In social science research there are a number of instruments that use a rating scale such as a Likert response scale. For a number of reasons, a respondent's response vector may not contain responses to each item. This study investigated the effect on a respondent's location estimate when a respondent is presented an item, has ample time to answer…

A Monte Carlo simulation approach was taken to investigate the effect of sample size on a variety of latent growth models. A fully balanced experimental design was implemented, with samples drawn from multivariate normal populations specified to represent 12 unique growth models. The models varied factorially by crossing number of time points,…

The purpose of this study was to determine: (1) the extent to which effect sizes vary by chance; (2) the proportion of standardized effect sizes that achieve or exceed commonly used criteria for small, medium, and large effect sizes; (3) whether standardized effect sizes are random or systematic across numbers of groups and sample sizes; and (4)…

Nonparametric procedures are often more powerful than classical tests for real world data, which are rarely normally distributed. However, there are difficulties in using these tests. Computational formulas are scattered throughout the literature, and there is a lack of availability of tables of critical values. This paper brings together the…

Real world data often fail to meet the underlying assumptions of normal statistical theory. Many statistical procedures in the psychological and educational sciences involve models that may include a system of statistical equations with non-normal correlated variables (e.g., factor analysis, structural equation modeling, or other complex…

Four of the most commonly used multiple comparison procedures were compared for pairwise comparisons and relative to control of per-experiment and experimentwise Type I errors when conducted as protected or unprotected tests. The methods are: (1) Dunn-Bonferroni; (2) Dunn-Sidak; (3) Holm's sequentially rejective; and (4) Tukey's honestly…

A method of introducing a controlled degree of skew and kurtosis for Monte Carlo studies was derived. The form of such a transformation on normal deviates is given. Analytic and empirical validation of the method is demonstrated. (Author/JKS)

Stochastic approximation is suggested as a useful technique in areas where individuals have a goal firmly in mind, but lack sufficient knowledge to design an efficient, more traditional experiment. One potential area of application for stochastic approximation is that of formative evaluation. (CTM)

Used Monte Carlo simulation to compare the properties of a goodness-of-fit (GOF) procedure and a test statistic developed by E. Bartfay and A. Donner (2001) to the likelihood ratio test in assessing the existence of extra variation. Results show the GOF procedure possess satisfactory Type I error rate and power. (SLD)

A Monte Carlo simulation was conducted to study the Type I error rate and power of the 1994 approximation developed by R. A. Alexander and D. M. Govern as an alternative to the analysis of variance "F" test. Conditions under which this test is the best approach are discussed. (SLD)

A Monte Carlo study compared the accuracy of different estimates of the standard error of correlations corrected for restriction in range. The procedure suggested by P. Bobko and A. Rieck (1980) generated the most accurate estimates of the standard error. Aspects of accuracy are discussed. (SLD)

Monte Carlo study results show that the runs test yields markedly asymmetrical error rates in the two tails and that neither directional nor nondirectional tests are satisfactory with respect to Type I errors. The test is not recommended for evaluating the independence of errors in time-series regression models. (SLD)

Compared two parametric procedures for detecting differential item functioning (DIF) using the graded response model (GRM), the GRM-likelihood ratio test and the GRM-differential functioning of items and tests, with a nonparametric DIF detection procedure, Poly-SIBTEST. Monte Carlo simulation results show that Poly-SIBTEST showed the least amount…

Estimated statistical power for three randomization tests used with multiple-baseline designs using Monte Carlo methods. For an effect size of 0.5, none of the tests provided an adequate level of power, and for an effect size of 1.0, power was adequate for the Koehler-Levin test and the Marascuilo-Busk test only when the series length was long and…