A. B. Silverstein

Statistical Power Lost and Statistical Power Regained: The Bonferroni Procedure in Exploratory Research

The Bonferroni procedure controls the risk of rejecting one or more true null hypotheses no matter how many significance tests are performed, but permits the risk of failing to reject false null hypotheses to grow with the number of tests. The loss of statistical power associated with the use of this procedure is demonstrated, and two options for alleviating the problem are explored. Setting a less stringent significance level for the set of tests is shown to be less effective than increasing the sample size.

Sex Differences and Sex Bias on the Boehm Test of Basic Concepts: Do They Exist?.

Form A of the Boehm Test of Basic Concepts (BTBC) was administered to 180 kindergartners—90 boys and 90 girls. Neither their mean scores nor their standard deviations differed significantly, and the distributions of scores for the two groups were virtually identical. Moreover, the application of three internal criteria for bias yielded totally negative results; thus, there was no evidence of either sex differences or sex bias in the data.

Social class differences on the Boehm test of basic concepts: Are they due to bias?

One of Jensens internal criteria for the detection of bias was applied to the standardization data for the Boehm Test of Basic Concepts (BTBC). Despite sizable social class differences in mean scores, differences in the rank orders of item difficulties were negligible. Thus, by this one criterion, there was little evidence of test bias. Instead, the findings suggest that children of different socioeconomic levels master basic concepts in about the same temporal order, differing only in the rate at which they do so.

A comparison of the 1976 and 1978 norms for the WRAT.

A comparison of normative data for the 1976 and 1978 editions of the WRAT showed that standard scores are quite closely comparable, except for those on Spelling II, but that grade ratings are not at all comparable at the upper ends of the distributions. In test-retest situations, the same set of norms should be used to evaluate performance on both administrations.

Reliability of the Concept Assessment Kit-Conservation for Educable Mentally Retarded Children.

Internal consistency, alternate form, and stability coefficients for the CAK-C were obtained for a sample of 135 educable mentally retarded children. The internal consistency and alternate form coefficients appear to be acceptably high, and the stability coefficients, while appreciably lower, still suggest some degree of consistency over a years time. The reliability of the CAK-C is judged to be satisfactory for EMR children.

Organization and Structure of the Detroit Tests of Learning Aptitude (DTLA-2).

The intercorrelations among the 11 subtests of the DTLA-2 for a portion of the standardization sample were subjected to both factor analysis and cluster analysis. None of the resulting factors or clusters corresponded exactly to any of the eight composites into which the subtests are partitioned, nor did the composites account well for the intercorrelations. Thus the organization of the DTLA-2 does not appear to reflect the actual structure of the instrument.

Multiple Regression Analysis of Split-Plot Factorial Designs

Cohen and Cohen (1983) and Pedhazur (1982) have describeddifferent procedures for the multiple regression analysis of split-plot factorial designs. In the present study both procedures wereapplied to a small data set previously analyzed by Kirk (1982), whonoted that two cases need to be distinguished when the groupscontain unequal numbers of subjects. The results demonstratedthat (a) Cohen and Cohens procedure and Pedhazurs give thesame results when applied to the same data, and (b) both procedures address the case in which the investigator deliberatelyselected unequal numbers of subjects for the groups, proportionalto the sizes of the corresponding populations.

Stability of Two Piagetian Scales with Severely and Profoundly Retarded Children.

Corman and Escalonas scales for Object Permanence and Spatial Relationships were administered to 98 severely and profoundly retarded children on three occasions, with intervals of six months between successive administrations. Correlations between the scores on the three administrations ranged from .76 to .87, and the mean change was less than one point on each of the scales. These findings demonstrate the high stability of the scales when environmental conditions are themselves highly stable.