Regression analysis showed that when the percentage of urban population is considered alone, it statistically accounts for a moderate 16.9 percent of the variation in imprisonment rates (significant at .0001). When the urban percentage is considered in combination with the crime rate for 57 counties, it disappears from the regression equation. This means that for these counties, the crime rate is a more powerful variable than the urban percentage. It further indicates a strong relationship between crime rate and urbanization. Thus, the urban percentage affects the crime rate, but the crime rate has a more direct causal effect on the imprisonment rate. Tabular data and one footnote are provided. For the main report, see NCJ 82306.