To be or not to be: Control and balancing of type I and type II errors

The various costs of Type I and Type II errors of inference from data are discussed. Unfortunately, other things equal, Types I and II errors are inversely related. Six methods of minimizing Type I error in studies are presented, each of which may be employed even after data are collected. Type II errors may also be minimized by a combination of study design and analytic means. Although some of these techniques must be built into the study others can be undertaken when beginning work on a data set. Only one of these techniques will increase Type I error, and those involving data reduction may decrease both types of error.     

Price vs. revenue stabilization through a buffer stock: Which is more financially feasible for LDCs?

Optimal control theory is used to analyze the implications of the adoption of price and LDC export revenue stabilization objectives by an international buffer stock for cocoa. The results obtained for the period 1956–76 suggest that the stabilization of either price or revenue at systematic trend would reduce the instability of both variables from that during the sample period. Although the stabilization of revenues at systematic trend decreases their average level, the stabilization of price has the opposite effect. Because of this, it may be financially feasible for the LDCs to provide the necessary resources for a price-stabilizing buffer stock.     

Predicted outcomes in a national sample: A source of comparison for program evaluation

The lack of comparison groups is a common problem for researchers conducting evaluations of small, specialized programs. When conventional comparison groups are not available, the authors suggest a possible alternative approach which involves: (a) locating a sub-group in a national study for whom relevant variables have been measured; (b) generating a statistical model which predicts the outcome of interest in the national data; and (c) applying the predictive model to program data, in order to estimate outcomes in the absence of progam intervention. The authors illustrate this new application by presenting the results of an analysis of data from Project Talent, a national manpower study, and discuss its application to the evaluation of a medical training program in New York City.     

Ethnic and class factors affecting mental health clinic service

Service delivery by a mental health clinic to clients from an older suburban catchment area containing both lower and middle class blacks and whites was studied. The epidemiologic method was employed, relating service delivery to the characteristics of the various population groups within the catchment. The findings indicate that low service rates for blacks were not related to ethnicity alone, but to the combination of black ethnicity and low social status; however, middle class blacks and middle class whites who entered treatment received similar service. Poor blacks had higher admission rates than poor whites with similar need indicators. In continued treatment, poor blacks received lowest service rates while minority status poor whites living in a predominantly black area received highest service rates. Social class mediated black service utilization. Severity of illness, interacting with social class and minority status mediated white service utilization. Outreach is suggested not only for poor blacks but also for poor whites, especially those living in areas in which they represent an ethnic minority.     

Behavioral epidemiology and social area analysis: The study of blackout looting

This article describes the use of social area analytic techniques in the study of the looting which occurred during the Con Edison power failure of July 13 and 14, 1977. Detailed procedures are outlined of the systematic collection and analysis of data which were developed from available data. Analyses contrasting competing explanatory theoretical models for blackout looting are presented. The central dependent variable is ratio of looted to retail stores. Both the numerator and denominator are aggregated to the census tract level.     

Sex stratification in the workplace: Male-female differences in economic returns to occupation

Employing data from the 1974–1977 NORC General Social Surveys, I investigate differences in the earnings attainment of currently employed white men and women age 25 to 64. I focus special attention on the explanatory effects of job characteristics other than those traditionally employed in prestige and status-defined earnings models. The results, based on a multivariate regression analysis and a regression standardization procedure, suggest that a nontrivial portion of the earnings gap between men and women is due to women's concentration in jobs which are low-paying and heavily female and because women are less likely than men to exercise authority in their jobs or to control the means of production. Including these factors in an earnings model statistically increases women's earnings as a percentage of men's by over 7%, accounting for approximately 13% of the earnings gap. Net of these job characteristics, gender differences in industry distribution are not substantively important in explaining why women earn less than men, accounting for only 0.4% of the earnings gap. When single women's earnings are adjusted to take account of their occupational concentration, 10% of the male-single female earnings gap is explained, providing preliminary evidence that the job characteristics I specify are not mere proxies for work experience. Including job characteristics as measures of the context of employment thus usefully extends the human capital and prestige or statusdefined models traditionally employed in explanations of the male-female earnings differential.     

Organizing evaluation to serve the needs of program planners and managers

Evaluation is often both conceptually and functionally separate from program planning and operations management. Though program managers may not see themselves as served well by the work of evaluation staffs, there are a number of potentially beneficial side effects to be realized from close collaboration between program administrators and evaluators. These flow only when the evaluation task is broadly defined and when both administrators and evaluators recognize their interdependence. Particular functional benefits to management from evaluation are described. Characteristics of interorganizational relations which might increase the probability of their attainment are then outlined and discussed.     

The use of the psychosocial maturity inventory in evaluations of youth conservation corps programs

Although work experience has been widely recommended as beneficial to adolescent development, there is little empirical support for this recommendation. The Youth Conservation Corps (YCC), combining as it does some of the best features of both paid work and unpaid community service, offers an opportunity to explore the developmental impact of work experience. However, the need to pay as much attention to the problems of measurement as to the programs themselves dictates simultaneous evaluation of both programs and measures.     

Supervised access: A judicial alternative to noncompliance with visitation arrangements following divorce

Visitation of children after divorce by noncustodial parents is a “natural” right, generally regarded to be in the best interest of children. Custodial parents frequently refuse to allow visitation, even when ordered to do so by the court. The purpose of this report is twofold: (a) to review issues and problems related to visitation; and (b) to describe a program, developed under family court sponsorship, which provides mediation to families in visitation disputes, and supervised visitation when more reasonable access is frustrated by noncompliant parents or when there is concern for the safety and well-being of visited children.