Examining the Predictive Validity of the Public Safety Assessment

• Author: Brian J. Brittain,Jim Martin,Leah Georges
• DOI: 10.1177/00938548211005836
• Published date: 01 October 2021
• Date: 01 October 2021
• Subject Matter: Articles

CRIMINAL JUSTICE AND BEHAVIOR, 2021, Vol. 48, No. 10, October 2021, 1431 –1449.

DOI: https://doi.org/10.1177/00938548211005836

Article reuse guidelines: sagepub.com/journals-permissions

© 2021 International Association for Correctional and Forensic Psychology

1431

EXAMINING THE PREDICTIVE VALIDITY OF

THE PUBLIC SAFETY ASSESSMENT

BRIAN J. BRITTAIN

LEAH GEORGES

JIM MARTIN

Creighton University

The purpose of this study was to examine the predictive validity of the Public Safety Assessment (PSA), an actuarial

pretrial assessment instrument, administered to 15,931 individuals in Volusia County, Florida, between 2016 and 2017.

A series of logistic regression models analyzed the influence of the PSA’s risk scores for Failure to Appear (FTA) and

New Criminal Activity (NCA), as well as gender, race, and the length of time spent in pretrial custody on incidents of

failure to appear and new pretrial arrest. The findings suggest that while both the FTA and NCA scales predicted pretrial

failure fairly well, the variation explained by the models suggest that there is much that we do not understand about

predicting pretrial failure to appear and new pretrial arrest, indicating the need for further research and refinement of

pretrial assessment instruments.

Keywords: actuarial risk assessment; pretrial risk assessment; Public Safety Assessment

Approximately 60% of America’s jail population consists of people in pretrial status

(Minton & Zhen, 2015). The decision to retain or release individuals before trial has a

significant impact on the outcome of an individual’s case and their future criminality

(“Developing a National Model for Pretrial Risk Assessment,” 2013). Furthermore, indi-

viduals who were detained for the entire pretrial process were 4 times more likely to be

sentenced to jail and 3 times more likely to be sentenced to prison when compared with

people who were released at some point during the pretrial stage (“Pretrial Criminal Justice

Research,” 2013). Sacks and Ackerman (2012) discussed the powerful adverse effects that

being incarcerated during the pretrial stage can have on a person’s family, living situation,

employment, and their ability to mount a sufficient legal defense. In addition, approxi-

mately 90% of individuals who remain in custody during the pretrial stage are detained in

American jails not because they are (a) dangerous, (b) pose a threat to the community, or

(c) have a history of failing to appear for court, but because of their inability to afford their

bond amount (Pretrial Justice Institute, 2013).

Actuarial pretrial assessment instruments are designed to place people into categories

based on a set of factors that assess the probability they pose to commit new crimes or fail

AUTHORS’ NOTE: Correspondence concerning this article should be addressed to Brian J. Brittain, 1502

Bent Oaks Blvd., Deland, FL 32724, USA; e-mail: bbrittainedd@gmail.com.

1005836CJBXXX10.1177/00938548211005836Criminal Justice and BehaviorBrittain et al. / EXAMINING THE PSA

research-article2021

1432 CRIMINAL JUSTICE AND BEHAVIOR

to appear for court during the pretrial stage of their case. While the idea of pretrial assess-

ment is well established in the field of criminal justice (McElroy, 2011; Vera Institute of

Justice, 2011), researchers most often reference probation, sentencing, sex offender, and

violence assessments when discussing pretrial assessment instruments. However, research

on the post-convicted population should not be considered interchangeable with research on

the pretrial population (Steadman et al., 2000; Stevenson, 2017). While pretrial assessment-

specific research progress has been made (DeMichele et al., 2020), more is needed. The

purpose of this study was to explore the predictive validity of one such instrument, the

Public Safety Assessment (PSA), by examining 2 years of jail and PSA data from Volusia

County, Florida (“Developing a National Model for Pretrial Risk Assessment,” 2013).

THE PSA

In 2013, Kentucky began piloting a new actuarial pretrial assessment instrument known

as the PSA (“Developing a National Model for Pretrial Risk Assessment,” 2013). The PSA

was designed to assist judges around the country with making more informed pretrial

release decisions by relying on data-driven analytics instead of limited criminal history

information and intuition alone (“Developing a National Model for Pretrial Risk

Assessment,” 2013). Using a data set composed of 746,525 cases from more than 300 juris-

dictions across the United States, Arnold Ventures examined risk factors that were most

indicative of pretrial failure to appear, defined as missing a scheduled court date, and new

offense, defined as being arrested for a new charge while in pretrial status. After testing

hundreds of risk factors, Arnold Ventures identified nine risk factors that were deemed most

predictive of pretrial failure to appear and new pretrial crime: (a) age at current arrest,

(b) current violent offense, (c) pending charge at the time of arrest, (d) prior misdemeanor

conviction, (e) prior felony conviction, (f) prior violent conviction, (g) prior failure to

appear in the past 2 years, (h) prior failure to appear older than 2 years, and (i) prior sen-

tence to incarceration (“Developing a National Model for Pretrial Risk Assessment,” 2013).

While Arnold Ventures (“New Data From Ohio Validates PSA Impact,” 2016; “Pretrial

Criminal Justice Research,” 2013) has produced several papers demonstrating the success

of the PSA, there remains a need for rigorous third-party examination of the PSA’s predic-

tive validity (“Pretrial Risk Assessment Now Available to All Interested Jurisdictions,”

2018). In addition, Arnold Ventures (“Pretrial Risk Assessment Now Available to All

Interested Jurisdictions,” 2018) has reported that more than 600 jurisdictions have expressed

interest in adopting the PSA. Thus, there is a growing need for research and examination in

jurisdictions where the tool is administered, ensuring that the PSA is operating as intended

and without bias.

THE UTILITY OF PRETRIAL RISK ASSESSMENT INSTRUMENTS

Despite being discussed since the 1920s, pretrial risk assessment practices and related

assessment instruments are still in their infancy and have only recently begun to gain

momentum (Klingele, 2015; Schnacke, 2014a, 2014b; Stevenson, 2017). As defined by

Douglas and Kropp (2002), actuarial risk assessment instruments are a type of structured

assessment, consisting of objective and empirical criteria, designed to predict human behav-

ior during specified time frames while providing probabilities of the likelihood of future

events occurring. When discussing risk assessment instruments in correctional settings,