Statistical Risk Management
Statistical risk is the chance or likelihood of something going wrong with your statistical processes that affects the quality or integrity of your statistical outputs.
The following papers on this page help to provide information on some quality initiatives that may help you in managing the risks around your data.
Developing Data Quality Statements
Among national statistical agencies, quality is generally accepted as "fitness for purpose". Fitness for purpose implies an assessment of an output, with specific reference to its intended objectives or aims. The ABS recommends that when assessing the quality of a statistical collection or product, a Data Quality Statement be developed. A Data Quality Statement is a presentation of information about the quality of a statistical collection or product using the ABS Data Quality Framework.
Data Quality Statement: A checklist
Have you ever gotten to the end of your statistical process and realised you made a mistake right back at the beginning?
Quality gates can help you manage the quality of your statistical processes by helping you to identify any data problems closer in the process to when they occur. This allows you to fix the potential errors in a more timely manner.
ABS 2010, Information Paper: Quality Management of Statistical Processes Using Quality Gates, Dec 2010, cat. no. 1540.0, ABS, Canberra.
Gilbert, N 2011, “Quality Gates Framework for Statistical Risk Management”, South African Reserve Bank Research Department Seminar. ( PDF, 36.39 KB)
Schubert, P., et al. 2006, “Using Quality Measures to Manage Statistical Risks in Business Surveys”, Proceedings of Q2006 European Conference on Quality in Survey Statistics. ( PDF, 53.36 KB)
Quality Management of Administrative Data
Administrative data is useful for many statistical purposes. However, because of the nature of administrative data, there are some unique challenges and risks that need to be managed in order to assure the quality of statistical outputs. The paper below outlines common challenges and strategies for managing the acquisition of administrative data.
ABS 2011, Information Paper: Quality Management of Statistical Outputs Produced from Administrative Data, March 2011, cat. no. 1522.0, ABS, Canberra.
Statistical Quality Incident Response Plan
A quality incident occurs when the quality of the data are questioned. A Statistical Quality Incident Response Plan is a guide for resolving serious doubts about the statistical results by examining the processes that contributed to them. The Statistical Quality Incident Response Plan is a step-by-step guide to contingency planning; from discussing the issues in the first meeting; to investigating and identifying possible causes; through to resolving the issue and incorporating the lessons learnt from the situation into the normal day to day process.
ABS 2012, Information Paper: Statistical Quality Incident Response Plan, June 2012, cat. no. 1542.0, ABS, Canberra.