A general model for the analysis of multilevel data

A general model is developed for the analysis of multivariate multilevel data structures. Special cases of the model include repeated measures designs, multiple matrix samples, multilevel latent variable models, multiple time series, and variance and covariance component models.We would like to acknowledge the helpful comments of Ruth Silver. We also wish to thank the referees for helping to clarify the paper. This work was partly carried out with research funds provided by the Economic and Social Research Council (U.K.).     

Reconsideration of measurement of error in human motor learning

Human motor learning is often measured by error scores. The convention of using mean absolute error, mean constant error, and variable error shows lack of desirable parsimony and interpretability. This paper provides the background of error measurement and states criticisms of conventional methodology. A parsimonious model of error analysis is provided, along with operationalized interpretations and implications for motor learning. Teaching, interpreting, and using error scores in research may be simplified and facilitated with the model.     

Accuracy of estimations of body-frame size as a function of sex and actual frame size

Body-frame size is an important factor in determining an optimal body weight for a given height. Previous studies have indicated that many individuals incorrectly estimate their body-frame size, and, as a result, incorrectly assess their ideal weight. The present study investigated the accuracy of estimation of body-frame size as a function of sex and actual frame size. The subjects were 66 men and 52 women participating in a community adult fitness program. Data indicated that medium-framed individuals were the most accurate in their estimations of body-frame size. Also, women were twice as likely to be accurate as were men. These results are interpreted to mean that most people assume they are medium-framed and that there is a sex difference in the way body-frame size is estimated.     

SETTING CUTOFF SCORES: LEGAL, PSYCHOMETRIC, AND PROFESSIONAL ISSUES AND GUIDELINES

Cutoff scores are used in a variety of settings to help managers make personnel decisions. Although cutoff scores are used widely, information about their derivation and appropriate use is scattered in the legal, psychometric, and professional literatures. The purpose of this paper is to analyze critically and to integrate these diverse literatures, to summarize what is known and what is unknown about the use and misuse of cutoff scores. Alternative methods are examined in relation to legal precedents set by the courts, to psychometric principles, and to the various sets of professional guidelines. Directions for future research are suggested, and we conclude with a set of recommendations regarding acceptable professional practice in this area.     

Effects of eye color on frisbee toss

Light-eyed individuals generally perform better at self-paced activities while dark-eyed individuals perform better at reactive activities. Using multiple regression it was found that dark-eyed students hit a target with a frisbee more times than did light-eyed students.     

Follow-up on the youngest REHABIT client: importance of caution

The youngest subject (age 5) participating in the 1986 evaluation studies published in this Journal returned 25 mo. after finishing REHABIT training. This boy's test scores showed a significant decline in verbal ability. It was concluded that follow-up studies of REHABIT clients are necessary and that such clients should be cautioned that regression may occur.     

Relation between musical aptitude and intelligence among mentally retarded, advantaged, and disadvantaged subjects

The purpose of this study was to investigate differences in scores of 59 mentally retarded adults (mild, moderate, severe, or profound), 133 advantaged, and 130 disadvantaged (low income) children in Grades 1, 2, 3 on the Primary Measures of Music Audiation. Test-retest reliabilities for the Tonal and Rhythm subtests were .81 and .86, respectively, for the retarded group. Analysis of variance indicated that the mildly retarded children performed significantly better than other groups of retarded children on these two tests. 2 x 3 analyses of variance indicated that third graders from both advantaged and disadvantaged groups performed significantly better than the first or second graders on both subtests and that advantaged children performed significantly better than the disadvantaged. However, disadvantaged children made greater gains, especially for third graders, than the advantaged.