Theory and Decision

, Volume 53, Issue 4, pp 343–369

Sophisticated Voting Under the Sequential Voting by Veto1

  • Fany Yuval

DOI: 10.1023/A:1024196415513

Cite this article as:
Yuval, F. Theory and Decision (2002) 53: 343. doi:10.1023/A:1024196415513


The research reported here was the first empirical examination of strategic voting under the Sequential Voting by Veto (SVV) voting procedure, proposed by Mueller (1978). According to this procedure, a sequence of n voters must select s out of s+m alternatives (m≥n≥2; s>0). Hence, the number of alternatives exceeds the number of participants by one (n+1). When the ith voter casts her vote, she vetoes the alternative against which a veto has not yet been cast, and the s remaining non-vetoed alternatives are elected. The SVV procedure invokes the minority principle, and it has advantages over all majoritarian procedures; this makes SVV a very desirable means for relatively small groups to make collective decisions. Felsenthal and Machover (1992) pointed out three models of voting under SVV: sincere, optimal, and canonical. The current research investigated, through laboratory experiments, which cognitive model better accounts for the voters' observed behavior and the likelihood of obtaining the optimal outcome as a function of the size of n (when s=1). The findings suggest that while voters under SVV use all three models, their choice is conditioned by group size. In the small groups (n=3), the canonical mode was a better predictor than the sincere model. In the larger groups (n=5), the sincere model was a better predictor than the canonical model. There is also evidence of players' learning during the experiment.

Minority principleStrategic votingVetoVoters' behavior

Copyright information

© Kluwer Academic Publishers 2002

Authors and Affiliations

  • Fany Yuval
    • 1
  1. 1.Department of Political ScienceUniversity of HaifaMount Carmel, HaifaIsrael