Meta-analysis and comparative analysis: illustrations using perceived crowding

Abstract

Meta-analysis is a quantitative research synthesis that uses a specific quantitative measure (e.g., an effect size) to indicate the strength of relationship between variables for the studies making up the synthesis. Comparative analysis is another quantitative research synthesis technique where data (typically datasets) are aggregated over specific variables to reveal patterns in findings and identify factors that are not evident in a single study. Both analysis strategies (a) turn the focus away from one statistically significant result, (b) give credence to repeated results that are in the same direction across several studies, even if one or more are statistically insignificant, and (c) can assist researchers in making judgments about practical significance by encouraging examination of findings from multiple studies. For the field of human dimensions of natural resources, quantitative research synthesis can be useful for analyzing outdoor crowding research. Since the mid-1970's, perceived crowding has been measured using a single item 9-point scale. This single item measure has been used in approximately 187 studies involving 627 evaluation contexts. The dissertation contains three stand-alone papers. Chapter 2 provides an introduction to meta-analysis literature and discusses the challenges of applying meta-analysis to human dimensions research. Specifically, the various definitions of meta-analysis, the advantages and disadvantages, and the basic steps of meta-analysis are introduced and discussed. Chapter 3 expands on the methodological literature associated with the analysis step in meta-analysis, and provides an empirical example using a subset of the perceived crowding data. Chapter 4 builds on an earlier comparative analysis of perceived crowding by: (a) describing perceived crowding studies from 1975 to 2005, (b) evaluating alternative approaches to summarizing the 9-point crowding scale, (c) assessing generalized factors that influence crowding, and (d) exploring previously proposed crowding standards. Overall, this dissertation has attempted to describe meta-analysis and comparative analysis in the context of practical significance, and examine the potential of both quantitative synthesis strategies in the field of human dimensions of natural resources. Researchers are encouraged to further explore the use of meta-analysis and comparative analysis in other natural resource applications.