Analyzing Contribution Patterns of Volunteered Geographic Point Features in Relation to Errors and Demographics.

摘要

This dissertation explores the relationship between the contribution patterns of volunteered geographic point features in relation to error and demographic properties. Recent research on Volunteered Geographic Information (VGI) has asserted that a correlation exists between population density and data quality. Others have shown that the relationship may be more complicated than population density alone. Within this research, an algorithm is developed to compare two datasets with each other to analyze the spatial accuracy and completeness. The algorithm is developed in Python so that it can be implemented as a tool within the ArcGIS framework. Datasets from the United States federal government and the volunteered geographic community are used to examine accuracy and completeness for schools within a study area in the Denver, Colorado area. In an effort to extend the research to include more points, the study area is then extended to include OpenStreetMap geographic point features across the state of Colorado. The larger dataset was used to conduct an analysis of the relationship between contribution patterns and demographic data. While this research failed to confirm the assertion by others that a relationship exists between data quality and demographics properties; however, this research furthers the understanding of patterns of volunteered geographic point feature contribution, error, and the relationship with demographics. Furthermore, analyses of the results of this research indicate that a relationship may exist that is more complicated than demographics alone and provides some suggestions for additional research areas that may be pursued to better understand the relationship.