Online Program

Data Quality Assurance - Quality Gates Framework for Statistical Risk Management
*Narrisa Gilbert, Australian Bureau of Statistics 


Keywords: quality gates, statistical risk, statistical data quality, quality management, quality assurance

Statistical collections are often exposed to the risk that the components of the process fail to meet the quality standard expected, such that the quality of the statistical outputs are affected. We refer to this kind of risk as "statistical risk". Statistical risk arises for various reasons, some of which may include inadequate inputs, processes not being well defined, changes to existing processes, or human error.

This paper introduces an approach to managing the quality of statistical processes called quality gates. Quality gates are a risk mitigation tool designed to improve the early detection of errors or flaws in statistical processes. They improve a user's ability to manage statistical risk by providing explicit evidence relating to the statistical process, improving knowledge management and information sharing.

This paper will provide an explanation of the concepts underpinning quality gates so that they can be applied for statistical risk management. A discussion of some of the challenges that the Australian Bureau of Statistics has experienced in implementing quality gates will also be provided.