Studied the sample size requirements for accurate estimation of squared semi-partial correlation coefficients through simulation studies. Results show that the sample size necessary for adequate accuracy depends on: (1) the population squared multiple correlation coefficient (p squared); (2) the population increase in p squared; and (3) the…

Varied the number of items and categories per item to explore the effects on estimation of item parameters in the nominal response model. Simulation results show that increasing the number of items had little effect on item parameter recovery, but increasing the number of categories increased the error variance of the parameter estimates. (SLD)

Derived an approximate test and confidence interval for coefficient alpha and used the approximate test and confidence interval to derive closed-form sample size formulas that can be used to determine the sample size needed to test coefficient alpha with desired power or to test coefficient alpha with desired precision. (SLD)

The primary objective of this study was to investigate how incorporating prior information improves estimation of item parameters in two small samples. The factors that were investigated were sample size and the type of prior information. To investigate the accuracy with which item parameters in the Law School Admission Test (LSAT) are estimated,…

Investigated empirical Type I error rates and the power of omega (index of answer copying developed by J. Wollack, 1997) when item and trait (theta) parameters were unknown and estimated from datasets of 100 and 500 examinees. Type I error was unaffected by estimating item parameters, with power slightly lower for the smaller sample. (SLD)

Reviews interval estimates of the Pearson, Kendall tau-alpha, and Spearman correlates and proposes an improved standard error for the Spearman correlation. Examines the sample size required to yield a confidence interval having the desired width. Findings show accurate results from a two-stage approximation to the sample size. (SLD)

P. Rousseeuw and A. Leroy (1987) proposed a very robust alternative to classical estimates of mean vectors and covariance matrices, the Minimum Volume Ellipsoid (MVE). This paper describes the MVE technique and presents a BASIC program to implement it. The MVE is a "high breakdown" estimator, one that can cope with samples in which as…

Two general categories comprise the various effect size indices that have been proposed for use in meta-analysis: (1) the "d"-type estimator (based on magnitude of mean difference); and (2) the "r"-type estimator (based on magnitude of correlation). In meta-analyses, researchers often must convert these effect size indices to a common metric to…

Simple noniterative estimators of the polyserial correlation coefficient are developed by exploiting a general relationship between the polyserial correlation and the point polyserial correlation to give extensions of the biserial estimators of K. Pearson (1909), H. E. Brogden (1949), and F. M. Lord (1963) to the multicategory setting. (SLD)

Studied the use of different weighting techniques in structural equation modeling and found, through simulation, that the use of an effective sample size weight provides unbiased estimates of key parameters and their sampling variances. Also discusses use of a popular normalization technique of scaling weights. (SLD)

A multiobjective programming method is proposed for determining samples of examinees needed for estimating the parameters of a group of items. This approach maximizes the information functions of each of three parameters. A numerical verification of the procedure is presented. (SLD)

Discusses determining sample size for estimation of the squared multiple correlation coefficient and presents regression equations that permit determination of the sample size for estimating this parameter for up to 20 predictor variables. (SLD)

Shows that the conventional noncentrality parameter estimator of covariance structure models, currently implemented in popular structural modeling programs, possesses asymptotically potentially large bias, variance, and mean squared error (MSE). Presents a formal expression for its large-sample bias and quantifies large-sample bias and MSE. (SLD)

Evaluated the bootstrap method under varying conditions of nonnormality, sample size, model specification, and number of bootstrap samples drawn from the resampling space. Results for the bootstrap suggest the resampling-based method may be conservative in its control over model rejections, thus having an impact on the statistical power associated…

Whether "more is ever too much" for the number of indicators (p) per factor (p/f) in confirmatory factor analysis (CFA) was studied by varying sample size (N) from 50 to 1,000 and p/f from 2 to 12 items per factor in 30,000 Monte Carlo simulations. For all sample sizes, solution behavior steadily improved (more proper solutions and more…