This paper explores the use of K. G. Joreskog's congeneric modeling approach to reliability using censored quantitative variables. In addition, the compound problem of non-normality and attenuation that occurs when estimating censored continuous variables is discussed. (TJH)

The superiority of multiwave designs over two-wave designs in the measurement of individual growth and change is explored by examining the increase in reliability afforded by collecting additional waves of data. Results are discussed in terms of the impact on longitudinal studies of individual change. (SLD)

Two Likert-type scales were used by 25 undergraduates to assess their reading and writing; assessments were compared to those of three judges. The study found a significant relationship between self-evaluation accuracy and reading and writing performance. Prior positive reinforcement did not significantly relate to reading or writing achievement.…

Developed behavior observation system for group supervision, designed to measure occurrence of verbal problem-solving behavior during group supervision meetings. Obtained reliability data from four counseling teams' group supervision meetings at residential facility for children with social and emotional problems. Results indicated that system had…

Reviews the history of verbal data in a variety of fields, places protocol analysis in its historical context, and examines more recent claims and criticisms regarding protocol analysis. Concludes that protocol analysis, when conducted according to certain principles, can be an important tool for researching the composing process. (MG)

Discussion of user satisfaction with computers focuses on a study that tested the construct validity and reliability of the End User Computing Satisfaction Instrument. Highlights include the relationship of user satisfaction and computer attitudes, computer anxiety, computer skill, and computer usage; and implications for the management of…

Twelve adult stutterers were instructed to self-rate the speech naturalness (how natural their speech sounds) and feel naturalness (the amount of attention they are paying to the way they are speaking) of their speech under a variety of conditions across repeated rating occasions. Results showed that subjects were relatively consistent and valid…

Four traits of voluntary organizations (group size, age composition, sex composition, and educational composition) were measured with three different methods: reports from a probability sample of members, reports from a group official, and direct survey of members at an organizational meeting. In general, the three methods of gathering data…

A proportional reduction in loss (PRL) measure for reliability of categorical data is explored for the situation in which each of "N" judges assigns a subject to one of "K" categories. Calculating a lower bound for reliability under more general conditions than had been proposed is demonstrated. (SLD)

Problems and illustrations are presented from a meta-analysis of crime reduction programs. Eight criteria are specified for including studies in a meta-analysis, and some problematic studies are discussed using the criteria. Rules for handling problem studies in order to retain consistency throughout the analysis are discussed. (SLD)

The reliability coefficient is predicted from the test information function (TIF) or two modified TIF formulas and a specific trait distribution. Examples illustrate the variability of the reliability coefficient across different trait distributions, and results are compared with empirical reliability coefficients. (SLD)

Examines consistency of results and variables that may confound validity in child sexual abuse cases. Concludes that the ways in which information on the incidence and prevalence of child sexual abuse has been collected, analyzed, and reported have been flawed to the extent that few reliable conclusions can be made. Includes suggestions for…

Three methods for grouping items in an opinion survey were compared for their utility in subscale construction: rational organization according to content, factor analysis, and multidimensional scaling. Only subscales based on factor analysis could be refined to meet the criteria of reliability, additivity, and interpretability simultaneously. (SV)

An approximate statistical test is derived for the hypothesis that the intraclass reliability coefficients associated with two measurement procedures are equal. Control of Type 1 error is investigated by comparing empirical sampling distributions of the test statistic with its derived theoretical distribution. A numerical illustration is…

Discusses online search techniques that can be used to identify inaccuracies and inconsistencies in records of online and CD-ROM databases. Methods that can be used to sample a database for errors in descriptor, name, or coded fields are described, including browsing, matching, looking up cross-references, and searching for implausible record…