Comparing Multiple Attribute System Selection and Social Choice Preference Aggregation

Comparison of information systems by evaluation of several specified criteria is a critical and often arduous process in organizational IT. The process is often done by compilation and consolidation of utility values using weights for the criteria. There is ample room for manipulation and misrepresentation of system aspects via the use of utility values and weights. In this work simulation results are presented which show that preference aggregation methods used in Social Choice can be applied to this problem, avoiding the use of utility values and weights altogether. Instead, only ordinally scaled expert judgment along the criteria is necessary, greatly facilitating the process, while in a large majority of simulated cases arriving at the same results.