« Reactions to our look behind the robes with federal sentencing judges? | Main | What data in the federal system would indicate the Biden Administration is drawing down the federal drug war? »

November 25, 2020

A final (too brief) foray into what metrics and data matter for assessing a sentencing system

After our first few weeks of class discussing theories of punishment, you likely recall that we never reached any kind of firm conclusive resolution as to what goals a sentencing system must or should pursue.  For that reason (and others), it should not be a surprise that there is no firm conclusive view of what metrics or data matter most for judging a sentencing system or the criminal justice system more generally.  But this is not for lack of possible options, and here I will list some broad categories and sub-categories for you to consider as possible data points of greatest interest or concern for a sentencing system:


  • Overall crime rates and/or total crimes (per year or over an extended period)
  • Violent crime rates and/or total violent crimes
  • Murder rates and/or total murders


  • Overall total arrests (per year or over an extended period)
  • Arrest rates relative to crime rates
  • Arrests for certain crimes


  • Total convictions, felonies and misdemeanors (per year or over an extended period)
  • Conviction rates relative to crime rates or arrest rates
  • Conviction rates for certain crimes


  • Total sentences for many types of punishments (death, prison, jail, probation, fine)
  • Prison/jail sentences for more any crimes
  • More severe sentence of death, life or more than, say, 20 years in prison


  • Total persons in prison, jail. supervision at one snapshot in time
  • Total persons in prison, jail. supervision over a year or longer
  • Total persons in prison, jail. supervision for certain crimes


  • Persons arrested within one year (or 5 or 10 years)  after arrest
  • Persons sentenced for a new crime within one year (or 5 or 10 years) after release from custody
  • Persons convicted of a new serious crime within one year (or 5 or 10 years) after prior CJ contact


  • All persons ever subject to arrest/conviction
  • All persons who see or secure record relief from a prior conviction
  • All persons receiving longer sentences due to criminal past


  • Total expenditures by governments on entire criminal justice system
  • Total expenditures by governments on prisons and jails
  • Social/human costs (and benefits) from interactions with criminal justice system

I could go on and on, but I am sure you get the idea and that you can now reflect a bit on how many different possible sentencing "outputs" could  be a focal point for data collection, review and analysis.  This is the broad topic I am eager for us to cover in our last few classes, and I especially want to highlight that this long list of possibilities does not begin to engages various social justice issues — e.g., should we focus on "output" numbers in any or all of these categories particularly for people of color?  for women?  for juveniles?  for persons with mental illness?  for veterans?

We can perhaps start the discussion here in the comments, but know I will be asking you in our final classes to share your views on these issues through this particular question: What two of three metrics or data points should the incoming Biden Administration give special and sustained attention to in the coming months and years?

November 25, 2020 in Class activities, Data on sentencing, Who decides | Permalink


Never underestimate Yahoo as a source for diverse and on-point news!

Case in point, a decent article regarding Oklahoma and how it is quickly becoming the weed capital of the U.S. https://www.yahoo.com/news/one-reddest-states-became-nation-120904908.html

Free markets for all!

Posted by: Chris Wald | Nov 30, 2020 10:17:01 AM

Thanks, Chris. This was a good (and looooong) Politico article grabbed by Yahoo (https://www.politico.com/news/magazine/2020/11/27/toke-lahoma-cannabis-market-oklahoma-red-state-weed-legalization-437782). It is on-point for our class in some sense, as there are many hard questions concerning what data/metrics should matter in the marijuana reform space for folks interested in criminal justice reform -- e.g., should future arrests or expungements of past convictions matter more for reformers after legalization? In a few states, arrests for juveniles for marijuana possession went UP after marijuana legalization. Is this a big or small problem? Does the race and gender and age of the juveniles arrested matter for this assessment? And on and on....

Posted by: Doug B. | Nov 30, 2020 10:58:43 AM

After our conversation about crime rates a few weeks ago, I am inclined to stray away from a focus on overall crime rates and violent crime rates. Since the early 90s, we've seen those rates drop, and we've also seen periods of time where those drops are not at all related to increasing rates of arrests. Clearly the answer is not more arrests = safer community with less crime. Out of the remaining metrics and data points, I'd like the Biden Administration to focus instead on data related to sentences imposed and data on recidivism rates. I think these two areas are closely related, and by analyzing the intersection it's possible that we can address some of the failings of the system. Perhaps the data will show that we don't need long sentences for certain crimes to deter both generally and specifically.

Posted by: Shelby Mann | Dec 1, 2020 1:02:43 PM

In the press, with Trump doing pardons and commutations, there is some talk whether he will do any more than just for his few friends. This plays into his obsession with his own image. I would think he will do a bunch more so he can cover for his friends as well as look good for any re-election possibilities. If he only takes care of friends and himself that will not play well for reelection. I am just a curious observer in this matter and would like to hear what others think about this issue.

Posted by: Gary Josephson | Dec 1, 2020 4:00:37 PM

The comments to this entry are closed.