Universities conduct a variety of cost-allocation studies that require the collection and analysis of the library cost-data. Cost accounting methods are used in most studies; however, costs are attributed to library user groups in a variety of ways. Cost accounting studies are reviewed and allocation methods are discussed. (Author)

The "Biglan model" has become a generally accepted new eponym (a term or phrase incorporating the originator's name) in the field of higher education. As eponyms become diffused through literature and citations disappear, the veracity of bibliometric studies of intellectual influence structures are undermined. (Author/MLW)

Discusses sources of available data on reference questions asked and answered in public libraries, and analyzes data collected between 1970 and 1985 in nine states. National data from libraries in the United States, Canada, and Great Britain are also presented. (Author/MES)

Comments on the uses of statistical analysis in social sciences within the context of Freedman's "As Other's See Us: A Case Study in Path Analysis." Points out that there is an unacknowledged gap between statistical principles and the data analysis that social scientists use. (RB)

Examines D. A. Freedman's criticism of path analysis, agreeing with Freedman's criticism of its application to nonexperimental data in the social sciences. Argues that Freedman's overall conclusions, however, are too pessimistic. (RB)

A step-by-step guide to conducting readership surveys by computer is presented. Planning a computerized survey, handling the responses, and analyzing the data are discussed. (MLW)

Although dental school faculty as well as practitioners are interested in evaluating products and procedures used in clinical practice, research design and statistical analysis can sometimes pose problems. Sequential trials methods provide an analytical structure that is both easy to use and statistically valid. (Author/MLW)

The advantages and disadvantages of balanced incomplete block designs are clarified and their use is demonstrated with an empirical example. A procedure for reducing data of this type to analyzable form is proposed, and an analytical approach that is appropriate for the resulting data is illustrated. (Author/BW)

The situational specificity hypothesis predicts that, if the setting does not vary, validity will not vary. Using data from a large-sample validity study numerous small-sample studies were generated for which the setting was held constant. Results indicate variability across studies in observed validity coefficients, significance levels, and…

The advantages of using a computer to analyze needs assessment data for continuing education are (1) it allows the expression of organizational needs, (2) all learners are able to declare their own needs, and (3) it provides rapid access to large amounts of information. (SK)

Four aspects of the relationship between auditing and evaluation in their approaches to program assessment are examined: (1) their different origins; (2) the definitions and purposes of both, and the questions they seek to answer; (3) contrasting viewpoints and emphases of auditors and evaluators; and (4) commonalities of interest and potential…

As complex, sophisticated techniques for statistical analysis become easier to obtain and apply, it becomes increasingly critical for researchers to examine their data prior to analysis to ensure that the data are suitable for analysis. Exploratory Data Analysis provides methods for subjecting data to appropriate pre-analysis scrutiny. (PGD)

The authors demonstrate how exponential smoothing can play a role in the identification of the intervention component of an interrupted time-series design model that is analogous to the role that the sample autocorrelation and partial autocorrelation functions serve in the identification of the noise portion of such a model. (Author/BW)

This paper demonstrates the variety of positive-skew phenomena and discusses their theoretical, research, and practical implications in education. (PN)

Factor and cluster analyses are distinctly different multivariate procedures with different goals. However, when used in a complementary fashion, each set of methods can be used to enhance the interpretation of results found in the other set of methods. Simple examples illustrating the joint use of the methods are provided. (Author)