Power and Preferences: An Experimental Approach

A methodological note on a weighted voting experiment

“…The proposal that first obtains as many votes as the quota or more is adopted as the allocation of points among the members, and then the round is terminated. The protocol used by Montero et al (2008) and Aleskerov et al (2009) requires (a) random role (RR) reassignment of each subject to player i ∈ N in each round and (b) multiple approval (MA) of proposals that subjects can make simultaneously at a time. Under MA, all of the subject's votes are cast for each proposal he or she approves.…”

“…Subjects were allowed to make as many proposals as they wished during each round, but they could make only one proposal at a time. We randomized the order of the players' appearance in the proposal-input table, following Aleskerov et al (2009), to eliminate the effect of a fixed order on the observations in this experiment. 8 After a proposal was made, it was immediately shown to the other members of the relevant subgroup in proposal-approval tables that appeared on the right-hand side of the screen.…”

“…Montero et al (2008) and Aleskerov et al (2009) conducted the experiments in a cooperative game environment in which no explicit structure as to how negotiations take place is specified for voters' decision on how to allocate a fixed amount of resources via weighted voting. 2 An empirical measure of "a posteriori" power of a voter was defined as the average payoff that the voter obtained in certain rounds of the experimental session.…”

“…3 The formation of MWCs is consistent with the "size principle" proposed by Riker (1962), but the main purpose of the experiment conducted by Montero et al (2008) was to test the "paradox of new members" i.e., adding a new member to a committee can increase the voting powers of some existing members, while reducing the powers of others. Aleskerov et al (2009) tested the effect of asymmetry regarding favorable coalition partners among committee members on their coalition formation and resource allocation.…”

“…Montero et al (2008) and Aleskerov et al (2009) required in their experimental protocol random reassignment of subjects' roles (RR) in each round of sessions and multiple approval (MA) of proposals that subjects can make simultaneously at a time. With small changes in protocol, however, Esposito et al (2012) observed that MWCs of three voters appeared more frequently than those in earlier research.…”