In spite of their evolution over the last few decades, accelerated most recently due to the Googlization of information, public libraries have been amazingly impervious to change in the arena of performance measurement. I found the following observations about library measures in the early history of American libraries:
There is no branch of library economy more important, or so little understood by a librarian as helps to himself, as the daily statistics which he can preserve of the growth, loss, and use (both in extent and character) of the collection under his care. The librarian who watches these things closely, and records them, always understands what he is about, and what he accomplishes or fails to accomplish. The patrons to whom he present these statistics will comprehend better the machinery of the library, and be more indulgent toward its defects.1
Interesting that use of library statistics for advocacy purposes was recognized in 1876. Early in the twentieth century our current ideas about performance measurement were already well understood, long before the practices of program evaluation, evidence-based management, and performance scorecarding were formalized. Arthur Bostwick, late director of St. Louis Public Library and Librarian of Brooklyn Public Library, wrote this in his book published in 1917:
No business can be properly carried on without a system of accounts. These may involve only money received and expended, but they may and should extend much further. The collection and tabulation of such [financial and performance] data have come to be regarded as indispensable by shrewd businessmen; and large corporations do not hesitate to spend considerable sums in employing a force of experts and clerks especially to gather data of this kind and to tell what they mean…2
Bostwick also dealt with the ideas of accountability, continuous improvement, careful analysis of library statistics, and a tiered approach to evaluation data-collection:
Information of this [financial and performance] kind is gathered with either or both of two different purposes in view—to satisfy the legitimate curiosity of the person managing the business, or of some one who has a right to know how it is going on, whether it is succeeding or failing and just what it is accomplishing; and, secondly, to furnish a basis for improvements or changes, to indicate weak points and points of strength, so that the business may be reënforced along the former and extended along the latter.
…If the latter [purpose is intended], a more detailed and analytical study is made of the data, which are compared and tested in all possible ways to reveal unsuspected facts. When something is thus brought to light that seems to call for further investigation, additional data are collected.3
He also preached about the imprecision of library statistics, a topic conveniently overlooked in our profession nowadays (as is the importance of assuring the validity of data presented in library advocacy reports also):
It should not be forgotten, either by those who collect and report these statistics, or by those who read them or use them, that they are of various degrees of exactness…In any kind of scientific measurement the limits of probable error are always mentioned to give an idea of the degree of accuracy. The less the probable error, the greater the accuracy. It is never stated that there can be no error and that the accuracy is exact; this would be simply ridiculous. The same holds good in library statistics. In the average report nothing at all is said of accuracy; the reader is left to conclude that all the data are exact, or at least that there is no difference in their degree of exactness.4
Finally, the level of interest in this topic among Bostwick’s 20th century peers strikes a familiar chord:
But how much intelligent study of library statistics goes on in librarian’s offices, and how much modification or improvement in library methods and material results from such study, is something that we shall never know. It appears to be certain, however, that large numbers of librarians…look upon their statistics in the light of a necessary evil. They must be collected, because some thing of the kind is expected in the annual report, but they should be minimized, and, once in print, they should be dismissed from the mind. This attitude reminds one of the rural workman who used a dull saw because the amount of work before him gave him no time to stop and sharpen it…5
1 Warren, S. R. & Clark, S. N., Eds., 1876, Public Libraries in the United States of America, Washington, DC: U.S. Bureau of Education, p. 714.
2 Bostwick, A. E., 1917, The American Public Library, New York: D. Appleton & Co., p. 253.
3 Bostwick, A. E., p. 254.
4 Bostwick, A. E., p. 262.
5 Bostwick, A. E., p. 255.