A Bayesian Model II approach to the estimation of proportions in m groups is extended to obtain posterior marginal distributions for the proportions. The approach is extended to allow greater use of prior information than previously and the specification of this prior information is discussed. (Author/RC)

Six in-general samples have been generated to fit different developments of the SVIB, Twenty experimental homogeneous scales were used to measure the similarities and differences among the six in-general samples. Generally, all samples were strikingly similar with the greatest differences appearing between male and female in-general samples.…

Four different extensions to McNemar's problem concerning the hypothesis of equal probabilities for the unlike pairs of correlated binary variables are considered, each for testing simultaneous equality of proportions of unlike pairs in c independent populations of correlated binary variables, but each under different assumptions and/or additional…

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)

Major interest is the derivation of a mathematical formulation and a demonstration of the effects of pooling two subgroups on the total group correlation coefficient, and additionally, a procedure to guide the decision concerning the pooling of data for the purpose of calculating a single correlation coefficient. (Author/RC)

A method of applying the discriminant function to correlated samples was discussed for the researcher interested in utilizing multivariate statistics to pre-post data. By finding the linear combination that maximizes differences in the groups, the t-statistic can be computed for correlated samples. (Author)

The problems addressed, and solutions devised, by educational ethnographers in selection of and sampling from phenomena for internal representativeness external comparison are examined. The process of sampling is designated as a specialized form of the more general process of focusing and choosing in research, which is called selection. Common…

This report documents a methodological study conducted for the National Science Foundation to investigate the effects of different media environments on relationships between local television news viewing and local knowledge. The study serves two purposes: it establishes tne reliability of data collected in an investigation by Rand of the effects…

To explore the feasibility of sample selection to predict agricultural employment, a stratified sample of 360 employers was drawn from the employers reporting employment in the Disability Insurance Program for agricultural workers in Fresno County. Month-to-month changes were examined for seven industry strata, and estimates were made by the…

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…

The power of a statistical test is, in part, a function of the reliability of the dependable variable being analyzed. The substitution of sigma square divided by the reliability coefficient for sigma is proposed. This enables the researcher to incorporate dependent variable reliability information when determining the sample size required for a…

Significant cost savings can be achieved with the use of matrix sampling in estimating population parameters from psychometric data. The statistical design is intuitively simple, using the framework of the two-way classification analysis of variance technique. For example, the mean and variance are derived from the performance of a certain grade…

The primary purpose of the present study was to investigate empirically the effect of multiple hypothesis testing on error rates in factorial ANOVA designs under a variety of controlled conditions. The per comparison, per experiment, and experimentwise error rates were investigated for three hypothesis testing procedures. The specific conditions…