Beliefs about memories and about the ability to recall memories may affect the individual's recollection of facts and "events." What people believe about memory is investigated using previously analyzed and reported responses to the Beliefs about Memory Survey (BMS). The survey sample (N=1046) was randomly divided for calibration and…

The development and psychometric properties of the Mathematics Anxiety-Apprehension Survey (MAAS), a measure of instructional and/or classroom mathematics phobia are discussed. This instrument was designed to assess levels of mathematics anxiety and apprehension in students with varying levels of education and socio-demographic factors. The highly…

It is the purpose of this paper to suggest the orthogonal analysis of variance as a device for simplifying either the analytic or iterative problem of finding LS (least squares) estimates for the parameters of particular nonlinear models. (Author/RK)

The scales of the EPI have both a favorable and an unfavorable psychometric property which is related to the instructions and format of the inventory. (Author)

A consideration of the factors which influence development and development predictions leads to the view that, with better instruments, improved but still very modest predictions might become possible during later infancy. Paper read to the Standing Conference of Societies Registered for Adoption: Medical Group: Day Conference on Genetic and…

Rasch's psychometric model is presented as a special case of additive conjoint measurement. Two data sets, each of which was analyzed using Rasch scaling methods as well as some of the methods of conjoint measurement, are compared. (Author/JKS)

Modern statistics are considered as a branch of psychometrics and the question of how the central problems of statistics can be resolved using psychometric methods is investigated. Methods developed in the fields of test theory, scaling, and factor analysis are related to principle problems of modern statistical theory and method. (Author/JKS)

Two lower bounds to reliability in classical test theory, Guttman's lamda and Cronbach's alpha, are shown to be terms of an infinite series of lower bounds. All terms of this series are equal to reliability if and only if the test contains items which are tau-equivalent. (Author/JKS)