Results of this study comparing coverage and features of Psychological Abstracts and National Institute of Mental Health (NIMH) online databases demonstrate a 49 percent citation overlap between the files. A complementary, not competitive, relationship is indicated by comparing citation overlap, journal coverage, descriptive indexing, treatment of…

The perceptions of a representative sample of GRE test takers who were asked to indicate their views of the importance of eight factors in graduate admissions were examined. Candidates perceived undergraduate grades as the most important factor in graduate admissions, followed by recommendations and test scores. (Author/MLW)

Binomial models for domain-referenced testing are compared, emphasizing the assumptions underlying the beta-binomial model. Advantages and disadvantages are discussed. A proposed item sampling model is presented which takes the effect of guessing into account. (Author/CM)

A psychological theory of educational productivity is tested, and the usefulness of the National Assessment of Educational Progress data for secondary analyses for policy purposes is explored. Class as a social psychological factor and the didactic quality of instruction appear to be the only unequivocal and potentially manipulable causes of…

Job change patterns of academic administrators were studied, focusing on trends for different groups of administrators. The prevalent pattern was for persons to assume their current positions from within institutions. Job changes within and across colleges and universities differed based on personal and career characteristics including gender and…

In time-on-task observations, the effects of variations in methodological characteristics on the importance of time-on-task for student achievement were examined. Substantive conclusions were affected by variations in five areas with the greatest differences due to changes in the duration and number of days of observation. (CM)

Concludes little relationship exists between the three substances marihuana, alcohol and tobacco. Youthful subjects tend to overestimate the relationships between the three substances and cannot be generalized to other populations. Suggests an explanation of this youthful association focuses on simultaneous experimentation rather than causal…

Describes two types of criterion-referenced testing that districts can use to measure achievement outcomes of their instructional programs: a random sampling assessment plan and an item-response theory assessment plan. (Author/JM)

A Monte Carlo experiment was used to evaluate four procedures for estimating the population squared cross-validity of a sample least squares regression equation. One estimator was particularly recommended. (Author/BH)

Investigators are urged to plan factor analytic studies prior to collecting the data, to formulate a hypothesized factor structure, to develop several relatively pure measures of each factor expected, and to select an appropriate sample of at least 200 cases. Continuous variables should be used rather than dichotomous variables. (Author)

Nonreturn of mail surveys was described in this paper in terms of selectivity bias, that is, the bias introduced in survey results due to respondents and nonrespondents differing in unmeasured and unmeasurable ways that might yield inconsistent population estimates. (CT)

Discusses a set of neuropsychodiagnostic rules proposed by Selz and Reitan. That study was found to be seriously limited on theoretical grounds and in terms of research methodology. Sampling problems were indicated, and issues of neurological integrity and learning disabilities were poorly conceptualized. (Author)

A major problem in social science research is that of successfully carrying out the random assignment of persons to experimental and control groups. In this study a computer-based random assignment procedure operated successfully on a weekly basis for 17 consecutive weeks in a program serving over 360 ex-offenders. (CTM)

Discusses differences in results of factor analyses of ten diverse samples which have been studied using the Brazelton Neonatal Behavioral Assessment Scale (NBAS). Concludes that a single common factor structure accounts for the intercorrelations among NBAS items. (Author/BH)

To clarify the findings of R. Gillett (1991) about the inequality of the means of test scores of minority and majority examinees, the standard errors of the quota-selected sample means and the sampling distribution of these means were studied through Monte Carlo simulation. Results explain that the quota selection inequality results from…