A time-saving and space-saving algorithm is presented for computing the sums of squares and estimated cell means under the additive model in a two-way analysis of variance or covariance with unequal numbers of observations in the cells. The procedure is illustrated. (Author/JKS)

Transcripts analyzed by computer programs reveal that when individuals shift from truthfulness to deception, there are corresponding shifts in lexical diversity, such as fewer words, decrease in the number of unique words used, fewer past tense verbs, and a decrease in the perceptual-cognitive activity. (PD)

Authors reanalyze the data of the High School and Beyond research study and question its conclusions that private schools produce greater cognitive outcomes and are more racially desegregated than public schools. (EF)

A confirmatory, longitudinal design was used to determine whether deviant adolescent peer culture influences subsequent drug use beyond that which may be obtained from knowledge of present drug use and perceived sources of supply and support. Peer culture measures did not increase prediction over a one year period. (Author/JKS)

Factor analysis of coping behavior in adolescents in Frankfurt, Germany, and Tel Aviv yielded similar structures, but Israelis demonstrated a marked preference for internal coping strategies as opposed to coping by means of social resources. Discusses the impact of environmental and historical conditions. (DM)

A canonical analysis disclosed significant relationships between personality traits and value choice among 334 upper-level undergraduate and graduate students at a midwestern state university. These relationships were interpreted in terms of their usefulness in understanding human behavior. (Author/TE)

Multi-group discriminant analysis was used to identify the pre-enrollment demographic and academic factors that best separate the success classes in the first-year examinations of science students at the University of Benin. The predictor of success was performance in specialist subjects of the matriculation examination. (Author/MLW)

Study of 843 children under 36 months of age enrolled in a prepaid health plan from September 1985 through March 1986 revealed an increased incidence of illness in children attending day care centers, but not day care homes. The number of other children in the room is the major risk factor in a child care setting. (BJV)

The paper reviews a construct validation framework for research on the selection and evaluation of residents. The application of the proposed methodology to surgery residents is described. The need to measure non-cognitive and neuropsychological factors in addition to cognitive knowledge and technical ability is emphasized. (Author/MLW)

Using a quota sampling procedure, 514 Indiana citizens were asked about specific circumstances surrounding capital punishment and about their support for various goals of sentencing. Results suggest that although general favorability toward capital punishment is quite high, such a result is misleading with only nine percent supporting it under all…

Discussion of explanations and knowledge representation within knowledge-based systems focuses on the development of an expert system for purification of proteins in a Swedish laboratory. Topics include domain conceptualization; a multiple-explanation construction model; a study of laboratory staff that investigated construed explanations; and…

A method is presented for determining the optimal number of conditions to use in multivariate-multifacet generalizability designs when resource constraints are imposed. A decision maker can determine the number of observations needed to obtain the largest possible generalizability coefficient. The procedure easily applies to the univariate case.…

Studied methods for structuring a performance domain for a certification test in emergency nursing based on task frequency ratings from 659 emergency nurses or task similarity ratings from 21 experts. A 125-job analysis survey was used. Similarity judgment results are more easily interpreted and adequately modeled by multivariate analysis. (SLD)

Multilevel models used to assess and compare school effectiveness, particularly their application to multivariate data, are critiqued. Problems associated with standard errors of measurement and the various components of "school effect" are highlighted in terms of how they could affect incautious application of the models. (TJH)

Techniques are presented for use in meta-analysis to obtain effect sizes when information is missing from the original report. A method of reconstructing the mean squares within groups is presented for one-way analysis of variance and factorial designs. (SLD)