Usage and citation data

Many repositories see the advantages of showing data for downloads of papers and will have features such as ‘most downloaded paper’. There are add-ons for both EPrints and DSpace to generate reports of usage statistics.

A JISC-funded project, Pirus2 (Publisher and Institutional Repository Usage Statistics) has been investigating ways to improve the data on usage at the individual article level. This is partly because funders have been showing a greater interest in having a reliable standard metric for article usage. Given that articles may be available from a number of different places – a journal publisher’s website, a database, a subject repository and an Institutional Repository, for example – this is not as straightforward as it may seem. The project aimed to enable all hosts of online journal articles, including repositories, to produce standard COUNTER-compliant usage reports which could then be consolidated at a higher level. It was considered that this would help repositories by ensuring their data was seen as authoritative, and connecting the repository more closely to the overall research agenda. The standard could also be applied to other material in the repository. Software plug-ins were developed for EPrints, DSpace and Fedora.

What I’m keen to do is… close that loop with the researcher. You persuade people to engage, they deposit their research… they need to see its been worth the investment of their time. So that means demonstrating impact, it means really good statistical feedback to them about the usage of their papers…, collating any kind of impact the deposited materials is having and feeding that back regularly.” Rachel Proudfoot

One repository has implemented a feature which gives individual authors information on downloads of their papers when they log in. This is a by-product of the use of unique institutional identifiers for authors and log-in using those institutional IDs. The repository hopes to do the same with citations through links with citation databases such as Scopus.

Hong Kong University has used the Scopus API to integrate citation data into its repository and researcher profile pages.

Business intelligence

In repositories forming core parts of the research management system – where the repository is the central publications database and links with other management information systems – it can generate valuable intelligence for the university on research performance. Where it acts as the full-text element in a CRIS-type set-up, it can both drive and measure usage and impact of research outputs. A repository can hold not just conventional research outputs but the evidence of impact activities, such as reports aimed at practitioners, video/audio of media interviews or public lectures and so on. Usage of these, as well as of the outputs themselves, can also be measured.

No comments yet.

Leave a Reply