A brief history of library analytics

We are just finalizing a chapter for a forthcoming Facet publication, the following section didn’t make the final cut, but we thought we would reproduce it here for anyone interested.

The literature shows an interest in the relationship between library use and undergraduate attainment stretching back to the 1960s and 1970s (Barkey, 1965; Lubans, 1971; Mann, 1974), however, until recently literature reviews looking into this area have found little evidence of more research until the last few years.

Some studies have investigated the relationship between university library usage and undergraduate student outcomes (De Jager, 2002a; De Jager, 2002b; Emmons and Wilkinson, 2011; Han, Wong and Webb, 2011), however, all lack information on electronic resource information use, De Jager points out that further investigation is necessary to discover where electronic resources play a part in achievement. Additionally, recent research has considered the relationship between library value and impact on research and learning (Oakleaf, 2010; Tenopir and Volentine, 2012). These studies have found that the library supports key academic research activities and thus can be considered to make a vital contribution to university value.

Over the past few years’ more detailed research on ‘library analytics’ has been gathered in the UK, US and Australia; Huddersfield (Stone, Pattern and Ramsden, 2012; Stone and Ramsden, 2013; Stone and Collins, 2013; Collins and Stone, 2014), Wollongong (Cox and Jantti, 2012; Jantti and Cox, 2013) and Minnesota (Soria, Fransen, and Nackerud, 2013; Nackerud, Fransen, Peterson and Mastel, 2013). These three projects have all independently suggested a correlation or statistical significance between library usage  – e-resources use, book loans and gate entries – and student attainment. It is important to note, however, that this relationship cannot be considered a causal one.

The advantage of a more data driven approach over surveys (Chrzatowski , 2006, Whitmire, 2002) is that data can be captured for every student in an institution, or across institutions, which removes the issue of low survey return rates and potential bias in survey responses or interpretation. Another benefit of using linked data from student registry systems is that far more information can be interpreted, for example demographic characteristics and discipline in addition to degree classifications and grade point average. Student retention can also be investigated using this data.

With regards to research into demographic data in academic libraries, a number of studies have been undertaken in the United States (Whitmire, 2003; Jones, Johnson-Yale, Millermaier and Perez, 2009; Green, 2012).  Of the more analytics driven studies, Cox and Jantti (2012) reported on gender and age differences. Many of the more recent studies have also looked at discipline, in some cases producing consistent finding, for example, arts and humanities are usually found to be the biggest users of physical library materials (De Jager, 2002a; Maughan, 1999; Whitmore, 2002) and many studies have found engineering students to be the least engaged library users across resources (Kramer and Kramer, 1968, Bridges, 2008 and Cox and Jantti, 2012, Nackerud et al, 2013).

The references included here can be in the bibliography of library analytics maintained by this blog.