Examines the effects of high school performance, high school and university mathematics background, previous computer experience, and age on computer science aptitude and achievement as measured by a computer science aptitude test and final exam respectively. Methodology and analysis of results by stepwise regression are presented. (EAO)

Describes a procedure for identifying a group of potential users from among current nonusers in a sample of adults drawn from the library's community. This procedure was designed for use in conducting marketing surveys for public libraries. (Author/CHC)

Simple methods for detecting attrition (nonresponse) in longitudinal evaluations are reviewed, focusing on regression-based analyses of data from a longitudinal evaluation. The approaches are illustrated with data from the Fort Bragg Evaluation, an evaluation of a major demonstration in children's mental health services. (SLD)

Uses Texas data to explore using an indirect measure of local wealth as the basis for calculating indexes of wealth neutrality. The proposed measure, the percentage of school district revenues derived from local sources, can easily be calculated from data contained within the U.S. Department of Education's Common Core of Data. The measure…

Combined multiple regression and classification and regression tree (CART) analyses to study the interrelationships of variables that affect the item difficulty of items on tests of second language reading and listening comprehension. Results for 214 computerized items completed by 87 nonnative English speakers show the value of synthesizing…

Examined the relationship between fathers' parental stress and their perceived sense of competence in parenting roles. Regression analyses of data from 94 fathers suggest that fathers' depression in their parental role and perceptions of their children's demandingness are the best predictors of their perceived parental competence. (Author/TE)

Using a 1983 survey of Pennsylvania employees' hiring in seven occupations, this article reports differences in employers' observations concerning the effect of postsecondary technical education on employee performance, training periods, turnover, and worker development after hiring. Generally, employers gave more training to workers with a higher…

Data from the 1979 follow-up of the National Longitudinal Study of the High School Class of 1972 suggested that early career experiences of undergraduate humanities majors differed little from contemporaries in other liberal arts fields, but contrasted with those in which job-major fit was better or pay higher. (Author/MLW)

Examines cross-metropolitan variation in Puerto Rican poverty rates using an instrumental variables regression model. The analysis highlights the roles of residential segregation and economic restructuring on Puerto Rican poverty rates in 1970 and 1980. During the 1970s, the contributory effects of segregation grew stronger, whereas the…

Analysis of 28,000 student evaluations of faculty in 46 departments of the University of Dayton. Overall instructor effectiveness was predicted most strongly by ratings on three items: instructor was prepared, presented subject matter clearly, and was interesting. Predictors of students perceiving they "learned a lot" were instructor was…

Identifies predictors of symptomatic distress in emergency services personnel exposed to traumatic critical instances. Evaluated predictors included exposure, social support, and psychological traits. Replicated analyses showed that levels of symptomatic distress were positively related to the degree of exposure to the critical incident. Level of…

An alternating least squares method for iteratively fitting the longitudinal reduced-rank regression model is proposed. For social and behavioral science applications, the developed algorithm does not rely on the assumption of multinomial errors and allows for scaling of input and output variables. (SLD)

A trilevel (macro, intermediate, and micro) analytical framework was used to determine problems in the organizational context that reduce or impede the application of training outcomes. A law enforcement job-skill training program was studied via survey responses of 322 course graduates. Factor analyses identified predictors of training use. (SLD)

A new measure of health status, the Health Status Scale, is presented. This five-indicator scale is used as the dependent variable in a regression equation using six common predictors of health status. Results indicate that health facilities are the most significant predictors of health status. (SLD)

A version of the randomized response survey unrelated question technique is presented in which the answer to the unrelated question is known to be yes. Raw data become a direct linear transformation of response to the sensitive question and can be used in analysis at the individual level. (SLD)