Generalizations of the Scheffe, Tukey, and Bonferroni-t techniques are presented, each of which controls the experimentwise error rate for a particular type of partially or fully planned analysis. All three procedures provide more power than multivariate analysis of variance (MANOVA) tests. (Author/BJG)

Games (1973) criticized the Marascuilo and Levin (1970) "nested" or simple effects design as a poor substitute for the usual logic of a factorial analysis of variance. Marascuilo and Levin's use of "nesting" is completely idiosyncratic and contrary to well-established usage, thus confusing the student. Proper usage is…

The central messages of this paper are that (a) unequal variances may be more prevalent than typically imagined in educational and policy research, and (b) when considering tests of equal variances one needs to be cautious about what is being referred to as "Levene's test" because Levene's test is actually a family of techniques. Depending on…

This study investigated the performance of native and non native speakers of Modern Greek language on morphology and syntax tasks. Non-native speakers of Greek whose native language was English, which is a language with strict word order and simple morphology, made more errors and answered more slowly than native speakers on morphology but not…

Generalised access to the Internet and globalisation has led to increased demand for translation services and a resurgence in the use of machine translation (MT) systems. MT post-editing or the correction of MT output to an acceptable standard is known to be one of the ways to face the huge demand on multilingual communication. Given that the use…

The probability of experiment-wise error is explored. Overall, the experiment-wise error rate is directly related to the test-wise error rate--the alpha level set by researchers to curtail the existence of a Type I error. A Type I error occurs when a true null hypothesis is rejected in a given study or experiment. The experiment-wise error rate is…

Monte Carlo methods were used to investigate the robustness of techniques used in judging the magnitude of a sample correlation coefficient when observations are correlated. Empirical distributions of r, t, and Fisher's z were generated. A technique for controlling error rates in certain situations is suggested. (Author/BJG)

Investigated empirically through post mortem item-examinee sampling was the feasibility of the jackknife as a procedure for approximating standard errors of estimate in multiple matrix sampling. The parameters estimated were the mean test score, second through fourth central moments of the test score distribution, and the variance of the item…

When Torgerson's multidimensional scaling model is used in conjunction with the method of tetrads, derived coordinates are based on data which is assumed to be distributed normally. The object of this study was to determine the amount of error contained in derived coordinates when the normality assumption is violated. Torgerson coordinates were…

Selected parameters for a negatively-skewed and a normally distributed normative distribution were estimated in a post-mortem item-examinee sampling investigation. Manipulated systematically were number of subtests, number of items per subtest, and number of examinees responding to each subtest. Each item-examinee sampling procedure was replicated…

Through simulation, the power of the U3 statistic was compared with the power of one of the simplest person-fit statistics, the sum of the number of Guttman errors. In most cases, a weighted version of the latter statistic performed as well as the U3 statistic. (SLD)

It is well known that the two-sample Student t test fails to maintain its significance level when the variances of treatment groups are unequal, and, at the same time, sample sizes are unequal. However, introductory textbooks in psychology and education often maintain that the test is robust to variance heterogeneity when sample sizes are equal.…

One of six summaries of workshop sessions (See TM 000 130), designed to strengthen the evaluation of costly programs and their effects, this handbook presents an analysis of both random and nonrandom sampling errors by application of the Bayesian model. This model attempts to formalize the process and procedures of inference from data through…

Analyzes accuracy of Library of Congress cataloging for items with Cataloging in Publication records and items not in that program. Types and frequencies of errors are observed with a separate analysis of significant errors. Recommendations are made for copy cataloging based on the different types of errors found. (Author/EM)

A listing of the program is available from the author at Northwestern University. (MB)