The present article reviews the procedures that have been developed for measuring the reliability of human observers' judgments when making direct observations of behavior. These include the percentage of agreement, Cohen's Kappa, phi, and univariate and multivariate agreement measures that are based on quasi-equiprobability and quasi-independence…

Problems in the measurement of latent variables (LVs) in structural equations causal models are presented, with examples from recent empirical studies. Main issues addressed are (1) two general types of increased operative variables; (2) criteria for indicators of LVs; and (3) disattenuation effects in LV models. (SLD)

A recent recommendation by Holt (EJ 200 576) that coefficients resulting from estimating log-linear and similar models should not be interpreted is argued to be based on lack of clarity about the substantive and theoretical importance of the choice between dummy and effect coding for categorical variables. (Author/GDC)

First- and second-round commentaries on an article by L. Guttman are presented. The following authors responded, with two articles each: (1) E. E. Roskam and J. Ellis; (2) P. H. Schonemann; (3) A. R. Jensen; (4) J. C. Loehlin; and (5) J.-E. Gustafsson. (SLD)

Criticizes methodological tendencies of economics and sociological researchers who do not ask "how large is large?" Reliance on statistical significance and existence theorems shows the limits of formalistic rather than humanistic research approaches. Argues that serious scholars develop whole arguments and understand the rhetoric of…

Argues that Jensen's article contains an inaccurate and misleading account of Spearman's work and distorts the basic concepts of factor analysis. The target article has failed in all its main objectives; its major failing is a result of the irrelevance of factor analysis to the study of group differences. (SLD)

Computer generated data are used to show that Slinde and Linn's criticism of the usefulness of the Rasch model for equating (EJ 189 585) may have been the result of an artifact produced by the manner in which the samples were chosen in their study. (CTM)

Analyzes effect sizes in tails of distribution of scores in Feingold's study of joint effects of gender differences in mean and variability on 28 cognitive-ability scales. Effect sizes are smaller than Feingold assumed. Evaluates joint effect of gender differences by number of males and females in extreme score ranges. (SLD)