Case study meta-analysis in the social sciences. Insights on data quality and reliability from a large-N case survey

Does stakeholder participation improve environmental governance? Evidence from a meta-analysis of 305 case studies

Participation and collaboration of citizens and organized stakeholders in public decision-making is widely believed to improve environmental governance outputs. However, empirical evidence on the benefits of participatory governance is largely scattered across small-N case studies. To synthesize the available case-based evidence, we conducted a broad case-based meta-analysis across 22 Western democracies, including 305 individual cases of public environmental decision-making. We asked: How do 'more' participatory decision-making processes compare against 'less' participatory ones in fostering - or hindering - strong environmental governance outputs, (i.e. environmental provisions in plans, agreements or permits)? Which design features make a difference? What role does the decision-making context play? How do results change if we control for the intentions of the leading governmental agency? To capture the central design features of decision-making processes, we distinguish three dimensions of participation: the intensity of communication among participants and process organizers; the extent to which participants can shape decisions ("power delegation"); and the extent to which different stakeholder groups are represented. Our regression analysis yields robust evidence that these three design features of participation impact upon the environmental standard of governance outputs, even when controlling for the goals of governmental agencies. Power delegation is shown to be the most stable predictor of strong environmental outputs. However, communication intensity only predicts the conservation-related standard of outputs, but not the environmental health-related standard of outputs. Participants' environmental stance was another strong predictor, with considerable variation across different contexts. While our results remain broadly stable across a wide range of contexts, certain contextual conditions stood out in shaping the relation between participation and environmental outputs. Overall, our findings can inform the design of participatory processes that deliver governance outputs of a high environmental standard.

Human–Machine Systems Reliability: A Series–Parallel Approach for Evaluation and Improvement in the Field of Machine Tools

Machine workshops generate high scrap rates, causing non-compliance with timely delivery and high production costs. Due to their natural characteristics of a low volume, high-mix production batches, and serial and parallel configurations, generally the causes of their failure are not well documented. Thus, to reduce the scrap rate, and evaluate and improve their reliability, their system characteristics must be considered. Based on them, our proposed methodology allows us to evaluate the system, subsystem, and component–subsystem relationship by using either the Weibull and/or the exponential distribution. The strategy to improve the system performance includes reliability tools, expert interviews, cluster analysis, and root-cause analysis. In the application case, the failure sources were found to be mechanical and human errors. The component maintenance/setup, institutional conditions/attitude, and subsystem process/operation were the machine factors that presented the lowest reliability indices. The improved activities were monitored based on the Weibull β and η parameters that affect the system reliability. Finally, by using a life–effort analysis, and the method of comparative analysis of two sequential periods, we identified the causes that generated a change in the Weibull parameters. The contribution of this methodology lies in the grouping of the tools in the proposed application context.