Most borrowed is most cited? Library loan statistics as a proxy for monograph selection in citation indexes

TL;DR

This study investigates whether library loan data can serve as a reliable proxy for monograph use and influence inclusion in citation indexes, finding limited correlation and highlighting data challenges.

Contribution

It provides an exploratory analysis of library loan data's potential as a selection criterion for citation indexes, revealing current limitations and data aggregation issues.

Findings

High percentage of loans in national languages (96%)

Low correlation between loans and citations

Loan data currently insufficient for use in citation index selection

Abstract

This study aims to analyse whether library loans statistics can be used as a measure of monograph use and as a selection criteria for inclusion in citation indexes. For this, we conducted an exploratory study based on loan data (1000 most borrowed monographs) from two non-Anglo-Saxon European university libraries (Granada and Vienna) with strong social sciences and humanities components. Loans to scientists only were also analysed at the University of Vienna. Furthermore, citation counts for the 100 most borrowed scientific monographs (SM) and textbooks or manuals (MTB) were retrieved from Web of Science and Google Scholar. The results show considerable similarities in both libraries: the percentage of loans for books in national languages represents almost 96 per cent of the total share and SM accounts only for 10 to 13 per cent. When considering loans to scientists only, the percentage of English books increases to 30 per cent, the percentage of SM loans also increases (approx 80 per cent). Furthermore, we found no significant correlations between loans and citations. Since loan statistics are currently insufficient for measuring the use of monographs, their suggested use as an applicable selection criterion for book citation indexes is not yet feasible. Data improvement and aggregation at different levels is a challenge for modern libraries in order to enable the exploitation of this invaluable information source for scientometric purposes.