Longitudinal changes in consensus as a function of acquaintance and agreement in liking.

By using a round-robin design, groups of freshmen reported their impressions of dormmates at 4 different times during the year. Consensus on W. T. Norman's Big Five (1963) did not increase over the year, even though reported acquaintance did increase. Agreement in liking predicted agreement in the trait ratings, such that groups that agreed in their evaluations of one another were more likely to agree in their trait impressions of one another, and vice versa. These patterns, obtained by using trait ratings, were replicated with open-ended descriptions of the targets. Correlations with self-judgments did not increase with acquaintance. The data suggest that increased acquaintance does not produce increased consensus. Instead, the role of agreement in evaluations is explored as a determining factor in the level of obtained consensus.