The Community for Technology Leaders
Green Image
<p><b>Abstract</b>—This paper describes an experiment that explores the consequences of providing information regarding the quality of data used in decision making. The subjects in the study were given three types of information about the data's quality: none, two-point ordinal, and interval scale. This information was made available to the subjects, along with the actual data. Two decision strategies were explored: conjunctive and weighted linear additive. Two decision environments were used: a simple environment and a relatively complex environment. Various combinations of these factors were employed to explore several issues. These include <it>complacency</it>, <it>consensus</it>, and <it>consistency</it>. The paper provides preliminary insights into which type of data-quality information is most effective and the circumstances in which data-quality information is most effective. Such knowledge would be of value to those responsible for designing databases that support decision-makers. Overall, we find that in a situation where subjects are confronted with clearly differentiated alternatives, the inclusion of data-quality information impacted the selection of the preferred alternative while maintaining group consensus.</p>
Data quality, data tagging, decision making, decision complacency, decision consensus, decision consistency.
Donald P. Ballou, Harold L. Pazer, InduShobha N. Chengalur-Smith, "The Impact of Data Quality Information on Decision Making: An Exploratory Analysis", IEEE Transactions on Knowledge & Data Engineering, vol. 11, no. , pp. 853-864, November/December 1999, doi:10.1109/69.824597
86 ms
(Ver 3.3 (11022016))