Up to now, the research on urbanism has found only modest effects of urbanism on social psychological characteristics of residents. However, there is a possibility that previous researches underestimated urban effects because previous analyses have two problems. The first is only few attempts have so far been made at effects other than population size. The second is little attention has been paid to variances between places of residence and not individuals.
    So this research re-examined urban effects on unconventionality using multilevel model with individual-level data, which have been linked with aggregate-level data. By analysis of multilevel model, I obtained three findings. (1) There are significant variances of unconventionality between places of residence, independent of the individual attributes of residents such as socioeconomic status. (2) Consistent with Fischer's view, population size and residential mobility have significant positive effects on unconventionality. (3) Urban effects on personality may be stronger than previously thought. These results warrant further attention to urban effects on personality.