At our February 25 RIOT, Robyn started with including us all, reminding us it’s not just the sciences requiring data management: the NEH (National Endowment for the Humanities) and the IES (Institute of Education Sciences) require data management, too. (Tellingly, perhaps, the NEH data management plan executive summary is a PDF. The IES offers a detailed data sharing implementation guide.)
In her post, Robyn outlined a dream workshop building graduate students’ information literacy through data management skills. At the session, we worked through how to move towards such workshop.
Start with a small, focused group: One starting point would be to focus the workshop by working with a small group of graduate students in one lab, or starting with one Principal Investigator, or doing a drop-in session at the Pickle campus. Then, move outward to a department or area. Research says data management education gets too expansive in interdisciplinary groups, so it’s best to focus on a small group.
Perspective of scale: Schools doing this well have teams of dedicated data curation librarians. Purdue (linked to in the original post), for example, has money to invest in infrastructure like this. So, we need strategies to make this scalable as we get started.
So what does this look like? From the librarian’s perspective, data curation might start with a website or online modules teaching students how to organize and manage their files, preparing students to think about and create effective metadata so that the data is truly shareable.
Starting where we are: Structurally, we can start by sharing what we have. Let’s let students know about Box and the UTDR. We also have data sets available through subscription to ICPSR (Inter-University Consortium for Political and Social Research); using this tool, we could build data literacy.
Then build skills: The critical scaffolding for using Box and UTDR effectively includes issues like copyright and the ethics of open data. These conversations open up space to think about how graduating students leave data behind, how well-structured data can increase citation.
Sharing data: This approach could build capacity within UT, since people want free data sets to use with undergraduate classes, particularly with the growth of digital humanities. Building UT awareness of data management frameworks and metadata best practices would allow UT classes to use UT student and faculty data sets, an exciting prospect.
Connected to Learning Commons Initiative: During focus groups for the learning commons, students and faculty said they wanted help using Excel to manage data. This would include both help using Excel and help understanding data analysis (as well as understanding that changing how we analyze the data can change our results).
- Build data management discussion/awareness into graduate student orientations.
- Add data management resources to research guides.
- Curriculum mapping to find out where students are using/creating data.
- Get started with small, scalable groups.
Our role is in building best practices, connecting researchers with data networks for the future.