Public Safety Assessment Research Articles

How Long is Long Enough: Using Abbreviated Criminal Histories for Pretrial Assessment Instruments?

The debate over pretrial assessment instruments focuses largely on the potential for instruments to exacerbate racial disparities. This study examines the effects of using different recall periods for the criminal history items on the New Criminal Arrest (NCA) scale from the Public Safety Assessment. Using jail admission data from a large jurisdiction in the southeast of the US, we varied the recall period from 1 to 10 years for the criminal history items on the NCA scale and examined scoring differences and the predictive validity of the tool by race. The results showed that shorter criminal history recall periods could result in reductions in the proportion of individuals scored as high risk and no loss in validity. The current study demonstrates that shorter recall periods for criminal history items on pretrial instruments could potentially provide a practical strategy to mitigate disparities across groups, while maintaining predictive validity of the assessment.

Justice involvement prediction as individuals age: An age-graded evaluation of the public safety assessment.

Some scholars have criticized pretrial assessments for perpetuating racial bias in the criminal legal system by offering biased predictions of future legal system outcomes. Although these critiques have some empirical support, the scholarship has yet to examine the predictive validity and differential prediction of pretrial assessments across individuals by their age. Following the guidance of the life-course literature, the present study serves as the first age-graded evaluation of the Public Safety Assessment (PSA) focused on assessing whether the predictive validity and scoring predictions of the tool vary across the lifespan. We expected that the predictive validity of the PSA scores would vary across the lifespan, such that the PSA underpredicts for younger individuals and overpredicts for older individuals. The present study relied on pretrial information collected from 31,527 individuals during the Advancing Pretrial Policy and Research project. Logistic regression models were estimated to evaluate the differential prediction of the PSA for individuals ranging from 18 to 68 years of age. The results of bivariate models were used to produce area under the curve estimates at each year of age. The results of the present study provided some evidence that the PSA differentially predicted pretrial outcomes for individuals from 18 to 68 years of age. Specifically, the predictive validity of the New Criminal Arrest and the New Violent Criminal Arrest scales appears to improve as individuals become older, suggesting that these instruments are better able to predict pretrial outcomes for older individuals relative to younger individuals. The results suggest that the PSA is a valid predictor of pretrial outcomes and that the predictive validity of some PSA scores is impacted by age. These findings suggest that the age of the defendant should be accounted for when interpreting the new criminal arrest and new violent criminal arrest scores. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Testing Predictive Biases at the Intersection of Race-Ethnicity and Sex: A Multi-Site Evaluation of a Pretrial Risk Assessment Tool

Pretrial assessment instruments provide courtroom actors with information about the likelihood that someone who has been charged with a criminal offense would engage in criminal behavior if released prior to trial. Although prior research supports the ability of pretrial instruments to predict pretrial outcomes, there are concerns that pretrial instruments may inadvertently exacerbate racial-ethnic and sex disparities found in the larger criminal legal system. In the current study, we conduct multi-site predictive bias tests of a widely used pretrial instrument—the Public Safety Assessment (PSA)—across six racial-ethnic and sex groups (i.e., White males, White females, Black males, Black females, Hispanic males, and Hispanic females). Study results support the PSA as a valid and consistent predictor of failure to appear, new criminal activity, and new violent criminal activity across these six racial-ethnic and sex groups. These findings support the use of the PSA and identify several areas for future research.

Sensible Pretrial Reform: The Importance of Accounting for Risk of Violence in Judicial Release Decisions

This study uses data collected from three diverse counties in the United States between 2017 and 2019 to compare the performance of a pretrial risk instrument—the Public Safety Assessment (or PSA)—to a revised version of the tool that accounts for risk of future violence in predicting three pretrial release outcomes. Results indicate the suggested three tiered release levels produced from the current PSA release conditions matrix (RCM) are better at predicting failure to appear in court, while those generated from the violence-informed RCM are better at predicting arrest for a new violent crime. The suggested release levels from both versions of the RCM were similarly able to distinguish between likelihood of a new arrest for any reason.

Supplemental Material for Justice Involvement Prediction as Individuals Age: An Age-Graded Evaluation of the Public Safety Assessment

Supplemental Material for Justice Involvement Prediction as Individuals Age: An Age-Graded Evaluation of the Public Safety Assessment

What Predicts Failure to Appear for Court Hearings?

Despite its importance, the pretrial process has not been subjected to the type of scrutiny directed at judges’ sentencing decisions. The goal of this project, which uses data on defendants released pretrial in four Arizona counties, was to identify the factors that predict failure to appear (FTA) for court hearings and to determine whether these predictors vary across defendants. We found that the likelihood of FTA was affected by a combination of defendant characteristics, case characteristics, and the jurisdiction where the case was adjudicated; the FTA prediction score, which is obtained from the risk assessment instrument used in all Arizona counties, also had a significant effect on the likelihood of FTA. Discussion focuses on inter-jurisdictional differences in FTA rates, the fact that the type of pretrial release did not predict FTA, and the fact that the defendant’s race/ethnicity affected the FTA prediction score and most of the Public Safety Assessment factors.

Victor Richmond Jose's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Victor Richmond Jose's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

How accurate are rebuttable presumptions of pretrial dangerousness? A natural experiment from New Mexico

AbstractIn New Mexico and many other jurisdictions, judges may detain defendants pretrial if the prosecutor proves, through clear and convincing evidence, that releasing them would pose a danger to the public. However, some policymakers argue that certain classes of defendants should have a “rebuttable presumption” of dangerousness, shifting the burden of proof to the defense. Using data on over 15,000 felony defendants who were released pretrial in a 4‐year period in New Mexico, we measure how many of them would have been detained by various presumptions, and what fraction of these defendants in fact posed a danger in the sense that they were charged with a new crime during pretrial supervision. We consider presumptions based on the current charge, past convictions, past failures to appear, past violations of conditions of release, and combinations of these drawn from recent legislative proposals. We find that for all these criteria, at most 8% of the defendants they identify are charged pretrial with a new violent crime (felony or misdemeanor), and at most 5% are charged with a new violent felony. The false‐positive rate, that is, the fraction of defendants these policies would detain who are not charged with any new crime pretrial, ranges from 71% to 90%. The broadest legislative proposals, such as detaining all defendants charged with a violent felony, are little more accurate than detaining a random sample of defendants released under the current system, and would jail 20 or more people to prevent a single violent felony. We also consider detention recommendations based on risk scores from the Arnold Public Safety Assessment (PSA). Among released defendants with the highest risk score and the “violence flag,” 7% are charged with a new violent felony and 71% are false positives. We conclude that these criteria for rebuttable presumptions do not accurately target dangerous defendants: they cast wide nets and recommend detention for many pretrial defendants who do not pose a danger to the public.

Eugenie Hunsicker's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Eugenie Hunsicker's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Authors' reply to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment.’

Authors' reply to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment.’

Kuldeep Kumar's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human-decision-making: application to pretrial public safety assessment’ by Imai et al.

Kuldeep Kumar's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human-decision-making: application to pretrial public safety assessment’ by Imai et al.

Pier Luigi Conti's contribution to the Discussion of ‘Experimental Evaluation of Algorithm-Assisted Human Decision-Making: Application to Pretrial Public Safety Assessment’ by Imai et al.

Pier Luigi Conti's contribution to the Discussion of ‘Experimental Evaluation of Algorithm-Assisted Human Decision-Making: Application to Pretrial Public Safety Assessment’ by Imai et al.

Efrén Cruz Cortés and Debashis Ghosh's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human-decision making: application to pretrial public safety assessment’ by Imai et al.

Efrén Cruz Cortés and Debashis Ghosh's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human-decision making: application to pretrial public safety assessment’ by Imai et al.

Mats Stensrud, Vanessa Didelez and Aaron Sarvet's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al., part 2

Mats Stensrud, Vanessa Didelez and Aaron Sarvet's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al., part 2

Jane Luise Hutton's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Jane Luise Hutton's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Proposer of the vote of thanks and contribution to the Discussion of ‘Experimental evaluation of algorithm- assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Proposer of the vote of thanks and contribution to the Discussion of ‘Experimental evaluation of algorithm- assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Aaron Sarvet and Mats Stensrud's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al., part 1

Aaron Sarvet and Mats Stensrud's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al., part 1

Peng Ding's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Peng Ding's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Tyler VanderWeele's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Tyler VanderWeele's contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Seconder of the vote of thanks and contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.

Seconder of the vote of thanks and contribution to the Discussion of ‘Experimental evaluation of algorithm-assisted human decision-making: application to pretrial public safety assessment’ by Imai et al.