Two stage multiple-comparison procedures give an exact solution to problems of power and Type I errors, but require equal sample sizes in the first stage. This paper suggests a method of evaluating the experimentwise Type I error probability when the first stage has unequal sample sizes. (Author/BW)

Analysis of individual-level data from the 1970 Census confirms earlier findings, based on census tract data, about Hispanic and Black spatial assimilation in selected cities. Besides validating the theory of spatial assimilation, the results also indicate that errors resulting when individual-level inferences are made from ecological data are…

Introduces structural modeling with nonnormal continuous variables, using the equations language of the micro-mainframe program EQS in the context of a longitudinal study of adolescent development that followed about 700 adolescents across an 8-year span into young adulthood. Two models are developed to assess the influence of drug use on…

Considers problems which arise when researchers do not have the optimally large sample sizes desired in structural equation modeling. Discusses the ways in which small sample size affects assessment of model fit. Provides a new estimator that may be beneficial for use in small-sample situations. (Author/RH)

States that regardless of the content or level of a statistics course, five goals to reach are: (1) overcoming fears, resistances, and tendencies to memorize; (2) the importance of intellectual honesty and integrity; (3) understanding relationship between deductive and inductive inferences; (4) learning to play role of reasonable critic; and (5)…

States that the teaching of logit regression analysis is much neglected in statistics courses within sociology. Claims that logit regression allows the researcher to evaluate the impact of a set of predictor variables on a dichotomous dependent variable without the problems associated with discriminant analysis, weighted least squares regression,…

Multiple discriminant function analysis was conducted with data from Strange Situations. Results enable researchers to obtain attachment classifications directly from scores on interactive behavior and crying during reunion episodes. (PCB)

Research on refugees in the United States accentuates the normal problems of anthropological field work, such as blurred roles and role conflict, the degree of familiarity of the research environment, and cultural shifts made in changing environments. (MSE)

Contends that success of supported employment service programs depends on the ability of new programs to replicate results of supported work demonstration projects. Notes that the capacity to verify accurate replication independently is limited, as is knowledge of the extent to which supported employment demonstration programs generalize to a wide…

Used data from 195 children and 207 adolescents to assess and compare quality of data from those age groups. Interviewer rating of child's behavior, quantity of missing data, and percentage of agreements between parents and children on objective family characteristics all suggest that quality of data was high for children and significantly higher…

Presents three studies investigating discrepancies in published American research on participation and its outcomes. The first study, a longitudinal review of research published between 1950 and 1985, showed societal trends affected research questions, methodology, and magnitude of reported effects. The other two studies corroborated these…

The dangers of equating "statistical significance" with "real world" significance are summarized. When a finding is said to have "statistical significance," it means only that the same results would be likely to occur again if the study were repeated, not that the finding has any true personal or societal importance. (VW)

Examines narrative choices in experimental (interpretative) and traditional (analytical) ethnographies. Discusses probability in quantitative and qualitative research, ethnographic narratives as ways of knowing and telling about the world, perspective as a consequence of both narrative voice and stance, and the economics of producing…

A comparison of two Finnish data sets on use of time and on drinking habits raises the question of the validity of data in studies of use of time. The nature of drinking as an activity is discussed and data on time-related aspects of drinking in various contexts are presented. (Author/MT)

Ratcliff (1985) simulated data from three letter-matching experiments with his diffusion model. The necessity of including a comparison criterion is consistent with the conclusion of Proctor, Rao, and Hurst (1984) that bias of response criteria, alone, is insufficient to generate the fast-same phenomenon. (Author/LMO)