Librarians’ attitudes to resource usage measurement and decision making in UK academic libraries: A content analysis

Purpose:
Libraries provide users with shared access to resources, and librarians balance limited funding with user requirements when managing their collection. Academic librarians use usage measures for a variety of decisions about their stock, as well as providing usage figures to senior university management to influence their decisions. There is a perception, among some academics and library researchers that there is an over-reliance on usage measures, to the point of substituting numbers for expert judgement, although librarians report using measures as part, rather than the whole of their decision-making process. The purpose of this paper is to explore how and why librarians use usage measures in decision making in practice and whether these behaviours and attitudes have changed over time by analysing the content of professional communications on mailing lists and comparing this to the literature review.

Design, methodology or approach:

This study was an unobtrusive, naturalistic inquiry into the attitudes of librarians, using the content analysis of publicly accessible, anonymised mailing list messages from 1998 to 2011. The research explored how and why librarians use usage measures for decision making in practice and purposively selected messages were analysed to identify common topics. These were examined for trends and key themes, based on and compared with an extensive review of the literature of resource usage measurement.

Findings:
The literature review thoroughly explored the issues surrounding the use of usage measures by librarians and provided sufficient thematic guidance to inform the creation of the coding classification scheme and validate the findings.
The major themes of standardisation and the impact of technological developments were identified and explored in depth, while topics specific to e-books measurement were isolated and examined as this area posed a number of challenges. A chronological survey explored trends in the use of usage measures by librarians and the identification of any changes in practice or attitudes. This process identified a major shift due to the widespread adoption of electronic resources, and the challenges posed by measuring resources not physically held in the library. Following this, the introduction of the COUNTER code of practice made reliable and comparable vendor-supplied statistics possible and widely available. Challenges in the measurement of e-books continued to pose problems for librarians, from the middle to the end of the study period. The introduction of e-measures questions in the annual SCONUL return prompted many to begin gathering electronic resource statistics on a routine basis, and this increased expectations of usage measures, especially with the promise of automated systems to analyse the numbers.

The research was unable to conclusively prove that librarians never substituted usage measures for professional judgement in practice, but there was strong evidence that they took the views and needs of their user community very seriously. They used measures to exclude high use items from further consideration rather than automatically removing low-use materials, although many did make reference to the cost per use calculation to determine whether a subscription or document delivery option would provide better value for money. Measures were very often used to determine patterns of use, rather than being concerned with absolutes, and the upward or downward trends for a resource or collection carried more weight. Zero-use items were a special case, especially if this was an ongoing situation, as they were not providing even limited benefit to the user community. However, even these were investigated to ensure that the library had done everything possible to make them accessible to their users.
Although the research found few areas which had not been covered in the literature, it did confirm that the literature represented an accurate view of the attitudes of librarians towards the use of usage measures. While published case studies may demonstrate best practice techniques, the aspirations for the assessment of library collections were in alignment. Several of the most frequent contributors to the lists had also written papers on the subject, suggesting that the published literature has a strong basis in real practice.

Research limitations:
This study applied a method which enabled the researcher to assess the accuracy and appropriateness of the published literature through the unobtrusive examination of documents produced in the course of professional discussions of librarians. The research was carried out by one individual, and therefore does not have inter-coder inaccuracies which might be present in other content analyses, although this also means that the coding was not tested
by an alternative analyst, and as a result it is possible that the coding reflects too closely the attitude of the researcher. However, the grounding of the coding scheme in the published literature, and the closeness of the topics and attitudes identified to existing survey and other research suggests that this balances with the potential bias inherent in other methods.

Conclusions:
Librarians make use of usage measures to manage large collections, but also employ other measures to ensure their assessments balance value with cost effectiveness. Usage statistics are particularly valuable for the assessment of large resources, such as big deals, although librarians expressed concerns over their quality and reliability. The COUNTER code of practice has been the most significant agent of change in the measurement of resource usage, promoting both comparability and ease of use. Technological developments such as JUSP offer librarians the potential for sophisticated analysis, but can also impact on the meaning of figures, and the value of librarians who understand both the limitations of the statistics and the nuances of their collections cannot be underestimated. Open communication between librarians, and with providers, is key to improving the provision and quality of usage statistics, and ensuring the measures are applied to decisions effectively.

Originality and value of the proposal:
This research fills a gap in the literature for naturalistic studies of librarians’ behaviour and attitudes in relation to usage measures. It will be of benefit to librarians, as well as statistics providers and publishers who will gain a better understanding of the needs and priorities of librarians and how they apply the data supplied to decisions in practice.