Open Research and New Technology as Drivers of a Metrics Revolution

Beth Norgrove - Research Services Support Assistant, University of Manchester Library.

Our team (Library's Research Metrics Team) was recently challenged to think about whether the products and tools that we use in our day-to-day activities could be defined as “open” or “closed”. If they are closed, are there any open alternatives available that we could use? In the Research Metrics team, we regularly use bibliometric databases such as Scopus and SciVal to map research activity across departments, schools and across broad research themes, to explore our research collaborations, and to help researchers identify the “impact” of their publications. I was struck by the fact that most of the products that we currently use for these analyses are proprietary, commercial sources, where the data is hidden behind a paywall, or limited in terms of re-use by restrictive licences. This led me to reflect on why this is the case and, more importantly, what can we do about it? 

Commercial sources such as Elsevier’s Scopus and SciVal have the significant advantages of comprehensive training courses, access to specialist support and expertise, and user-friendly interfaces. The data that comes from these tools carries a certain amount of credibility and, for a long while, these products were the only tools librarians could use to help answer questions senior University staff posed about our research outputs. 

It goes without saying that we need to meet business needs, but how can we, as Library staff who advocate for openness, be rooted to using closed, commercial sources? The University of Manchester is fortunate to be able to afford the hefty price tags that come with Scopus, SciVal, etc, but of course this is not universally the case. Open infrastructure in this area is critical to ensuring that all have access to bibliometric data, regardless of their financial means. Another drawback, especially from the perspective of open evaluation, is the lack of transparency that comes with using closed sources. Though the Library is not directly involved in research assessment, we are committed to upholding the University’s Responsible Metrics statement, so we must ensure that data collection processes are as open and transparent as possible, and that the data that we provide can be verified by others. 

The movement to open up bibliographic data has gained momentum over the past decade. Various initiatives, such as the Initiative for Open Citations (I4OC), and community-developed infrastructure, such as OpenAlex, have emerged, helping to deliver greater accessibility, reproducibility and transparency. While it’s impossible to cover them all in one article, if you are interested in learning more, I would encourage you to read Section 1.1 of the Literature Review presented in the recent Harnessing the Metric Tide: indicators, infrastructures & priorities for UK responsible research assessment report. 

OpenAlex, whose name is inspired by the ancient Library of Alexandria, is a global index of research literature, journals, researchers and institutions, developed by the non-profit OurResearch. An alternative to Scopus and Clarivate’s Web of Science, unlike these sources, OpenAlex is free to access, and both the underlying data and source code behind the database is openly available. The database makes significant use of PIDs to pull data from various open sources, including Crossref, ORCID and the Directory of Open Access Journals (DOAJ). 

When OpenAlex launched in January 2022, the data could only be accessed via a (free) API or database snapshot. This wasn’t ideal given that none of us in our core team are familiar with using APIs to extract data. This is set to change though, as the emergence of Artificial Intelligence (AI) tools such as ChatGPT means that we can rapidly enhance our use of Python and, in turn, our ability to work with APIs. Technological advances such as ChatGPT are also creating exciting opportunities for analysis and visualisation of bibliometric data, above and beyond the capabilities of SciVal, which we’ve previously relied on. The opening up of bibliographic data, coupled with new technologies, is game-changing for everyone, but particularly for institutions with little (or even no) budgets for research intelligence. 

In late April 2023, OurResearch launched the alpha version of a web-based graphical user interface (GUI) for OpenAlex. The beta release is scheduled for July. This interface will not only make it easier for our team to access the data, but also for researchers and the non-academic community alike. 
Like other bibliometrics practitioners, our team will be following developments in this space in the coming months. We’re excited to get stuck in to using OpenAlex and other open bibliometric tools. We’re also in the process of creating a Research Metrics LibGuide for our website, where we plan to incorporate information on OpenAlex and how to use its GUI. By doing so, we hope to encourage greater awareness and adoption of these tools within our research community. 

Special thanks to my colleagues in the Research Metrics team for their feedback and suggestions.