An Indicator Function for Insufficient Data Quality - A Contribution to Data Accuracy

Owing to the fact that insufficient data quality usually leads to wrong decisions and high costs, managing data quality is a prerequisite for the successful execution of business and decision processes. An economics-driven management of data quality is in need of efficient measurement procedures, which allow for a predominantly automated identification of poor data quality. Against this background the paper investigates how metrics for the DQ dimensions completeness, validity, and currency can be aggregated to derive an indicator for accuracy. Therefore existing approaches to measure these dimensions are analyzed in order to make explicit, which metric addresses which aspect of data quality. Based on this analysis, an indicator function is designed returning a measure for accuracy on different levels of a data resource. The indicator function’s applicability is demonstrated using a customer database example.