A test of the racial contact hypothesis from a natural experiment: Baseball's All-Star voting as a case

Objective. The contact hypothesis is difficult to test because of selectivi ty bias: the direction of causation between contact and attitudes cannot be definitively determined. Selectivity bias can be avoided given a sample fo r which, for reasons unrelated to group member attitudes, whites divide int o groups that differ systematically in the amount of contact they have with blacks. This paper uses a natural experiment that provides such a test: Al l-Star voting by baseball players versus fans. The average white player has more substantial and persistent contact with African Americans (as team ma tes) than does the average white fan, and the contact hypothesis would thus predict that players will discriminate less than fans, all else equal. Met hods. Logistic, ordinary-least-squares (OLS), and tobit regressions are use d to analyze the effect of candidate race on votes received, controlling fo r other factors. Results. No evidence is Found of a differential in discrim ination related ro the amount of contact, although discrimination by both g roups is found to decline between 1970 and 1980, Conclusion. The contact hy pothesis is not supported; exogenous variations in the amount of interracia l contact are nor associated with difference in levels of discrimination.