Evaluating Lubbock County’s Behavioral Therapy Program to Reduce Violence Among Inmates

The “Step Up” program teaches individuals how to manage their emotional responses and specifically teaches the skill of not resorting to violence as a means of resolving conflict. The underlying philosophy of the program is that individuals routinely resort to violence due to not having developed practical coping skills when confronted with emotional responses. If given better tools for coping with their emotional responses, individuals could be more likely in the future to put a stop to violence, sustain employment, and avoid encounters with law enforcement.

Once someone is selected for Step Up, they will be transferred to a designated Step Up residential “pod” within LCDC. Once transferred, individuals participate in roughly 15 hours of programming per week, with an optional hour of individual counseling available. Participants stay in the program for an average of about 40 weeks, typically until they leave LCDC. A sample of the courses is listed below:

Conflict Resolution -- Learning communication skills that will de-escalate heated situations and prevent potential physical altercations
“Emotion 2.0” -- Developing the capacity to recognize their own and others’ emotions, distinguish between different feelings and use emotional information to guide behaviors
Anger Management -- Learning to recognize signs of anger before they get out of hand
“Therapeutic Writing” -- Designed to help those that struggle with expressing themselves through usual means of communication

2022-02-07

2024-02-05

Key outcome areas include in-custody behavior, recidivism, and workforce success. These areas could be captured by data on disciplinary infractions in custody, re-arrests, and employment, respectively. For more details about capturing these outcome areas of interest, see the attached pre-analysis plan.

The RCT design for this project will be integrated into LCDC’s existing operations, taking advantage of excess demand for the program and its waitlist system. Within their typical referral and enrollment process, the research team will introduce a lottery for individuals on the waitlist who are eligible and interested in joining Step Up.

As spots in the program become available, individuals will be randomly selected for participation in the program from those who are next available to fill these spots on the waiting list. To conduct randomization, LCDC staff will send an anonymous waiting list each week to the research team, including both those who did and did not consent to research, and the number of open spots noting any exclusionary restrictions for each, in the order that they became available. For each new available spot, the research team will conduct pairwise randomization of the two eligible individuals who have been on the waitlist the longest, excluding anyone who violates specific restrictions for the open spot.

Sometimes capacity constraints and safety concerns require temporary exclusions from the waiting list. Specifically, LCDC must take into account (1) a Gang Intelligence screen that monitors the group composition of Step Up, (2)“keep separate” orders that restrict who a person can come in contact with, and (3) “bed restrictions” that allow for specific sleeping accommodations (e.g., a person that cannot be bunked because of a physical disability). For example, suppose a bed becomes available in a room currently occupied by individual A. Individual B is one of the next available participants on the waiting list, but has a “keep-separate” order with individual A. In this case, Individual B will be passed over and the next available person on the waiting list will be used in the randomization instead. Individual B would be included in the next randomization for a spot that matches their specific requirements.

Not available

Randomization done in office by computer. Using Stata, we will enter a randomly selected seed number between 1 and 999999999. We then will draw from a uniform distribution between 0 and 1. If the drawn number is at or below 0.5, the individual at the top of the list (on the list the longest) will be assigned to the Step Up program; otherwise they will exit the waitlist and return to business-as-usual services at LCDC. The other individual in the pair will then enter the opposite group, with a 50 percent chance of being selected for Step Up.

Individual

No

600 individuals (see below, not clustered)

600 individuals

300 individuals to control
300 individuals to treatment