A BETA‐BINOMIAL MODEL FOR LIBRARY SURVEY DATA

It has become of increasing importance to the UK library community to estimate the proportion of library acquisitions in the United Kingdom for which a UK MARC record is available at the time of cataloguing. This proportion is called the hit rate. The libraries have for the purpose of the study been divided into two sectors, academic libraries and public libraries. This paper uses a beta‐binomial model to enable us to attribute standard errors to each of our overall hit rate estimates and to investigate the difference between the hit rates of the two populations. The model also provides an estimate of the individual hit rate for any given library. Such estimates are useful for identifying libraries which are not typical of the library population from which they are drawn. The model is applied to data obtained from the Centre of Catalogue Research.