The recommendation to use group means when there may be nonindependence among observational units is unduly restrictive. When random factors are properly identified and included in the analysis, the results are identical in balanced analysis of variance designs, irrespective of whether group means or individual observations are employed.…

Sampling experiments were used to assess the Type I error rates of the t test in situations where classes were randomly assigned to groups but analyses were carried out on individual student scores. Even small amounts of between-class variation caused large inflations in the Type I error rate of the t test. (Author/BW)

This paper demonstrates that gain score analysis and analysis of covariance (ANCOVA) will always yield conflicting results unless the slopes of within-group regression lines of posttest on pretest equal one. In Lord's specific hypothetical example, trait instability invalidates the ANCOVA. An analysis that corrects for this problem is…

A method was designed to provide history students an opportunity to make interpretations and form hypotheses about historical events and eras. The teacher created representative "facts" and questions about a topic and let students use them to make their own hypotheses. A sample unit on European history, 1300-1450, is included. (KC)

The interpretation of significant left-tailed analysis of variance (ANOVA) F-ratios is supported by considering the case of a fixed effects ANOVA model. The conclusions of this case are generalizable to other standard ANOVA models. (Author/PN)

Attempts are made to improve the previously used procedure for multidimensional scaling in the measure of conceptual level. Monte Carlo simulation runs and a laboratory experiment highlight potential incongruities related to the degree of error in the input data required by that process. Methodological recommendations are briefly evaluated.…

Tests two hypotheses concerning success in debate: (1) that success would be enhanced by a fast speaking rate, large amounts of evidence, and use of jargon; and (2) that a high correlation exists among these same variables. Neither hypothesis is supported and several explanations are offered. (JMF)

Causes of evaluator bias are: overemphasizing concrete, salient, and retrievable information; reporting only evidence which confirms hypothesis; focusing on stable personality factors, rather than on situation and environment; developing positive perceptions of a program as both an evaluator and a highly involved participant; statistical naivete;…

High state-anxious subjects solved fewer problems than middle or low state-anxious subjects under no memory-aid conditions, and all anxiety groups performed comparably with memory aids. High state-anxious subjects tended to use less focusing strategy when memory aids were unavailable. (Author/CP)

Symbolic interaction theory, exchange theory, and systems theory were major schools of thought in the 1970s. The general theory of the family may be unnecessary. Theorists in the 1980s should improve existing theory and continue to develop metatheory and methodologies of building theory. (Author)

Twelve variance-analytical and nonparametrical coefficients of reliability for rating scales designed for rating persons were compared to each other theoretically and empirically. Preference for two coefficients was established. The intraclass correlation coefficient appeared to be useful for the estimation of reliability as well. (Author/RL)

Results of a study of the relationship between teachers' perceptions of five attributes of an innovative curriculum and the observed degree of its implementation generally supported the hypothesis that perceptions of the innovation's relative advantage, compatibility, trialability, and observability correlate positively with degree of…