ABSTRACT

The current study uses data from the reality television show Survivor to determine whether contestants discriminate against other contestants on the basis of sex. In Survivor, contestants initially compete in teams, and are strategically incentivised to eliminate weaker contestants to boost the performance of their team. Later in the season, teams are merged and the contestants compete individually. Stronger players then become a threat, and so strategic incentives encourage contestants to eliminate stronger players. Voting patterns allowed inferences to be made about taste-based and information-based theories of discrimination. Taste-based discrimination predicted that targeted groups would receive high mean votes both before teams are merged and post-merge. Likewise, a preference for in-group members would result in voters voting for members of the opposite sex both pre- and post-merge. Neither sex showed a tendency to vote consistently for contestants of the opposite sex. Information-based discrimination predicted high mean votes pre-merge and low mean-votes post-merge, and both sexes showed voting behaviour consistent with information-based discrimination against female contestants.