Evaluation Review

The dramatic secular increase in the labor force participation of women with preschool children has led to arguments for some type of federally subsidized system of day care centers. In order to analyze the need for federal action several studies of the demand for different modes of childcare by households of different socioeconomic characteristics are critically reviewed here. These studies indicate that a family's income has little effect on modal choice and most working mothers choose at-home care for economically rational reasons. The economic arguments for federal intervention m the child care market are discussed and analyzed. The weight of the evidence indicates that the case for a public role in this market is weak as the private sector is now providing the child care arrangements households demand.

This article reports the results of an experiment to determine if increased enforcement of sanitation laws improves the cleanliness of city streets. The experiment was conducted over a six-week period in New York City Increasing the number of sanitation policemen was found to decrease the rate of production of litter as measured by daily cleanliness ratings. Implications for future research and public policy are discussed.

When participants with varied background and interests join in a collaborative activity, their different viewpoints may make the evaluation of the activity more difficult. The emphasis placed on different kinds of success may differ greatly among project partici pants, causing them to disagree over the worth of components of the program, irrespective of the technical merits of the evaluation of these components. Our experience as evalu ators in a multidisciplinary biomedical research center indicates the need to augment conventional evaluative techniques with ones that explicitly recognize the influence of values on program evaluation.

There is general agreement that human services (i.e., services that depend on direct inter personal contact between a deliverer and a client) are difficult to evaluate. This article points out some of the sources of this difficulty: first of all, the theories lying behind the delivery systems are often deficient. Second, delivery of such services is highly operator- dependent and hence often radically transformed in delivery, in ways that tend to negate the intended treatment effects. A strategy for the evaluation of human services delivery is proposed consisting of several steps, in which the underlying theory is first tested, next the ability of any system to deliver the services in question, and finally whether given delivery systems can deliver a relatively pure version of the service.

The SEEK program in the senior colleges of the City University of New York provides special services (e.g., stipends, counseling, tutoring) to disadvantaged students. Cohorts of entering freshmen were followed through eight, ten, and twelve semesters, graduation data for SEEK students are compared to those for students who were eligible for SEEK, but denied admission to it by random procedures, and who entered the senior colleges anyway.

This research is an evaluation of the impact of engineering and policy interventions on bicycle accidents occurring on the campus of the University of California'at Santa Barbara between October 1970 and March 1977. The number of bike accidents per day were regressed on 15 predictor variables including the engineering and policy innovations. Generalized least squares estimation procedures were used to adjust for serial correlation. None of the bikeway modifications were found to have been effective in reducing bicycle accidents on the campus.

Evaluation of the productivity of government-applied social research is considered from the aspects of the several major government and public constituencies concerned, in terms of "Whose benefits? Whose costs?" Public and government concepts of the public good in evaluation research are differentiated. Government-imposed constraints on the quality and quantity of evaluation research are outlined and their impacts considered. Constraints on research quality imposing unnecessary costs include misregulation of sponsored social research by rigid procurement regulations inappropriately modeled on military hardware procurement, rigid overspecification of research tasks, counter- productively cumbersome and costly OMB survey clearance procedures, and inefficient long-term investment. Some impacts on research community employment security, professional development, and R&D investment are considered. A need to disseminate more explicit standards of effective and effectively managed evaluation research is expressed.

Our customary strategies for protecting ourselves against the charge of subjectivity in our reports often prevent us from addressing phenomena of central importance: the expec tations, fears, beliefs, and motivations of both the evaluated and the evaluator. We need new strategies that will support our commitment to science and at the same time preserve the integrity of our endeavor. The article discusses three strategies: one dealing with our philosophical foundations, one with the subjectivity of the "objects" of our studies, and one with the subjectivity of the observer.

Maltz and McCleary have suggested a split-population exponential distribution to predict parolee recidivism where one portion of the population will never return and the remainder will return according to an exponential distribution. They have estimated parameters to fit the distribution to some sample data. A reanalysis of their data shows that the parameter estimates change systematically as more and more observation time is allowed in the computation. It is argued that this feature limits extrapolability. Local data concerning readmissions of discharged psychiatric inpatients are used to illustrate this point.

This bibliography lists references to cover 300 randomized field experiments undertaken in schools, hospitals, prisons, and other social settings. The list is divided into 10 major categories corresponding to the type of program under examination. They include: criminal and civil justice programs, mental health, training and education, mass media, information collection, utilization, commerce and industry, welfare, health, and family planning. The main purpose of the bibliography is to provide evidence on feasibility and scope of randomized field tests, since despite their advantages, it is not always clearfrom managerial, political, and other constraints on research that they can be mounted.

In recent years, many researchers have reported that the utilization of evaluation infor mation by public administrators has been limited This article describes a new approach for building a bridge between the evaluation and policy-marking functions in the Florida Department of Health and Rehabilitative Services The formal linkage, termed the evaluation analysis and policy development process, has important implications for the utilization of evaluation findings, as well as for the role and level of satisfaction of the career evaluator.