Clare, Donald A.

Integrated Solid Waste Management systems consist, in order of their occurrence, of source reduction, recycling, composting, combustion, and landfilling. Present theory and practice emphasize recycling as the stage with the greatest potential for dealing with the problem of rapidly growing solid wastes. The central proposition of this dissertation is that the rate of recycling is largely dependent on local political, demographic, economic, and technological conditions. To test this, the recycling rates of 221 counties of Florida, Minnesota, and Pennsylvania were examined. Primary data were gathered using a questionnaire-based survey sent to all 221 county recycling coordinators. An overall response rate of 62.9 percent was achieved. Twenty-four variables were tested for their relationship to recycling rates. Of these, four demographic variables, two economic variables, and four technological variables were identified as being significantly (p<0.05) related to the recycling rates in one or more states. No political variables were found to be significantly related to recycling rates in any state. A recycling model was developed for each state using multiple regression techniques. The Florida and Minnesota models suggest that recycling is highly influenced by local demographic factors, while the Pennsylvania model suggests that in that state the dominant influence is exerted by technological factors. The study developed the importance of the concept of per capita daily generation of municipal solid wastes. This concept, labeled "CMSW" was not found in the literature, but emerged from the study itself. CMSW, combined with the concept of non-recycled CMSW, were highlighted as practical and theoretically relevant variables for use in future research in the area of solid waste management.

Measuring performance in the public sector has been a topic of heightened concern during the past decade. While enthusiasm remains high for its use, performance measurement has also been subject to increasing scrutiny. In this research, problems with definitional ambiguity and concerns of conscious errors in the reporting of performance are theorized to limit the usefulness of performance measurement and benchmarking. The research examined the use of "response time" as a key benchmark in the delivery of paramedic services in the State of Florida. The research used both quantitative analysis based on survey response data from 120 paramedic agencies and qualitative analysis based on field visits to six selected agencies. The major findings of the study were that definitional ambiguity exists and that it adversely affects the ability to conduct benchmarking. Definitional ambiguity existed from two perspectives. There was ambiguity in the definitions used among agencies---thereby limiting the ability to benchmark. There was also ambiguity between definitions used by agencies and the definitions they attribute to citizens and elected officials---thereby limiting accurate communication of performance results to the public. While the research was not designed to examine intentions of paramedic agencies, the findings were consistent with Juran's theory of conscious errors---the reporting of performance that provides a more favorable impression than may otherwise be justified.

Policy implementation is recognized as a major component of the public policy process, and an important area for public administration research. The literature calls for the examination of public projects to improve policy design and implementation, for the comparison of expected implementation with actual implementation, and for the examination of the role of states in implementing federal policy. This dissertation addressed those areas by analyzing the implementation of the Omnibus Budget Reconciliation Act of 1987 (OBRA '87) which contains federal requirements for certification of nurse aides (nursing assistants) who are employed in Medicare and Medicaid funded nursing homes. The dissertation focused on the state of Florida and three levels of analysis: policies and practices of the state, nursing home operations and procedures, and individuals in the certification process. Multiple methods of data collection and data analysis were used and included: an analysis of the history, development and content of the federal policy, interviews with state administrators, a survey of nursing home administrators, an analysis of state certification statistics, documentary analysis, factor analysis and content analysis. Some of the data indicate that, although Florida has had a nurse aide certification process in place since 1982, there may be some uneven implementation of the certification process at the local level, and that race may have an impact on the certification rate. These areas were recommended for further study. A descriptive model, developed from the research design, outlined three dimensions that should be examined in order to conduct a comprehensive analysis of policy implementation: the origin of the policy, the agent of the policy, and the target of the policy. Additionally, a comprehensive analysis should examine the communication between the dimensions. In this dissertation, the federal legislation was the origin, the state of Florida was the agent, and nursing homes and prospective nurse aides were the targets. The descriptive model can be used for future implementation studies whether the origin of the policy is federal, state or local legislation.