The availability heuristic: effects of fame and gender on the estimated frequency of male and female names.

In two experiments, Canadian undergraduates heard a list of 13 male names and 13 female names; then they estimated how many male and female names there seemed to be. In Experiment 1, the list consisted of 26 famous names or 26 nonfamous names. Both male and female participants gave similar estimates for the number of male and female names, contradicting hypotheses of a bias toward males or toward one's own gender. In Experiment 2, where the list contained names of famous men and nonfamous women or names of famous women and nonfamous men, participants gave higher estimates for the gender that was famous (effect size d = 0.78). This result confirmed Tversky and Kahneman's (1973) fame availability effect and showed it to be moderate to large in size.