Value-driven data quality assessment

Techniques for assessing data quality along different dimensions have been discussed in the data quality management (DQM) literature. In recent years, researchers and practitioners have underscored the importance of contextual quality assessment, highlighting its contribution to decision-making. The current data quality measurement methods, however, are often derived from impartial data and system characteristics, disconnected from the business and decision-making context. This paper suggests that with the increased attention to the contextual aspects, there is a need to revise current data quality measurement methods and consider alternatives that better reflect contextual evaluation. As a step in this direction, this study develops content-based measurement methods for commonly-used quality dimensions: completeness, validity, accuracy, and currency. The measurements are based on Intrinsic Value, a conceptual measure of the business value that is associated with the evaluated data. Intrinsic value is used as a scaling factor that allows aggregation of quality measurements from the single data item to higher-level data collections. The proposed value-based quality measurement models are illustrated with a few examples and their implications for data management research and practice discussed.