Measurement error in dependent variables reduces the power of statistical tests to detect mean differences of specified magnitude. Procedures for determining power and sample size that consider the reliability of the dependent variable are discussed and illustrated. Methods for estimating reliability coefficients used in these procedures are…

The results of 16 pairs of "equivalent papers" were used to estimate the reliability of the papers and the extent to which each paper correlated with the year's average test grade. Estimates were also made of the work of the grade for each paper as a predictor of true subject grades. It is shown that a "profile" of grades would mislead.…

Twelve children with Down's syndrome (ages 3-5 years) were tested six times over 2.5 months on three Piagetian infant search tasks. Results suggested that cognitive ability of this population may be poorly measured by single-session testing and that caution is necessary when using tests designed for and validated on younger, nonretarded subjects.…

This study of the stability of parent behaviors toward toddlers over a 3-week period used both aggregated and nonaggregated data. Comparison of stability correlations indicated higher stabilities for aggregated scores, with the level of stability increasing as scores from additional single sessions were aggregated. (PCB)

Two new methods for constructing a credibility interval (CI)--an interval containing a specified proportion of true validity description--are discussed, from a frequentist perspective. Tolerance intervals, unlike the current method of constructing the CI, have performance characteristics across repeated applications and may be useful in validity…

This article discusses the ethical implications of using tests intended only to protect the blood supply as diagnostic or prognostic tests for Acquired Immune Deficiency Syndrome (AIDS). The reliability of these tests and the confidentiality of their results are discussed, along with ethical issues of research regarding AIDS. (MT)

Drew from the occupational prestige literature in describing the development of prestige estimates for occupations contained in the Minnesota Occupational Classification System III (MOCS III). Results provided estimates for 60 occupations and a comparison of prestige scores for eight benchmark occupations. Devised equation for estimating…

Reviews arguments in favor of administrative use of public opinion information; demonstrates insights to be gained from examination of poll data; and considers methodological weaknesses of such data and strategies to counter them. (JOW)

Evaluated a short form of the Roommate Rapport Scale that would maintain the scale's reliability and eliminate potentially objectionable items using students (N=320) who resided in dormitories. Results showed the short form to be reliable and unidimensional. (ABL)

The validity generalization model of meta analysis is described and modified for application to concurrent validity studies of motor behavior. An example is presented. (Author/MT)

Self-, peer, and supervisor ratings were obtained for simulated interviews using seven different measures of counseling performance. Results indicated significant differences among self, peer, and supervisor ratings and among four of the seven dependent measures. (Author/BL)

Two alternatives to the problems of conducting planned and post hoc comparisons in tests of concordance and discordance for G groups of judges are examined. The two models are illustrated using existing data. (Author/JKS)

Habituation to single female faces and to single geometric patterns was observed separately in two groups of infants who participated in two sessions separated by 10 days. Habituation was found to be distributed into three patterns and showed moderate but significant reliability between assessment sessions. (Author/RH)

Outlines issues pertaining to possible sex bias in job evaluation procedures and reviews relevant research. Gives attention to possible sex bias in job analysis procedures, choice and weighting of factors, and reliability and validity issues. Discusses future research needs, particularly reliability and validity aspects of job evaluation…

Showed that reliabilities in the .50 range can be obtained when raters rule out only 15-20% of the items on the Position Analysis Questionnaire as "Does Not Apply" and respond randomly to the remainder. (Author/ABB)