Abstract

Nowadays, data quality still becomes issue that is faced by organizations. As time goes by, organizations must deal with a lot of data that have been produced and several of data sources. Of course, it will be a challenge for organizations to manage their data in order to get high quality, so it allows improvement in decision-making process. Based on United Nations Statistics Quality Assurance Framework (UN-SQAF), good quality outputs are statistics fit for purpose from the user perspective, more specifically meaning that they are relevant, accurate, reliable, coherent, timely, accessible, and interpretable. To produce statistical outputs, BPS-statistics Indonesia implements quality gates to assure the statistical production phases comply with the standard operating procedures. Although, results from quality gates are used as early warning, but deadlines pressure, non-response, lack of systems, and human resource capabilities lead to data quality problems. There is a list of risks and mitigations, but an overlooked component is continous improvement in statistical business processes to lower the risk and avoid repeating the same mistakes. This paper explains the position of the statistical office as a data supervisor in carrying out tasks to ensure data quality and what recommendations can be given to improve data quality in the future. Some of the points that will be discussed are how to document the quality of each statistical phases, evaluate reports, and make continuous improvements in priority aspects. We can only improve what we can measure, otherwise if we only measure and do not improve then there will be no improvement in data quality.