Sample size determinations for the two rater kappa statistic

This paper gives a method for determining a sample size that will achieve a prespecified bound on confidence interval width for the interrater agreement measure,. The same results can be used when a prespecified power is desired for testing hypotheses about the value of kappa. An example from the literature is used to illustrate the methods proposed here.     

Generalized approaches to the maxbet problem and the maxdiff problem,with applications to canonical correlations

Van de Geer has reviewed various criteria for transforming two or more matrices to maximal agreement, subject to orthogonality constraints. The criteria have applications in the context of matching factor or configuration matrices and in the context of canonical correlation analysis for two or more matrices. The present paper summarizes and gives a unified treatment of fully general computational solutions for two of these criteria, Maxbet and Maxdiff. These solutions will be shown to encompass various well-known methods as special cases. It will be argued that the Maxdiff solution should be preferred to the Maxbet solution whenever the two criteria coincide. Horst's Maxcor method will be shown to lack the property of monotone convergence. Finally, simultaneous and successive versions of the Maxbet and Maxdiff solutions will be treated as special cases of a fully flexible approach where the columns of the rotation matrices are obtained in successive blocks.The author is obliged to Henk Kiers for computational assistance and helpful comments.     

Stress in time

The goals of this research were to determine whether speakers adjust the stress patterns of words within sentences to create an alternation between strong and weak beats and to explore whether this rhythmic alternation contributes to the characteristics stress differences between two major lexical categories of English. Two experiments suggested that speakers do alter lexical stress in accordance with rhythmic biases. When speakers produced disyllabic pseudowords in sentence contexts, they were more likely to place stress on the first syllable when the pseudoword was preceded by a weak stress and followed by a strong one than when the strong stress preceded and the weak followed. This occurred both when the pseudowords served as nouns and when they served as verbs. Text analyses further revealed that weakly stressed elements precede nouns more often than verbs, whereas such elements follow verbs more often than nouns. Thus, disyllabic nouns are more likely than disyllabic verbs to occupy contexts biased toward trochaic rhythm, a finding consistent with leftward dominant stress in disyllabic English nouns. The history of stress changes in English nouns and verbs also conforms with the view that rhythmic context may have contributed to the evolution of stress differences. Together, the findings suggest that the citation stress patterns of words may to some degree reflect adaptations of lexical knowledge to conditions of language performance.     

Learned helplessness in the cockroach (Periplaneta americana)

For 3 consecutive days cockroaches (Periplaneta americana) were exposed to either escapable, inescapable, or no shock in an escape task. Twenty-four hours later they were tested in a shuttlebox escape task. There were reliable differences between escapable and inescapable animals and between inescapable and control animals in both escape latencies and the number of failures to escape.     

Is it smart to play the horses? Comment on "A day at the races: a study of IQ, expertise, and cognitive complexity" (Ceci & Liker, 1986)

Ceci and Liker (1986b) presented data that they contended shows two things: (a) Handicapping harness races is a cognitively complex undertaking that can be captured by a multiple regression model, and (b) neither overall skill at handicapping nor the complexity of the mental model used is related to standard measures of intelligence. The first contention is not at issue. But the second contention, that handicapping performance is unrelated to IQ, is not supported by the data presented; in fact, the opposite conclusion seems more likely. The purpose of this comment is to point out errors frequently made in individual-differences research concerning population definition, sample selection, dependent and independent variable reliability and validity and interpretation of results.