Circulation statistics in the evaluation of collection development

Purpose - This article aims to describe how circulation statistics may be used to evaluate collection development policies. Design/methodology/approach - The circulation statistics of books acquired by a science library in a specific year are analyzed by publisher, publication date, and subject. Findings - The paper finds that older books circulated more than recently published titles purchased at the same time. Circulation averages varied considerably between publishers. Research limitations/implications - Checkouts are an imprecise measure of value. Number of items not purchase costs is the denominator of all averages used; there is a data bias against inexpensive books. Originality/value - The procedure outlined can be used generally to evaluate collection development policies.