The use of the arc-sine transformation in analysis of variance can lead to difficult inference situations and pose problems in interpretation. It can also produce tests of noticeably lower power when the null hypothesis is false, and is not recommended as a standard tool. Simulated illustrations are provided. (Author/GDC)

A student's understanding of a problem before he or she solves it is a critical component of successful problem solving. This understanding is based on how a problem is represented; that is, whether a problem is understood in terms of principles or solution methods or whether the focus is on features that are irrelevant to its solution. In this…

Bootstrap analysis, both for nonparametric statistical inference and for describing sample results stability and replicability, has been gaining prominence among quantitative researchers in educational and psychological research. Procedurally, however, it is often quite a challenge for quantitative researchers to implement bootstrap analysis in…

Examines potential problems in the use of inferential statistics for single population proportions, differences between two population proportions, and quotients of two population proportions. Discusses hypothesis testing versus interval estimation. Emphasizes the importance of selecting the appropriate formula for the standard error and…

Distinguishes two conceptions of sample and sampling that emerged in the context of a teaching experiment conducted in a high school statistics class. Suggests that the conception of a sample as a quasi- proportional, small-scale version of the population is a powerful one to target for instruction. (Author/KHR)

Presents and discusses examples that illustrate the nature and scope of elementary and middle school students' reasoning when they are faced with tasks that involve making inferences and predictions from data. Shows that the range in thinking is not so much dependent on age as on the experiences students have in data exploration. (KHR)

Discusses three topics related to interpretation of discriminant analyses (DA's): (1) partial F ratios and partial Wilks's lambdas for predictor variables in standard, step-down, and stepwise DA's; (2) relation of goals of classification to definition/evaluation of classification rules; and (3) significance tests for total hit rates in internal…

Investigates student interpretation of research by asking samples of undergraduates, graduates, and faculty questions concerning the implications of random sampling and random assignment. Finds that random sampling is understood while the role of random assignment in interpretation of results is misunderstood. Concludes there is a need for…

Studies binomial, negative binomial, and gamma regression models and gives a detailed description of inference procedures based on them. The process of model fitting and evaluation is illustrated by examples referring to the determination of endpoints in acute and chronic toxicity tests. (17 references) (Author/MDH)

Makes the case that science is best served when researchers focus on the size of effects and their practical significance. Advocates the use of confidence intervals for deciding whether chance or sampling variability is an unlikely explanation for an observed effect. Calls for more emphasis on effect sizes in the next edition of the American…

Comparison of 88 business students who completed experiential projects involving data collection and inferential analysis with 125 who received lectures only indicated that the active learning method resulted in better understanding of statistics through the application of theory to real-life situations. (SK)

Investigates significance tests as a tool for helping to identify real effects in the face of random variation. Speculates that equivalence tests improve the logic of significance testing when demonstrating similarity is important. Reverse tests can help show that failure to reject a null hypothesis does not support that hypothesis. (Contains 30…

Two methods, direct maximum likelihood (ML) and the expectation maximization (EM) algorithm, can be used to obtain ML parameter estimates for structural equation models with missing data (MD). Although the 2 methods frequently produce identical parameter estimates, it may be easier to satisfy missing at random assumptions using EM. However, no…

Wider use in psychology of confidence intervals (CIs), especially as error bars in figures, is a desirable development. However, psychologists seldom use CIs and may not understand them well. The authors discuss the interpretation of figures with error bars and analyze the relationship between CIs and statistical significance testing. They propose…

In this article, the authors (a) illustrate how displaying disattenuated correlation coefficients alongside their unadjusted counterparts will allow researchers to assess the impact of unreliability on bivariate relationships and (b) demonstrate how a proposed new "what if reliability" analysis can complement null hypothesis significance…