Since you asked: Can our recent series of reports about where incarcerated people call home be adapted to talk about other populations?
The data can help researchers better understand the harms of mass incarceration.
by Peter Wagner and Emily Widra, June 30, 2022
Now that we’ve started publishing a series of reports about where incarcerated people call home in each of the states that have ended prison gerrymandering, that raises the common question: Can you also do this for other populations under criminal justice system control, such as people on probation or parole? This comes up in a number of contexts, but to pick one example, someone might be running a campaign to change a state law that denies people on parole the right to vote, and they might want to make the numerical impact a little more clear for specific legislators who might support or oppose the legislation.
The short answer is: Because this data is about people in prison — not people on probation or parole — it can’t precisely answer this question, but it can illuminate the patterns and scale of correctional control in an area.
A simple strategy to apply our data to other populations — and one that might be enough for your purposes or a proof of concept for more in-depth research1 later — would be to make the relatively safe assumption that parolees (for example) are spatially distributed across the state in a way that is roughly similar to the distribution of incarcerated people. (If you have reasons to believe that the distribution is different, then this of course won’t work.)
In order to estimate the number of parolees in a particular geography, all you would need to know is the state-level ratio of parolees (or another relevant population) to incarcerated people and then apply that ratio to each geography in our report. For example, in New York State there were 44,917 people2 on parole in 2020 and our report says that 39,027 people were reallocated3 to their home addresses from state prisons that year, for a ratio of 115%. You simply need to multiply the number of incarcerated people in each legislative district, neighborhood, etc. by that ratio to get a fair estimate of the number of parolees in that area. A similar approach could be used for people on probation or any other population that you have good reason to believe is distributed similarly to people in prison.
If you find this general methodology helpful, here are two other suggestions and warnings as you proceed:
- Multiplying whole numbers of people by a percentage will almost surely result in fractional people. You’ll need to decide whether to round the numbers to keep this simpler or to keep the fraction as a way to emphasize that it’s an estimate. Both choices are fine and just depend on your intended use.
- As your geographies get smaller, the accuracy of a state-level ratio between parolees and the incarcerated people will become less accurate. How small is too small is a value judgement you’ll need to make based on your knowledge of your state and your own goals, but if you wanted a starting point, in our series of reports we generally didn’t rely on or highlight incarceration rate data where the total number of incarcerated people in that geography was less than 10.
Footnotes
-
The far more precise, although far more complicated way to answer this question would be to develop a relationship with a parole or probation agency, determine if they collect home address data, convince them to share that list of home addresses with other researchers under appropriate privacy protections, and then have the researchers map all of those addresses and then aggregate them up to the various geographies of interest to you. ↩
-
As of January 1, 2020. See Probation & Parole in the U.S., 2020, BJS. Appendix Table 9. ↩
-
This number is not, as our report methodology explains, the total population of the New York State prison system. For this conversion to work you want to compare the total parole population to the number of people reallocated. ↩