Advancing Toward Spatial Data Quality Assurance
"I didn't know it existed"; "I couldn't get access to it"; and "It's not compatible with my system" are barriers that are falling away as implementations of the OGC's web catalogue, map, feature and coverage service standards become widely used. Another barrier, "I don't trust it", remains.
To know the quality of data, it is essential that assertions about the quality of the data are expressed in a standard and comparable way and that there is common knowledge of the measures used. ISO 19113, 19114 and 19138 standards (which are soon to be superseded by the new ISO 19157) provide principles for describing the quality of geographic data. They offer a consistent way to encode and report a dataset's quality information, with guidelines for evaluation procedures of quantitative quality information. Furthermore, ISO 19158 aims to provide a quality assurance framework for the producer and customer in their supplier-consumer supply chain relationship, to increase user confidence in the data that is provided.
The OGC Data Quality Domain Working Group (DQ DWG) is a global forum for producers and users of spatial data. The DQ DWG members seek to understand the use cases and describe a practical and open web services framework that builds on the ISO standards.
The DQ DWG's 2007/2008 Data Quality Survey measured interest and understanding of DQ within the geospatial industry. The survey yielded almost 800 responses from 107 countries, ranging from small to very large organisations, both public and private. Nearly all identified DQ as important, but almost half had no clear approach to managing it. A majority of those responding were willing to pay more for higher quality data in their projects, if they could just be sure that the quality was there. Yet there was great variety in how data quality was measured. So for most organisations, DQ appears to be something that is important, but difficult to quantify.
The group provides an opportunity to share experiences on the importance of good quality data (and the impact of poor quality data) for business domains such as national and international mapping, oil and gas, aviation and earth observation. Members
- Discuss best practices for monitoring and reporting on the levels of accuracy and uncertainty in relation to ISO 19157
- Collaborate with projects like GeoViQua and the European Location Framework (ELF), which are (respectively) developing concepts for visually informing data users about their data's usability and providing cloud based services for data providers to assess data quality using commonly agreed rule sets to improve cross-dataset consistency.
- Investigate approaches to improving currency and data completeness
- Share use cases on data quality approaches
Devising successful data quality interface and encoding standards requires such discussion among diverse users and providers of data. It is necessary to have a shared understanding of the role of DQ control all along the information supply chain, from change notification through data update to product generation, publishing and consumption. The standards emerging from this effort will enable knowledge of DQ levels to inform business intelligence that enables organisations to assess and mitigate the risk to their operations and future strategies.
Matthew Beare and Patrick Cunningham are co-chairs of the Data Quality Domain Working Group.
Principle Consultant, 1Spatial
President, Blue Marble Geographics
A version of this column appeared in GeoConnexion March 2012.