Compares the practice of assessing prior learning in the United States and Britain and reviews major issues: nature of the evidence, collection of evidence into portfolios, evaluation, and costs. (SK)

Two algorithms that automatically select subsets of variables (PACE algorithm) and reference variables (Fabin estimators), respectively, used for the noniterative estimators are presented. The PACE algorithm is based on a nonsymmetric matrix sweep operator. A Monte Carlo experiment compares the relative performance of these estimators and others.…

This study found that 47 mildly handicapped third graders in a full-time integrated program, Team Approach to Mastery (TAM), experienced greater achievement gains than 31 students in resource rooms, though differences were significant only in math. Nonhandicapped TAM students (n=176) made greater gains than mildly handicapped TAM students and…

Mothers (n=104) of children (mean age 10-11) receiving psychological evaluation completed a depression inventory. At least half of the mothers of children who were victims of intrafamilial abuse, extrafamilial abuse mothers, and mothers of nonabused children reported experiencing at least moderate depression. (Author/JDD)

A total of 93 elementary, middle school, and high school teachers rated the desirability and feasibility of adaptations for mainstreamed students in their general education classes. All adaptations were perceived as more desirable than feasible, and few differences were found among teachers of different grade levels. (Author/JDD)

Simulation studies (N=4,000 simulees) examined the effect of misspecification of the latent ability distribution (theta) on the accuracy and efficiency of marginal maximum likelihood (MML) item parameter estimates and on MML statistics to test sufficiency and conditional independence. Results were compared to those of the conditional maximum…

Two recent indices of fit, the Relative Noncentrality Index (RNI) (R. P. McDonald and H. W. Marsh, 1990) and the Comparative Fit Index (P. M. Bentler, 1990), are shown to be algebraically equivalent in most applications, although one condition in which the RNI may be advantageous for model comparison is identified. (SLD)

Discussion of structural knowledge assessment for large groups focuses on the features of a "KNOT" computer program which uses a network similarity index (NETSIM) to compare novice and expert concept maps. Treatment in three high school physics classes investigated reliability as well as content, construct, concurrent, and predictive…

Describes an experiment that evaluated features of TEXNET abstracting software, compared the use of keywords and phrases that were automatically extracted, tested hypotheses about relations between abstractors' backgrounds and their reactions to abstracting assistance software, and obtained ideas for further features to be developed in TEXNET.…

Reviews the literature on Web search engines evaluation, with the aim of identifying some of the issues and problems in this rapidly changing field. Suggests that a standardized set of tools be developed for the evaluation of search engines so comparisons can be made between them more effectively and variations in performance of any given search…

A study that compared two instructional strategies for teaching reading comprehension to fourth-grade students found no differences between a peer-assisted procedure involving 25 students working in dyads to ask and answer questions and a teacher-assisted procedure involving 23 students working alone to develop questions followed by class…

Colleges and universities should critically review their publications periodically to save money and also as a way to present a more cohesive institutional identity. Ten recommendations are made for conducting a productive publications audit; they include choosing an evaluation method, selecting participants, making the experience pleasant, and…

Discusses the advantages and disadvantages of what has become the standard approach to teaching relative clauses. To overcome the disadvantages, proposes a hierarchical teaching method that begins by introducing noun phrases that contain relative clauses before comparing them with similar structures in the learner's first language. Shows an…

Compared reliability of three widely used child protective service risk-assessment models (one actuarial, two consensus based). Found that, although no system approached 100% interrater reliability, raters employing the actuarial model made consistent estimates of risk for a high percentage of cases they assessed. Interrater reliability for the…

Addresses the problem of benchmarking image browsers. Existence of different search paradigms for image browsers makes it difficult to compare them. Currently, the only admissible evaluation method involves conducting large-scale user studies. An automatic image browser benchmark is proposed that uses structured text annotation of the image…