Validation and risk and data … oh, my! Corrections Trends for 2020

Northpointe Suite   •   An equivant product

Validation and risk and data ... oh, my! Corrections Trends for 2020

Customers come to us for help with everything from managing change to working through reform issues to navigating ongoing resource challenges. As we work with teams across the country, we’re able to spot trends as they unfold, and we’re constantly learning more and applying what we learn to new situations.

In 2019, customers asked questions that shed light on the current state of corrections and the trends that are starting to trickle down to agencies of all sizes. Check out our Top 3:

Question 1: Should we be validating our risk assessment?

Answer: Yes! The risk assessment you’re using should always be validated on your local population. Validation is a careful process, and the data sample must be of sufficient size. Data availability often drives the timing of the validation work from jurisdiction to jurisdiction, and you should begin the process as soon as you can. Some agencies in the country really shine in this area, ensuring validation occurs on a regular basis so results are current and readily available.

Trend: No matter the size of your agency, assessment validation is a necessity, not a nice-to-have.

Question 2: Is the recommended supervision level the same as risk level?

Answer: No. Risk level is based on static factors, such as age and criminal history (historical or unchangeable factors). Supervision level is an output of the COMPAS Risk and Needs Assessment, and it is based on combining the predictive general and violent recidivism risk scores and plotting the results in a matrix. The result is the recommended supervision level.

Overriding a person’s supervision level does not change a person’s risk level. It takes time to truly change a person’s risk level based on treatment provided against the primary needs of that person.

Trend: Risk assessments are an essential part of case planning, and it’s more critical now than ever before for practitioners to understand the nuances of the tools.

Question 3: Can we customize the length of our assessments?

Answer: Absolutely. The number of questions included is configurable – this means that different sets of questions can be turned on or off, depending on what data you want to collect and what you are going to do with it. For example, an agency that wants to collect data on substance abuse would turn on the substance abuse questions, and they would be incorporated in the final assessment design. With a substance abuse focus, other question sets can be turned off because they’re not essential to achieving the agency’s goals.

Trend: Data and measurement needs vary by community, and it’s becoming increasingly important for agencies to be able to collect the specific data they need with minimal time and effort.

What was your agency’s biggest question of 2019? What are the local trends you’re seeing? We’d love to hear more! Give us a call anytime to talk through your latest challenge or fill us in on your latest win. We’re listening and learning every day.