Description of common errors found in the statistical methodologies of research carried out by librarians, focuses on sampling and generalizability. The discussion covers the need to either adapt library research to the statistical abilities of librarians or to educate librarians in the proper use of statistics. (15 references) (CLB)

Examines the usefulness of the Salience Imbalance Theory of metaphor processing to aid in understanding how children comprehend metaphors in written text. Supports the prediction that the measure of salience of features is affected by the topic order and vehicle (asymmetry), but also reports children's responses not explained by the theory. (KEH)

Describes the use of a simulated learning environment (SLE) as an instructional aid in teaching multivariate statistics, particularly inferential statistics. A prototype microcomputer-based SLE called MVWORLD developed at Drexel University for upper level statistics courses is explained, and implementing a statistics laboratory for multivariate…

The use of hierarchical models in social science research is discussed, with emphasis on causal inference and consideration of the limitations of hierarchical models. The increased use of Gibbs sampling and other Markov-chain Monte Carlo methods in the application of hierarchical models is recommended. (SLD)

A mental model analysis was conducted with 27 graduate and 29 undergraduate students in the statistical hypothesis testing domain to determine the nature of relevant mental models that can be taught to novices. Results suggest the efficacy of using separate and diagrammatic strategies in teaching novices introductory hypothesis testing. (SLD)

Aggressive and nonaggressive Latino and African-American adolescents read scenarios describing negative outcomes initiated by a hypothetical peer and made judgments about their reactions to the outcomes and about the peer's intentions. Data supported a model in which emotion mediates aggressive behavior. (BC)

The use of sampling weights in deriving population estimates for the National Assessment of Educational Progress (NAEP) and the effects of nonresponse and undercoverage on those estimates are described. The estimation of sampling variability from complex sample surveys is also reviewed, concentrating on the jackknife repeated replication…

Possible interpretive arguments for licensure and certification test scores are examined, and evidence to support each type of argument is analyzed. Particular attention is given to a competency-based argument that involves a sequence of inferences from test scores to statements about competence and then to conclusions about expected performance.…

Through Monte Carlo simulation, respective normative and intuitive strategies for covariation assessment and Bayesian inference are compared. Results indicate that better performance in both tasks results from considering alternative hypotheses, although not necessarily using a normative strategy. Conditions under which intuitive strategies may be…

Two political scientists summarize what they have learned about the role science plays in policymaking. They explain how "greenhouse policy" is being driven by inadequate notions of scientific consensus, the improper use of scenarios, and a suspect analogy to buying insurance. (Author/MCO)

This study found that word retrieval, phonological awareness, sentence completion, and narrative discourse processing skills of 93 reading-disabled students (ages 8-14) differed from skills of 93 normally achieving students. Performance on sentence completion and word retrieval measures best accounted for variance in younger reading-disabled…

Inference is considered for the marginal distribution of "X" when ("X", "Y") has a truncated bivariate normal distribution. The "Y" variable is truncated, but only the "X" values are observed. A sample of 87 Otis test scores is shown to be well described by this model. (SLD)

Examines whether or not learning unknown words in a context is useful to Japanese high school students. Particular focus is on the learning effect of unknown word inference strategies in reading.(Author/VWL)

A communicative-competence orientation was undertaken to study the validity of test-score inferences derived from the revised Test of Spoken English (TSE). To implement the approach, a sample of undergraduate students, primarily native-English speakers, provided reactions to the test responses of a sample of TSE examinees. (Author/VWL)

A comparison of the performance of children (ages 6-12) with visual impairments (n=15) and sighted children (n=15) on two tasks involving inferences found some differences between the two groups when the information was visual, but not when it was nonvisual. Visual impairment affected some aspects of a word association task. (Contains references.)…