One of the first decisions a judicial officer makes after someone is arrested and taken into custody is whether to release or detain the person pending case resolution and, if the decision is to release, what conditions, if any, should be imposed.

Pretrial decisions have enormous consequences for both the individual accused of a crime and the greater community. People who are detained may incur enormous loss: spending just a few days in jail can cost a person their job, housing, and health care services, and significantly disrupt their family life. And studies show that people who are detained before trial are more likely to plead guilty, be convicted, and be arrested again at higher rates. On the flip side, releasing people who pose a danger to public safety may put a community in jeopardy.

The United States Supreme Court ruled that pretrial liberty is the norm and detention should be the carefully limited exception, and that the key factor for a judicial officer to consider when making these pretrial decisions is the likelihood that the person will not flee the jurisdiction and/or not pose a danger to others free while on pretrial release. Judicial officers should be given as much pertinent information as possible to help make these decisions fairly and accurately. A person’s inability to afford money bail amounts should also not determine whether they remain in jail or are released.

A growing number of jurisdictions are moving away from a reliance on money bail and providing their judicial officers with more comprehensive information to support their pretrial decisions. Often as part of a wider effort to improve their pretrial policies and practices, many jurisdictions incorporate an actuarial assessment into their pretrial decision-making processes.

The Public Safety Assessment (PSA) was designed by Arnold Ventures to provide judicial officers with information to help them assess a person’s likelihood of returning to court for future hearings and remaining crime free while on pretrial release.

Since its development in 2013, the PSA has been implemented in dozens of jurisdictions around the country. Based on data related to nine different factors, the PSA generates scores that predict three pretrial outcomes: failure to appear in court pretrial, new criminal arrest while on pretrial release, and new violent criminal arrest while on pretrial release. There are many features that make the PSA unique among pretrial assessments:

  • Nationally validated. The PSA was created using the largest, most diverse set of pretrial records ever assembled—approximately 750,000 cases from roughly 300 jurisdictions across the United States. Researchers analyzed the data to determine which factors were most predictive of failure to appear in court pretrial, new criminal arrest while on pretrial release, and new violent criminal arrest while on pretrial release. After its development, the PSA was validated using a dataset of different cases—over 500,000 cases from multiple jurisdictions. Since then, it has been re-validated in the State of Kentucky and Mecklenburg County, North Carolina, among other locations.
  • Predictive, objective factors. The research team that developed the PSA identified and tested hundreds of factors. Ultimately, the team isolated the nine factors that most effectively predicted failure to appear in court pretrial, new criminal arrest while on pretrial release, and new violent criminal arrest while on pretrial release. The factors include the person’s current age, prior convictions, pending charges, and prior failures to appear in court pretrial. Factors such as drug and alcohol use, mental health, employment, and residence were excluded because they did not increase the PSA’s predictive accuracy.
  • Accessibility. Unlike many other pretrial assessments, the PSA can be scored without interviewing the defendant. All nine factors are drawn from historical criminal records, eliminating subjectivity from influencing the assessment score. The nine PSA factors, weights, and method of calculation are fully transparent and publicly available. Jurisdictions are encouraged to make an individual’s PSA scores available to the person charged as well as to defense counsel and prosecution. The PSA is available to jurisdictions at no cost.
  • Evaluation. Arnold Ventures, which funded the development of the PSA, engages independent researchers to continuously subject the PSA to rigorous evaluation. Independent evaluators are validating the PSA in jurisdictions across the country to maximize predictive accuracy and minimize disparities. The results of the studies completed to date demonstrate that the assessment is predictive across different jurisdictions and, in combination with additional system improvements, is often associated with decreases in the use of money bail and increases in pretrial release rates. Further, these studies indicate that in jurisdictions where pretrial release rates have increased, new criminal arrests and missed court appointments have not increased. All studies to date have shown the PSA does not exacerbate racial disparities.

A person assessed using the PSA receives a score ranging from 1 to 6 on two separate scales – new criminal activity (meaning a new arrest) and failure to appear in court (with 1 signifying a greater likelihood of pretrial success and 6 signifying a greater risk of pretrial failure). The PSA also flags whether a person has an elevated likelihood of being arrested for a new violent crime if released during the pretrial period.

The PSA does not direct a judicial officer to release or detain a person or decide any conditions of release. To help judicial officers make use of the PSA scores in their pretrial decision making, local stakeholders develop policy frameworks (the Decision Framework and Release Conditions Matrix) that reflect local statutes, court rules, and policy preferences. The Decision Framework is designed to support justice professionals in making decisions that align with the law, and the Release Conditions Matrix is structured to help judicial officers associate the PSA scores with the appropriate supports that may benefit a person while on pretrial release. Ultimately, the PSA and these policy frameworks can help a jurisdiction achieve its pretrial justice goals and assist judicial officers in making more informed, more consistent, and fairer decisions for the person before them.

Implement the PSA