Creativity tests, like creativity itself, vary widely in their structure and use. These differences include instructions, test duration, environments, prompt and response modalities, and the structure of test items. A key factor is task structure, referring to the specificity of the number of responses requested for a given prompt. Classic…

Drawing on Glenn Fulcher's extensive work in performance-based language assessment of speaking, this paper explores the assessment of L2 speaking ability in local language testing contexts. For that purpose, I review Fulcher's influential work that highlights the relationship between the speaking construct, the task, the performance, and the…

Analyses were conducted to determine whether the two statistical tests of significance are identical. Such demonstrations regarding parallel correlational and analysis of variance procedures have brought further understanding to each of the domains in the past. When a researcher is investigating the relationship between two variables and thinks…

Multivariate normality is required for some statistical tests. This paper explores the implications of violating the assumption of multivariate normality and illustrates a graphical procedure for evaluating multivariate normality. The logic for using the multivariate bootstrap is presented. The multivariate bootstrap can be used when distribution…

It has often been asserted that the reliability of a measure places an upper limit on its validity. This article demonstrates in theory that validity can rise when reliability declines, even when validity evidence is a correlation with an acceptable criterion. Whether empirical examples can actually be found is an open question. (SLD)

This article presents arguments that the correlation measure "r" of halo is not conceptually more appropriate than the standard deviation (SD) measure. It also describes conditions under which halo effects occur and when the SD and r measures can be used. Neither measure is uniformly superior to the other. (SLD)

The development of the Survey of Attitudes Toward Statistics is described, and the results from several validity analyses performed with 1,203 college students are presented. Additional validity evidence was obtained through correlation with the Attitudes Toward Statistics Scale of S. L. Wise for a subsample of 221 students. (SLD)

Multivariate permutation tests are described, and some are suggested as substitutions for Hotelling's one-sample T2 test in common situations in behavioral science research. A Monte Carlo study shows advantages of these tests when the T2 test fails or is suspect. (SLD)

An important issue in educational and employment settings is the degree to which evidence of validity obtained in one situation can be generalized to another situation without further study of validity in the new situation. Theory, procedures, and applications concerning validity generalization are addressed. Meta-analytic techniques make possible…

As schools move toward performance assessment, there is increasing discussion of using these assessments for accountability purposes. When used for making decisions, performance assessments must meet high standards of validity and reliability. One major source of unreliability in performance assessments is interrater disagreement. In this paper,…

This essay explores what is known and what remains to be learned about the information value of performance items when they are used in large scale assessment. Within the context of the National Assessment of Educational Progress (NAEP), there is substantial interest in answering these questions as issues of national assessment are considered for…

Comparability of the Minnesota Multiphasic Personality Inventory (MMPI) and the MMPI-2 was explored by examining T-score means, profile configurations, score distribution, and rank-order correlations on validity scales for 84 undergraduates. Equivalency of the two forms was generally supported. (SLD)

A crucial, but often forgotten, role of educational assessment is to enhance students' learning. This author advocates that an assessment program designed for student learning differs from assessment for accountability in purpose, test format, measurement type, number, and spread of tests, use of test results, and amount of interval between…

Research evidence relating to the utility of the RMC evaluation models of compensatory education employing non-normed tests is examined. The history and evolution of five early models into the current norm-referenced model utilizing a non-normed test (Model A2), non-normed versions of a comparison model (Model B2), and the regression model (Model…

Changes in the predictive validity of the Scholastic Aptitude Test (SAT) that have been observed in the University System of Georgia (USGA) over the past two decades were studied. Several analyses were performed using student data from 27 USGA schools to determine the effectiveness of the SAT in predicting freshman grades. National trends…