Assuming vectors of item responses depend on ability through a fully specified item response model, this paper presents maximum likelihood equations for estimating the population parameters without estimating an ability parameter for each subject. Asymptotic standard errors, tests of fit, computing approximations, and details of four special cases…

How computers have affected diagnosis in mathematics is discussed. A distinction is made between errors and mistakes, followed by consideration of computers as a diagnostic tool. The development of algorithms to diagnose errors and of models of student performance is described. (MNS)

Analysis of errors children make is modeled both with and without computers. Patterns of thinking are traced for a fifth grader, with the discussion focused on getting clues for remedial instruction by analyzing the dialog. (MNS)

This study estimates the proportion of academic and public library acquisitions in the United Kingdom for which a UK MARC record is available at the time of cataloging. A beta-binomial model is used to attribute standard errors to each estimate and to compare differences between the two types of libraries. (Author/LRW)

Describes a new Boolean retrieval environment in which an extended soft Boolean logic is used to automatically construct queries from original natural language formulations provided by users. Experimental results that compare the retrieval effectiveness of this method to conventional Boolean and vector processing are discussed. (27 references)…

Discusses the motivation for integrating information retrieval and database management systems, and proposes a probabilistic retrieval model in which records in a file may be composed of attributes (formatted data items) and descriptors (content indicators). The details and resolutions of difficulties involved in integrating such systems are…

Presents suggestions for making statistical topics interesting and relevant in the classroom. For each of the topics, a general discussion is followed by a suggested teaching sequence with specific activities and examples. (PK)

National Assessment of Educational Progress reading data were scaled using a unidimensional item response theory model. Bock's full-information factor analysis and Rosenbaum's test of unidimensionality were applied. Conclusions about unidimensionality for balanced incomplete block spiralled data were the same as for complete data. (Author/GDC)

The estimation and evaluation of the effects of alternative educational or social policies is a major purpose of decision oriented, comparative evaluation studies. This paper reviews the rationale underlying the definition and interpretation of various measures of effect magnitude and examines their relevance to evalution research. (Author)

An alternative way of studying group differences is proposed based on ordering analysis using item hierarchies as a basis of comparison between two groups. Subjects were sets of twins in elementary school. Results showed that blacks and whites and males and females had similar item hierarchies for complex items. (Author/LMO)

A variety of tips about problem solving are included, with the focus on helping students recall an image. Manipulative materials and models using grids are included in most of the activities. (MNS)

Fourth-grade students' understanding of the order and equivalence of fractions was investigated in an 18-week teaching experiment. Three related characteristics of thinking are hypothesized to be related to success of such tasks, involving flexibility and reasoning that becomes increasingly independent of specific concrete embodiments. (MNS)

Explores alternative measures--Ph.D.s produced by the institution, number of faculty, number of interlibrary loans, number of publications attributable to the institution--to be used to test applicability to research libraries of the Cobb-Douglas production function, which relates three variables--production, labor, and capital investment.…

The application and use of hierarchical models is illustrated, using the example of the structure of attitudes toward a new product and a print advertisement. Subjects were college students who responded to seven-point bipolar scales. Hierarchical models were better than nonhierarchical models in conceptualizing attitude but not intention. (GDC)

The purpose of this study was to determise the relative loss in statistical power of traditional methods of analysis when response-shift bias is present. Five methods of analysis (posttest scores only; postminus pretests; postminus retrospective pretest; postcovaried by pretest; postcovaried by retrospective pretest) were compared. (Author/BW)