A procedure for determining the reliability of an examinee knowing k out of n possible multiple choice items given his or her performance on those items is presented. Also, a scoring procedure for determining which items an examinee knows is presented. (Author/JKS)

Comparison of black and white men in the experienced civilian labor force between 1972 and 1973 shows that race and class have become less important in the occupational hierarchy. Statistical tables are included. (ND)

The relation between average deviation, as determined using the Illinois Test of Psycholinguistic Abilities, and Stanford-Binet intelligence scores was examined using a preschool sample. Results revealed a curvilinear relation between total average deviation and Stanford-Binet intelligence scores. Use of average deviation as an index of…

This document contains statistical demographic information describing the gender, age, and ethnicity of the 57,772 students enrolled at City College of San Francisco (CCSF) during the fall 1997 semester, as well as the ethnicity of the faculty, classified staff and administrators working at CCSF at that time. The report includes 1996 figures for…

The mini-computer is found to be a viable alternative to large computer bio-medical information processing. (Author/RS)

Reviews the Sixth Berkeley Symposium on Mathematical Statistics and Probability, in which participants sought to establish a realistic cross-section of contemporary statistical thinking about the effects of pollution on health. Present information of health effects of pollutants is fragmental, and further studies are urgently required for…

Rebuttal to Gage's view that the elimination of racial differences depends on education, not on eugenics. (MF)

Discusses the heritability of intelligence. (MF)

Describes the problems, procedures, findings, and implications for education of a study designed to identify the ideal teacher as related to classroom behavior traits. (GR)

The reliability of simple difference scores is greater than, less than, or equal to that of residualized difference scores, depending on whether the correlation between pretest and posttest scores is greater than, less than, or equal to the ratio of the standard deviations of pretest and posttest scores. (Author)

Calls attention to limitations and dangers in routine application of multivariate analysis of variance (MANOVA), describes some alternative procedures, and laments the necessarily pervasive character of the statistical problem. (Author)

Data for four-, five-, and seven-position Likert formats from 292 undergraduates showed systematic error varied among formats, i.e., central tendency errors tended to increase with increasing number of categories and to reduce variances expected. (Author)

Basic rules are suggested to replace the signed coefficient rule to infer the form of moderator effects in terms of slopes of regression lines. Steps are provided to interpret the form of moderator effects that can be achieved using only the regression coefficients of the predictor variable and interaction term. (SLD)