The team is currently prepping for our first Community Advisory Board (CAB) meeting for the Jisc LAMP project. There’s a great deal to discuss, not least the use case ideas we have been drafting for feedback. Ben Showers and I met last week to talk about setting the context for the meeting, and we agreed that it would be useful to more broadly share the findings of the survey we ran back in November 2012. With the support and input of RLUK and SCONUL, Mimas worked with Jisc to run a community wide survey. We wanted to gauge the potential demand for data analytics services that could an enhance business intelligence at the institutional level and so support strategic decision-making within libraries and more broadly. Below is a summary of the results available through slideshare.
- Automated provision of analytics demonstrating the relationship between student attainment and resource/library usage within institutions
- Automated provision of analytics demonstrating e-resource and collections (e.g. monographs) usage according to demographics (e.g. discipline, year, age, nationality, grade)
- Resource recommendation functions for discovery services
Perhaps not surprisingly, the overwhelming response was positive – these tools would be valuable, yes (over 90 % ‘yes’ rate each time). But we also asked respondents to indicate which strategic drivers were informing their responses, i.e. supporting research excellence, enhancing the student experience, collection management, creating business efficiencies, demonstrating value for money, and others. What we found (based on our sample) was that the dominant driver was ‘enhancing the student experience,’ closely followed by the ability to demonstrate value for money, and then to support research excellence.
We also asked whether institutions would find the ability to compare and benchmark against other institutions would be of value. Whilst there was general consensus that this would be useful, respondents also indicated a strong preference to share data to be used as a benchmark for other institutions if it were anonymised and made available by a category such as Jisc Band (91%) (This compared to a 47% ‘yes’ rate when asked if they would, in principle, be willing to make this data available where users could see the source institution’s name). So, there is appears to be a strong willingness to share business intelligence data with the wider community, so long as this is done in a carefully managed way that does not potentially expose too much about individual institutions. In addition, there was far more hesitation over sharing UCAS and student data than other forms of transactional data (again, not surprising).
Are analytics a current strategic priority for institutions? Only nine respondents said yes it was a top priority at the present moment, with 39 stating that it was important but not essential. However, when asked whether it would become a strategic priority in the next five years, 40 respondents indicated it would become a ‘top priority.’
However, the question of where the decision-making in this area would reside evoked a wide range of different responses, indicating the organisational complexities we’d be dealing with here. Clearly the situation at each institution is complex and highly variable. Overall Library Directors and IT Directors are seen as the key decision-makers, but respondents also referenced Vice Chancellors, Registrars, Deputy Vice Chancellors. At certain individual institutions, the University Planning Office would need to be involved, or at another, the Director of Finance.
Other potential barriers to sharing include concerns over data privacy and sharing business intelligence, and our results revealed a mixed picture in terms of concerns over data quality, lack of technical expertise, and the fact that there are strong competing demands at the institutional level.
The LAMP project is now working to build on these findings and develop live prototypes to fully test out these use cases, working with data from several volunteer institutions. Our major challenge will be to ascertain to what extent the data available can help us support these functions, and that’s very much what the next six months is going to be focused on.