The Embedded Data Consulting Model and the "IDEAL" Collaboration

By Natsuko Nicholls posted

Embedded librarianship has increasingly shifted the emphasis from one-time reference services to developing highly focused, targeted, specialized research assistance. Although embedded liaison programs have been previously adopted across institutions, embedding data consultants is a relatively new approach that many academic libraries have introduced into the process of building research data services. In this blog post, I’d like to share my personal and professional experience as a researcher embedded in faculty research—as one form of data consulting—which enabled me to become fully engaged in the work of the team, group, or department in order to offer direct and personalized research assistance over an extended period of time to faculty and student researchers.

As the Research Data Consultant at Virginia Tech from fall 2014 until recently, I worked with faculty and PhD students as a collaborator and as an integral part of two research teams. My initial contact with the teams, Integrated Digital Events Archiving and Library (IDEAL) and Global Forum for Urban and Regional Resilience (GFURR), stemmed from data-related reference questions that would traditionally have been answered in a more cursory fashion within less than a week. Yet, because I quickly understood the teams’ unique research needs and how my domain knowledge and data curation expertise could be applied to the process of their research, I was able to help the teams for a more sustained period of six to eight months. 

Using the first team as an example, I collaborated with the IDEAL project, funded by NSF, mostly because the project involved digital libraries and archives. Four co-PIs from computer science, business, and sociology are leading this project—collecting, curating, analyzing, and giving access to tweets and web pages about important events. They have over 1,000 different collections of tweets from recent years, totaling about 1 billion tweets collected through careful selection of relevant key words, hashtags, and names of individuals and organizations. I helped the team compile information about important global events around the themes of, for example, natural disasters, climate change, and international conflicts. I also helped write a new grant proposal to NSF (serving as a faculty associate on the proposed project) to build on their current funded project, and co-wrote and reviewed their data management plan.

While attending weekly project meetings, it became apparent that the data, collections, and research output IDEAL was archiving have value for other researchers to re-use. During the process of data collection (tweets and websites in the case of IDEAL), the project team was already thinking ahead about how collected information and data would be discovered through a search engine /interface to be re-used/repurposed by other researchers. This is where individual researchers' goals intersect and complement library's mission to preserve and make accessible scholarly products, including research data.

There are many ways—and capacities—in which a data consultant can become embedded in faculty research teams. What helped me stay engaged in the research partnership was the interdisciplinary nature of the project, team members’ interest in accepting an external research partner, and, most importantly, the support from the library to dedicate some of my time and effort toward an equal research partner.

Challenges included sustainability, in terms of finite consulting time/ability; managing time across different projects; and being accepted as an equal/qualified partner, along with earning trust from a research team. None of these challenges are insurmountable, and their solution strengthens ties between their larger university community and the library with a large pay-off potential for funded research. 

Academic research libraries are quickly developing support for research data management, particularly through consulting services. I’d like to reiterate that data consultation takes different forms at various stages of the research lifecycle—ranging from one-time reference services, to pre-award data management plan reviewing, to post-award curation workflow support; therefore, the embedded data consulting model offers the potential for librarians and consultants to apply their knowledge and expertise in new ways that can directly influence the value proposition of library data services. The embedded consulting model shows the impact that consultants can and do have beyond the traditional functions of the library, and why librarians and data consultants are needed now more than ever as technology paves the way for the "data revolution."

Natsuko Nicholls is Data Manager at the Institute for Research on Innovation and Science (IRIS), Program in Innovation, Networks and Knowledge (INK), at The University of Michigan Institute for Social Research Survey Research Center