Studied the robustness of the sign test frequently used to examine group differences as measured by multiple indicators where observations are correlated. Findings of a simulation study show that modest intercorrelations only influence the ability to detect Type I errors to a limited degree, but high intercorrelations make use of the sign test…

The author provides a review of methods for comparing medians, with an emphasis on comparing two groups. A fairly broad perspective is provided with a summary of the relative merits associated with different strategies. Included as a special case are well-known rank-based techniques. In addition, the author reports new results on comparing the…

When governments and pressure groups attend only to research that suits their political purposes, it makes a mockery of the idea of evidence-informed policy and practice. In this article, the author cites examples of research that picked up weak and misleading evidence for political purposes. The author argues that researchers are tied in to the…

Many test-based accountability systems, including the No Child Left Behind Act of 2001 (NCLB), place great weight on the numbers of students who score at or above specified proficiency levels in various subjects. Accountability systems based on these metrics often provide incentives for teachers and principals to target children near current…

Quality measurement is essential in every form of research, including institutional research and assessment. Unfortunately, most survey research today (both published and unpublished) is lacking with regards to quality measurement. Reporting means and standard deviations based on ordinal measures is an inappropriate, yet widespread practice in the…

In this study, psychometric properties of the test of the three-mathematical minds (M3) were investigated. The M3 test was developed based on a multidimensional conception of giftedness to identify mathematically talented students. Participants included 291 middle-school students. Data analysis indicated that the M3 had a 0.73 coefficient as a…

Standard least squares analysis of variance methods suffer from poor power under arbitrarily small departures from normality and fail to control the probability of a Type I error when standard assumptions are violated. This article describes a framework for robust estimation and testing that uses trimmed means with an approximate degrees of…

This article focuses on the accuracy of satellite data, which may then be used in wave power applications. The satellite data are compared to data from wave buoys, which are currently considered to be the most accurate of the devices available for measuring wave characteristics. This article presents an analysis of satellite- (Topex/Poseidon) and…

Reliability Analysis for the Internationally Administered 2002 Series GED (General Educational Development) Tests Reliability refers to the consistency, or stability, of test scores when the authors administer the measurement procedure repeatedly to groups of examinees (American Educational Research Association [AERA], American Psychological…

Peer review is valued in higher education, but also widely criticized in terms of potential biases, particularly gender. We evaluate gender differences in peer reviews of grant applications, extending Bornmann, Mutz, and Daniel's meta-analyses that reported small gender differences in favor of men (d = 0.04), but a substantial heterogeneity in…

Instability in the school population between school entrance and school leaving is not "just a problem of missing data" but often the visible result of the educational problems in some schools and is, therefore, not merely to be treated as missing data but as indicator for the quality of educational processes. Even the most superior…

Three approaches to the analysis of main and interaction effect hypotheses in nonorthogonal designs were compared in a 2 x 2 design for data that was neither normal in form nor equal in variance. The Welch-James test with trimmed means and Winsorized variances provided excellent Type I error control. (SLD)

This study examines the effect of supplemental educational services (SES) on student test score gains and whether particular subgroups of students benefit more from No Child Left Behind (NCLB) tutoring services. The sample used includes information on students enrolled in 3rd through 8th grades nested in 121 elementary and middle schools over a…

Recent years have seen increasing interest in assessing the prestige and productivity of programs in communication. An NCA doctoral reputational study was published in 2004, and a new National Research Council assessment is underway that will for the first time include the communication field. Two systems relying on objective appraisal of faculty…

The Centre for the Economics of Education was asked to investigate the factors that influence a range of children's academic and non-academic outcomes, including their enjoyment of school, whether they take unauthorised absence from school and whether they feel they are bullied. The study also investigated whether schools can influence these…