Revisiting the Use of Total Incarceration Variables in Understanding Racial and Ethnic Disparities in Sentencing

Research on sentencing typically measures the sentence with either a dichotomous in/out variable that combines prison and jail sentences, or with a continuous variable of the length of the sentence. Holleran and Spohn called this into question, arguing that combining prison and jail into a “total incarceration” measure can lead to overly simplistic findings. We use data from Miami-Dade County, Florida ( N = 40,240) to replicate and extend Holleran and Spohn’s research, with additional model flexibility for both in/out and sentence length. We find that not only are there distinct disparities that would have been undetected using the total incarceration variable, but these disparities also vary across the sentence length distribution.