The first assumes that annual loss rate is more useful than a total inventory as a means of providing the information necessary to make decisions about the need for security in libraries. The second involves the manner in which some libraries interpret the cost of lost materials for the purpose of library security studies. The third deals with the role that factors beyond monetary value play in evaluating the need for security measures. In the process of analyzing these three strategies, the author discovered that the positions he takes in response to them are logically connected. (Author abstract)