An Experimental Analysis of the Effect of Data Aggregation on Decision Making
Journal of Accounting and Finance Research
This study used an experiment to study user preference for disaggregated versus aggregated data (Sorter, 1969). One independent and five dependent variables were examined. The independent variable, a proxy for the level of data aggregation, was databases versus an income statement. The five dependent variables studied included anticipated performance of the decision maker, the decision maker's satisfaction with the management information system (MIS), decision confidence, decision accuracy, and decision time. Graduate students from six universities acted as managers in an experimental decision-making simulation. Subjects were required to make a decision regarding whether to keep or drop a product in a hypothetical chemical company. The results indicate that users of aggregated data anticipated better performance, were more accurate, and took less time to make their decisions. No difference in satisfaction with the MIS or decision confidence was found between those using the databases and those using the income statement.
George Schmelzle, Thomas E. Buttross, and Janet C. Papiernik (2004).
An Experimental Analysis of the Effect of Data Aggregation on Decision Making. Journal of Accounting and Finance Research.12 (3), 114-121.