Compared the analysis of variance (ANOVA) "F" and several nonparametric competitors for two-way designs for empirical alpha and power through simulation. Results suggest the ANOVA "F" suffers from conservative alpha and power for the mixed normal distribution, but is generally recommended. (Author/SLD)

Some results on how the Alexander-Govern heteroscedastic analysis of variance (ANOVA) procedure (R. Alexander and D. Govern, 1994) performs under nonnormality are presented. This method can provide poor control of Type I errors in some cases, and in some situations power decreases as differences among the means get large. (SLD)

Illustrates the computation of confidence intervals for effect sizes for some analysis of variance applications and shows how the use of intervals involving noncentral distributions is made practical by new software. (SLD)

A generalization of the usual random-effects model based on trimmed means is proposed. The resulting test of no differences among J randomly sampled groups has advantages in terms of Type I errors and can yield gains in power when distributions have heavy tails and outliers. (SLD)

Considers a problem of simple random walks to study distributions of variances. Describes watching a drunk over a period of nights, taking a number of steps per night. Explores the full probability distribution for the variance of the data string and discusses the connection of the results to the problem of data binning. (Author/NB)

Formulae and graphs are presented allowing computation of the variances of three prototypical distributions over a finite number of categories. The uses of the variances of the maximum variance distribution, the uniform distribution and a unimodal triangular distribution to make inferences about distribution shapes are shown in several examples.…

The problem of determining which of several populations has the largest mean is considered. The procedure described by Dudewicz and Dalal is extended to the case of unequal sample sizes. (JKS)

This study examined the sampling distribution of the analysis of variance F ratio in the two sample cases when it followed a preliminary test for variance equality. When the population variances were equal, the sampling distribution approximated the theoretical F distribution quite well, but not when population variances differed. (JAZ)

The interpretation of significant left-tailed analysis of variance (ANOVA) F-ratios is supported by considering the case of a fixed effects ANOVA model. The conclusions of this case are generalizable to other standard ANOVA models. (Author/PN)

Describes a technique for statistically comparing bibliometric models and illustrates its use with two examples using Lotka's hypothesis of author productivity and one example using library circulation frequencies. Topics discussed include nested statistical models, analysis of variance, regression, log-linear models, and the likelihood ratio…

Two statistical measures are used to compare the Library of Congress and Dewey Decimal Classification systems by applying them to the library science portions of the two systems using a database drawn from MARC records. It is concluded that the measures characterize two different properties of dispersion. (Author/EM)

This paper presents 11 statistical procedures which test the equality of m coefficient alphas when the sample alpha coefficients are dependent. Several of the procedures are derived in detail, and numerical examples are given for two. (Author/LMO)

Describes the development and testing of the Bath County Computer Attitudes Scale (BCCAS), which is designed to measure elementary and secondary school students' attitudes toward computers. Preliminary and revised versions are reviewed, and a analysis of data supports the validity of the BCCAS as a measure of computer attitudes. (Author/LRW)

A cubic spline method for smoothing equipercentile equating relationships under the common item nonequivalent populations design is described. Statistical techniques based on bootstrap estimation are presented for choosing an equating method/degree of smoothing. Smoothing decreases the estimate of random error but results in an increase in…

This second report on a two-part study on the application of informetrics to information retrieval (IR) system design used models of database contents in a factorial simulation study at the University of Western Ontario. The study explored whether different file structures were better suited for different informetric environments. (32 references)…