Visualizing Data Quality Metadata for Decision Support: A Prototype and Evaluation

Data quality (DQ) metadata is the set of quality measurements associated with the data. Literature has demonstrated that the provision of DQ metadata can improve decision performance. However, it also showed that DQ metadata can overload decision-makers and negatively affect decision performance. In this paper, we examine a technique for reducing this overload by using visualization. Recognizing that visualization shifts the burden of absorbing DQ metadata to the perceptual capacity of the decision maker, we argue that it will reduce cognitive load. We theorize that decision-makers can hence absorb DQ metadata more easily. Decision performance will improve even when DQ metadata is provided. We describe the visual interface for visualizing data and DQ metadata and describe an experiment to test the impacts of visualizing DQ metadata on decision outcome. The results of this study offer insights for the design of decision support systems and the provision of DQ metadata.