Leymon 19
imprisonment rates with controls for percent African American and percent Hispanic. The results reveal
that “fixed” sentencing reforms are significantly and positively related to greater increases in
imprisonment rates for blacks, whites, and Hispanics in all three models, with the exception of
determinate sentencing among the African American imprisonment rate. This indicates that there is a
relationship between states that have implemented sentencing reforms and the increase imprisonment
rates for blacks, whites and Hispanics.
Table 1 About Here
Along with the logarithmic growth curve variables for “fixed” sentencing reforms, all three models also
included controls for percent African American and Hispanic. The results of all three models indicated
that percent Hispanic is positively associated with higher rates of imprisonment across all racial/ethnic
groups. Conversely, percent African American was negatively associated with rates of incarceration
across all three racial/ethnic groups. The negative significant effect of percent black on the black
imprisonment rate (Model 1) is unexpected.
18
The explanation for the negative association between
percent African American and the white and Hispanic imprisonment rate is also unexpected.
It should be noted that Table 2 represents analysis of three separate dependent variables. The
different dependent variables do not allow for direct comparisons between the results of the different
racial/ethnic groups as presented here. The larger coefficients for African Americans (for each of the
variables) over both whites and Hispanics and the larger coefficients for Hispanics than whites are at least
partly a result of differences in the dependent variables.
19
For example, African Americans and Hispanics
18
For example, while the South has the highest percent African American of all the regions in the United States, the South
does not necessarily have the highest rates of black imprisonment per 1000 blacks. A number of states outside the South
have extremely high rates of African Americans within their correctional systems, but have a relatively low percent
African Americans in their state. Furthermore, the majority of the highest per-capita rates of black imprisonment are not
in the states with the highest percentage of African-Americans.
19
The difference could also be partly caused by a difference in the effects of any one of the independent variables on the
dependent variable, but in the analysis presented in table 2 it is impossible to “tease out” the differential effects.
Convention