Activity Data Synthesis

[This is a draft Guide that will be published as a deliverable of the synthesis team's activities. Your comments are very much welcomed and will inform the final published version of this Guide. We are particularly interested in any additional examples you might have for the 'Additional Resources' section]

The problem:
Libraries use a range of software systems through which users interact with premises, services and resources. The LMS system is far from the only source, the OPAC and the LMS circulation module representing increasingly partial views of user attention, activity and usage in a changing world. So libraries wishing to build a picture of user interactions face the challenge of identifying the appropriate data – depending on their purpose, which may range from collection management (clearing redundant material, building ‘short loan’ capacity) to providing student success performance indicators (if correlation can be established), to developing recommender services (students who used this also used that, searched for this retrieved that, etc).
Let’s split the problem down. In this guide we consider the variety of sources available within library services, a list to which you may add more. In other guides we consider strategies for deriving intelligence from ‘anything that moves’ as well as from targeted data extraction and aggregation with reference to specific goals.

The options:
Libraries already working with activity data have identified a range of sources and purposes – Collection Management, Service Improvement, Student Success and Recommender Services. Potential uses of data will be limited where the user is not identified in the activity (‘No attribution’). Here are some key examples:

Data Source	What can be counted	Value of the intelligence
Turnstile	Visits to library	Service improvement, Student success
Website	Virtual visits to library (no attribution)	Service improvement
OPAC	Searches made, search terms used, full records retrieved (no attribution)	Recommender system, Student success
Circulation	Books borrowed, renewed	Collection management, Recommender system, Student success
URL Resolver	Accesses to e-journal articles	Recommender system, Collection management
Counter Stats	Downloads of e-journal articles	Collection management
Reading Lists	Occurrence of books and articles – a proxy for recommendation	Recommender system
Help Desk	Queries received	Service improvement

Taking it further:
Here are some important questions to ask before you start to work with user activity data:

• Can our systems generate that data?
• Are we collecting it? Sometimes these facilities exist but are switched off
• Is there enough of it to make any sense? How long have we been collecting data and how much data is collected per year?
• Will it serve the analytical purpose we have in mind? Or could it trigger new analyses?
• Should we combine a number of these sources to paint a fuller picture? If so, are there reliable codes held in common across the relevant systems – such as User ID?

Additional resources:
Consider also the Guides on Student Success and Data Strategies

The Library Impact Data Project (LIDP) led by the University of Huddersfield - http://library.hud.ac.uk/blogs/projects/lidp/