Document Type

Thesis

Degree Name

Master of Science (MSc)

Department

Geography & Environmental Studies

Program Name/Specialization

Geomatics

Faculty/School

Faculty of Arts

First Advisor

Dr. Colin Robertson

Advisor Role

Supervisor

Second Advisor

Dr. Rob Feick

Advisor Role

Member

Third Advisor

Dr. Trisalyn Nelson

Advisor Role

Outside Member

Abstract

Volunteered Geographic Information (VGI) is becoming a pervasive form of data within geographic academic research. VGI offers a relatively new form of data, one with both potential as a sensitive way to collect information about the world, and challenges associated with unknown and heterogeneous data quality. The lack of sampling control, variable expertise in data collection and handling, and limited control over data sources are significant research challenges. In this thesis, data quality of VGI is tackled as a general composite measure based on coverage of the dataset, the evenness in the density of data, and the relative evenness in contributors to a given dataset. A metric is formulated which measures these properties for VGI point pattern data. The utility of the metric for discriminating qualitatively different types of VGI is evaluated for different forms of VGI, based on a relative comparison framework. The metric is used to optimize both the spatial grains and spatial extents of several VGI study areas. General methods are created to support the assessment of data quality of VGI datasets at several spatial scales.

Convocation Year

2014

Share

COinS