| Aggregated quantitative information in the form of composite indicators facilitates the conceptualization, processing, reflection, and manipulation of the world surrounding us and our interactions with it. The number of environmental indicators, in particular, has grown considerably over the past three decades. Yet, the effective use of aggregated environmental data in environmental policy, management, and governance is diminished by our limited understanding of the effects of conceptual and methodological characteristics of composite indicators on their functionality. The present dissertation examines this knowledge gap along three distinct avenues. Consisting of three parts, the first part analyzes how structural elements of composite indicators influence the extent to winch they are useful information tools for environmental policy-making. The second part focuses on the problem of missing data and bridges the gap between existing statistical methods for analyzing missing data and their application (or lack thereof) in practice. In particular, based on a review of existing missing data methods and taxonomies, it generates a decision-tree for identifying appropriate missing data methods for use by practitioners and non-experts in the statistical analysis of missing data. In the third and final part, the dissertation introduces the concept of Desirability Indices, which originated in industrial process and quality control, to environmental performance measurement. Termed Environmental Desirability Index, the approach's adaptability to environmental contexts is tested in a simulation study of the effects of increasing levels of model complexity and uncertainty. |
