, and workflows, which in turn allowed us to target areas for instruction.
A Co-curricular approach
Because graduate students were our primary audience, we built upon previous in-
structional collaborations between the UTL and the School of Graduate Studies
(SGS) in the Graduate Professional Skills (GPS) program, through which graduate
students can attain a transcript notation for completion of 20 GPS workshop cred-
its. Workshops are offered in skill areas designed to prepare graduate students for
success in their studies and professional careers. Other campus partners delivering
workshops include the Academic Success Centre, the Centre for Community Part-
nerships, the Office of Research Ethics, and the Career Centre.
Sources of Content
To prepare our curriculum, we conducted an environmental scan of RDM courses
and workshops available at universities in the United States and Canada.
We wanted to find out:
• Were they offering workshops for their Graduate students?
• What kind of teaching models were they using?
• What content were they including and emphasizing?
We found research articles documenting Librarians’ experiences with developing
RDM workshops for their researchers1
and were especially interested in RDM pro-
gramming at the University of Massachusetts Medical School, University of Wash-
ington, University of Manitoba, and University of Houston.
The New England Collaborative Data Management
Curriculum (NECDMC)
Popular among Librarians looking to develop their own RDM workshops, the
NECDMC is a collaborative education project developed by New England Librar-
ians who work primarily in STEM-focused disciplines to educate researchers in
RDM skills. The NECDMC is comprehensive, covering all aspects of a research
data management program from defining to sharing and archiving research data.
The developers of the NECDMC have made all of their materials available online
in re-usable formats to promote current and future collaborations among research-
ers.
NECDMC: Strengths for us
• Linear, step by step framework to walk participants through RDM competencies
• Allows for subject agnostic teaching
• Easily flexible and adaptable
NECDMC: Weaknesses for us
• Consists of seven modules, each 3 hours in length.
• Often taught over a seven week period
1 Ishida, M. (2014). The New England Collaborative Data Management Curriculum Pilot at the University of Manitoba: A Canadian
Experience. Journal of eScience Librarianship 3 (1), 80-85. http://dx.doi.org/10.7191/jeslib.2014.1061
Kafel, D., Creamer, A.T., & Martin, E. R. (2014). Building the New England Management Curriculum. Lessons Learned From a Re-
search Data Management Pilot Course at an Academic Library. Journal of eScience Librarianship 3 (1), 60-66. http://dx.doi.org/10.7191/
jeslib.2014.1066
Muilenburg, J, Lebow, M., & Rich, J. (2014) Lessons Learned From a Research Data Management Pilot Course at an Academic Library. Jour-
nal of eScience Librarianship 3 (1), 67-73. http://dx.doi.org/10.7191/jeslib.2014.1058
Peters, C. & Vaughan, P. (2014). Initiating Data Management Instruction to Graduate Students at the University of Houston Using the New
England Data Curriculum. Journal of eScience Librarianship 3 (1) 86-91. http://dx.doi.org/10.7191/jeslib.2014.1064
Modifying NECDMC for use at UofT
Excited about NECDMC, we had to modify the curriculum to fit the GPS format.
Our requirements were:
• Three hours in length maximum
• Discipline and domain agnostic
• Workable for 30 participants and three presenters
• Include references to Canadian specific policies and examples
Distilling the main themes of each of the seven NECDMC modules
and compressing them to fit into one three hour workshop was our biggest chal-
lenge. Over several meetings and many iterations we developed a curriculum that
prioritized aspects of research data management and took into account our diverse
graduate student population and their strong subject expertise in their respective
fields.
Format
We aligned our curriculum with a DMP: each part and corresponding activity
covered a key aspect of RDM and at the conclusion of the workshop, participants
would have created their own simplified DMP. See the chart (right) for an outline
of the course.
We chose to format the course in this manner to ensure that students left the three-
hour workshop with a concrete starting-point for practicing RDM. Additionally,
we created a LibGuide in support of the workshop that included summary de-
scriptions of best practices, copies of the slides used in the class, and links to re-
sources on campus and beyond.
Introduction to
DM and DMPs
Activity One
Decide on and then describe your research data project to your
neighbour
Collecting and
Organizing
Activity Two
Identify your data formats, capture/creation processes, and who
will be responsible for each data task
Metadata
Activity Three
A year from now...What information do you need to make sense of
your data? How will you capture or create this metadata?
Storage and Se-
curity
Activity Four
How would you store/back up your research data differently from
other files? Do you have confidential or sensitive data?
Retention and
Preservation
Activity Five
What data will you be keeping? For how long? Who will be re-
sponsible?
Sharing and
Publishing
Activity Six
Are there reasons NOT to share your data? Who might be interest-
ed in the data? What are the possible future uses of the data?
Wrap Up
• ‘Data is Diverse’
• Data lifecycle
• Why manage your data
• DMP samples
• Creating a DMP
• Understanding your data
• Roles and responsibilities
• Documentation
• File structures
• Naming conventions
• Sharing requirements
• IP rights
• Repositories
• Data as publication
• Deciding what to keep (and
what not to)
• Preservation policies and
technical requirements
• Backing up research
• Secure Storage
• Confidential and sensitive
data
• Types of metadata
• Metadata standards
• Metadata for data
• Common fields
• RDM Resources at UofT
• Overview of RDM best prac-
tices
• Your simplified DMP
• Repositories
• Data as publication
Testing and Delivery
First test run: delivered to a small group of data management librarian specialists.
Having extensively modified the NECDMC, we wanted to first present the course to
subject specialists to ensure that we had not sacrificed depth for brevity.
Second and third test run: delivered to librarians throughout the UTL system.
Our two general test runs yielded a wide range of feedback on pacing, organization,
and areas for expansion and contraction while introducing librarians in many fields
to RDM, laying the groundwork for further training.
Delivery to graduate students: On March 5, 2015 and April 15, 2015, we delivered
the curriculum as part of the GPS workshop series, with 23 and 19 student attending,
respectively.
Demographics and Expectations
“Intro to Research Data Management” was marketed through the GPS and grad
events email newsletter and by including the registration link in the list of current
GPS offerings.
After registering for the workshop through the GPS website, students were sent a
pre-course survey via the GPS Survey Monkey account. They were asked about
their departmental affiliation and what they hoped to get out of the course. Students
who did not fill out the online survey were encouraged to fill out a paper copy of the
survey prior to the class beginning.
Of the 42 students who filled in pre-course surveys, 21 said that they wished
to learn how to manage data. More specific topics student hoped to learn
about including organization (19), software and tools for managing data
(7), data reuse (3), making data accessible and discoverable (3), and how
to manage large quanities of data (3). Several students had more specif-
ic desires, including encryption (1), research ethics (1), and data citation (1)
Feedback
At the conclusion of each session, we invited students to fill in a simple, anonymous
evaluation sheet. We asked four questions:
1. What do you like best about this course?
2. What topic or topics do you feel should have been addressed more in this course?
3. What would you like to change about the course?
4. Did you find this course helpful for your research practice?
We received 18 responses to the first class delivery and 17 responses to the second
class delivery, totalling 25 completed questionnaires.
Response to the course was overwhelmingly positive: respondants almost unani-
mously reported that they found the course helpful to their research practice in re-
sponse to question #4.
In response to what students liked best about the course (question #1), the most fre-
quent responses were the broad scope of the course (8); metadata (5); the interactive
format of the course (4); introduction to DMPs (4); and that the course was well or-
ganized (3).
In coding responses to questions #2 and #3, we found responses overlapped consid-
erably, and therefore merged the results of these questions. By far the most common
response to these questions were that students craved more practical, concrete exam-
ples of how to manage data (14). The next most common concern was a desire for
more discussion tools and software to support RDM (9). The third and fourth most
common responses were that students wished for a more domain- or discipline-spe-
cific approach to the material (5) and that students wanted more detail on the subject
of metadata (4). Other themes included a clearer and more detailed explication of
DMPs, their creation, and their function (3), and making the course longer or split-
ting it into multiple workshops (2).
What Students liked most
0 1 2 3 4 5 6 7 8 9
Well organized
DMPs
Interactivity
Metadata
Broad Coverage
What students would like more of
Interpretation and Future Plans
Based on the feedback we received, it was clear that students wanted concrete and
practical advice, whether that came in the form of explicit examples or specific tools
to perform particular tasks. Emphasis on these facets of RDM would likely necessi-
tate increased domain-specificity because tasks will likely rely on what kind of data a
researcher possesses (e.g. qualitative, quantitative, images, tabular data).
We were surprised that students engaged with metadata as much as they did, and ac-
knowledge that metadata, like specific RDM tasks, feeds into more domain-specific
approaches, especially when discussing discipline-specific metadata standards.
The format of the course hit the mark; students enjoyed the interactivity, organiza-
tion, and scope of the course. Students appeared to like the general, introductory na-
ture of the course and many described learning about topics that were entirely new to
them, including data sharing, data papers, and respositories for finding data.
Partnering with the GPS program gave us a recognized platform to deliver a work-
shop targeting graduate students from all disciplines, offering a GPS credit as incen-
tive to students for attending. We plan to continue with this general workshop as part
of the GPS offerings and roll out the session at our other UofT campuses as well as
modifying it for use in specific departments or groups such as our high performance
computing centre, SciNet. We are currently investigating ways to expand based on
the feedback received, with the option to develop more advanced sessions looking at
more discipline/domain focused issues of RDM and/or tool oriented sessions around
topics such as metadata.
Contact us!
leslie.barnes@utoronto.ca; dylanne.dearborn@utoronto.ca;
andrew.nicholson@utoronto.ca
0 2 4 6 8 10 12 14 16
Metadata
Domain specificity
Tools and software
Concrete examples