The library science community has long been discussing the important issue of university library standards. Differences in education systems, economic development and culture make a unified set of standards for all countries impractical. This paper proposes a methodology for establishing standards based on existing libraries. A regression analysis is conducted to find the relationship between each of the three major resources of the library - collections, personnel and buildings - and the university size in terms of the number of students and faculty. The standards are derived from the regression lines. The data of the twenty-four university libraries in Taiwan reveal that the power function has the best explanatory power for all the three resources. The exponents of the power function are 0.8349, 0.5756, and 0.8703 for collections, personnel and buildings, respectively. All these values are smaller than 1, a sign of decreasing returns. Compared to the standards of the Association of College and Research Libraries, this set of standards is lower in collections and personnel, but similar in buildings. Since the education system, economic development, culture and other factors of a country are taken into account implicitly, empirical standards are more realistic and are achievable by the university libraries. The methodology developed in this study can be applied to establish standards for the university libraries of other countries.
All Science Journal Classification (ASJC) codes