Research analytics: where do libraries fit in?

Photo by William Warby on Unsplash

A researcher in the economics department wants to find a colleague on campus with expertise on how different cultures form trust networks, to join a grant proposal for a project exploring the role of trust in market-based exchange. How can the researcher find an appropriate collaborator?

A staff member in the Research Office is collecting information on the research activity of the university, such as publication counts for a variety of research outputs, total external funding awards, and recent academic awards and recognitions. This information will be submitted to an important global rankings exercise – a key component of the university’s reputation and brand. Can this information be easily located?

As these questions suggest, research analytics – creating and delivering intelligence on a university’s research enterprise – is a matter of practical concern on many campuses. The data needed to support research analytics may be scattered across many sources, both internal and external to the university: for example, the institutional repository, personnel records, faculty CVs, and bibliometric and research impact databases like Web of Science or Google Scholar. Some of this information may be consolidated in a research information management system (RIMS) like Pure or Symplectic Elements, which in turn is populated by a combination of automated data feeds and manual entry.

As more and more data on research activity is collected, the range of questions that can be answered with it grows commensurately. A new service area is emerging around research analytics, but what is the role of the academic library? The OCLC Research Library Partnership (RLP)’s Research Support Interest Group recently welcomed Brian Mathews, Associate Dean for Innovation, Preservation, and Access, and David Scherer, Scholarly Communications Librarian and Research Curation Consultant, to lead a virtual discussion (actually two discussions, to accommodate RLP participants across many global time zones) on research analytics at Carnegie Mellon University, where the University Libraries recently deployed a new RIMS and is developing a research analytics service. Brian and David shared their experiences at Carnegie Mellon, which led to rich conversations and sharing of insights among participants from a number of RLP Partner institutions. Here are a few of the themes that resonated in the discussions:

  • Research analytics is a developing area, with lots of uncertainty – including what a service might look like. It was clear from the discussion that there is as yet no established path toward operationalizing research analytics as a service. Uncertainty touches even the most fundamental issues, such as how to define research analytics and what a research analytics service would look like. Brian and David from Carnegie Mellon observed that effective engagement with potential stakeholders required the ability to produce customized reports and visualizations; “stock” analytics would not be enough. Another participant suggested focusing on standardized services of value to many users, while at the same time training people to use analytics tools themselves for more specialized purposes.
  • Is research analytics a new service area for academic libraries? Several participants pointed out that libraries bring to bear a great deal of relevant expertise, such as bibliometrics, data management, and even experience in dealing with vendors and licensing (useful for purchasing systems like RIMS or securing access to data sources). At Carnegie Mellon University Libraries, where data is considered part of their collections, the move to research analytics is seen as a natural progression. Brian and David also emphasized that at their institution, the role of the Library is to provide lots of “carrots”, or incentives, for researchers and staff to contribute data. As David put it, they see themselves more like H&R Block (a US tax preparation service) helping researchers recognize and meet data requirements, rather than the Internal Revenue Service (the US tax collection authority). Encouraging the contribution of complete, accurate data is vital: research analytics will only be as good as the underlying data.
  • Interoperability is a big obstacle. The data needed for research analytics often resides on multiple university systems. To gather and synthesize it for analytics purposes, it often has to be re-entered or migrated into yet another system, such as the RIMS. A number of participants in the discussions raised the point that data interoperability across campus systems – and between campus systems and external systems – needs to be improved. The ideal, as one participant put it, is “one touch – enter and re-use”. Requiring researchers to duplicate effort by entering the same data into multiple systems is a clear disincentive for data contribution, and an important obstacle to overcome in seeking researcher buy-in.
  • What will research analytics be used for? The ability to generate information about a university’s – or an individual’s – research activity presents both opportunities and concerns. Research analytics has many practical applications, ranging from bringing together researchers with mutual interests, to helping the university manage and promote its scholarly reputation. Participants underlined the importance of having compelling use cases in hand when talking about research analytics with other campus units. A good starting place might be surfacing campus expertise and identifying networks of collaboration, which several participants indicated was a priority on their campuses. But be prepared to address concerns as well: in particular, those surrounding the use of metrics which might be construed as evaluating an individual researcher’s performance or impact.
  • Libraries need to team up and staff up. Developing a research analytics service can require a significant investment in staff resources. Several participants noted that staff scarcity is a key limitation they face in deploying RIMS and utilizing them for research analytics. A library building a research analytics service may initially assign this responsibility to existing staff, but for many libraries, this is not sustainable – as one participant pointed out, their institution currently employs only one bibliometrics librarian. Several participants emphasized the importance of leveraging limited staff by teaming up across campus units – for example, in cross-unit working groups – regardless of where the central administration of the RIMS (or research analytics service) is located. For example, the library might work with individual academic departments or the university press to ensure researchers sign up for ORCID IDs.

Teaching, research, and service are the three great missions of the modern university. Research analytics is a data-driven method for cultivating a better understanding of the research mission, including the types and scope of the intellectual capital produced by the university and its impact in the scholarly community and beyond. Our discussions, sparked by Brian and David’s experiences at Carnegie Mellon, highlighted both the promise and uncertainty of this new service area, as well as the possibilities it presents for academic libraries.

Conversations like this help share knowledge among RLP members, and also inform our work here in OCLC Research, such as our new project looking at institutional stakeholders in research support, as well as our continuing investigations into research information management practices. If you are affiliated with an RLP member institution and would like to join the Research Support Interest Group, please click here to sign up!

Thanks to colleagues Rebecca Bryant and Annette Dortmund for helpful suggestions for improving this post!