Empirical Evidence: What Judges Can Learn From Recent Social Science Research

Courts represent a critical moment that can influence a person’s trajectory through the criminal legal system,¹ and judges play a key role in shaping the courthouse environment and practices. Indeed, many judicial codes of conduct urge judges to establish the behavioral expectations to be followed by all who interact with the courts,² and research demonstrates the ability of judges to convince court professionals to adopt new policies and practices.³ Recognizing the power of judicial influence, agencies such as the Substance Abuse and Mental Health Services Administration (SAMHSA) suggest that judicial buy-in is crucial to successfully adopting new court practices.⁴

But how can judges decide which practices or policies to adopt or encourage within the court? One strategy is to look to the findings of social science research and consider the latest empirical evidence on the needs of diverse court-involved populations or the effectiveness of various practices in improving individual outcomes, case processing, and public safety.

In this article, we briefly summarize recent key findings across five areas of research. Two consider specific populations: indigent defendants and people experiencing frequent jail contact. The other three focus on practices: court reminders, court navigator programs, and pretrial risk assessments. We also share links to articles and other publicly available resources for those interested in learning more.

Key Research Finding 1:
Indigent Defendants Experience Worse Outcomes Than Defendants With Retained Counsel

Over two-thirds of defendants facing criminal charges in U.S. courts are deemed indigent by the courts (i.e., they cannot afford their own counsel and so the court provides one).⁵ A meta-analysis published in 2024 examined the findings of more than 40 studies across more than 2,400,000 cases comparing differences in pretrial outcomes, case outcomes, and sentencing outcomes for indigent defendants against those with retained attorneys.⁶ Results revealed that indigent defendants experience worse outcomes across all court processing stages than defendants with retained counsel. For example, indigent defendants were 75 percent less likely to be released pretrial, 50 percent more likely to have their case resolved by entering a guilty plea, 50 percent less likely to be acquitted, and 40 percent more likely to be convicted than defendants with retained counsel.⁷ Further, among defendants who were found guilty, indigent defendants were twice as likely to be sentenced to jail or prison as defendants with retained counsel.⁸ Outcomes were similar between indigent defendants with public defenders and those with assigned counsel, suggesting that the difference is not due to attorney performance.⁹ Instead, a person’s status as indigent may unfairly prejudice their outcomes, perhaps due to biases in court processes and decision-making.¹⁰

These findings highlight the need to make indigent defense systems more robust and to guard against bias in decision-making based on individual-level factors such as socioeconomic status. Judges could be instrumental in these efforts by conversing with indigent-defense attorneys to identify and help eliminate barriers they
may experience when representing clients. Judges could also encourage efforts to incorporate continuing education or training opportunities that address bias.¹¹

Key Research Finding 2:
People Experiencing Frequent Jail Contact Represent a Majority of All Jail Bookings

A second population that judges and court professionals frequently encounter is those who repeatedly cycle through the system and represent a disproportionate number of jail bookings — as many as half of all bookings annually.¹² Each unique jail contact requires processing by the courts, increasing the overall system burden. However, judges and court professionals may struggle to clearly identify the “frequent contact” group, its key characteristics, or effective strategies to address underlying needs and reduce the likelihood of repeat court involvement.

To address these issues, researchers partnered with three counties across the United States to understand the population of people who experience frequent jail contact.¹³ Researchers collected data from three sources: 1) county-specific documentation related to measuring and addressing frequent jail contact; 2) administrative court, jail, and police records; and 3) virtual and in-person site visits, including facility tours and interviews with practitioners and people with lived experience. Findings from this study revealed that, across all three counties, people with frequent jail contact represented a minority of the population of people booked into county jails — but a majority of all bookings.¹⁴ Further, people of color and people with mental health flags at booking consistently represented a disproportionately large percentage of the frequent jail contact population compared to both the county demographics overall and the total jail population.¹⁵ Interviews with practitioners and people with lived experience emphasized the importance of meeting basic needs, including housing support and behavioral health treatment, as well as relationship building to break cycles of frequent jail contact.¹⁶ Findings also underscored the need for local evaluations that consider jurisdiction-specific contexts and issues. To demonstrate, the definition of “frequent jail contact” differed across counties, as did the prevalence of mental health flags.¹⁷ Further, while all three counties had implemented behavioral health diversion strategies to reduce frequent jail contact, findings were mixed regarding their effectiveness.¹⁸

The research team developed nine policy recommendations to help communities address the needs of people experiencing frequent jail contact and, ultimately, to reduce such contact.¹⁹ These recommendations include creating a data-sharing ecosystem, establishing a formal definition of frequent contact, using validated behavioral health screening tools, implementing psychiatric advanced directives, facilitating jail in-reach programs, increasing peer-support programs, improving access to housing, increasing utilization of community-based services, and centering and evaluating efforts for racial equity. Judges can encourage the adoption of these recommendations by championing specific efforts or by lending their support to existing efforts in their community. Addressing the issue of frequent system contact requires collaboration beyond the courthouse to alleviate strain on the court and legal system.

Key Research Finding 3:
Court-Date Reminders Reduce Failures to Appear in Court But Are Not a Comprehensive Solution

Court-date reminders are often considered a strategy to improve case processing by reducing failures to appear in court. They involve sending postcards, text messages, or phone calls to remind people about the time and location of their upcoming hearing. Agencies such as the American Bar Association,²⁰ the National Institute of Corrections,²¹ and the National Association of Pretrial Services²² include reminders among their recommended best practices.

A meta-analysis published in 2022 tested the effectiveness of reminders by synthesizing the findings of studies that compared failure-to-appear rates between people who received a court-date reminder and those who did not.²³ It included findings across 12 studies, representing a combined 79,255 court-involved people.²⁴ The meta-analysis found that reminders lead to a modest reduction in rates of failure to appear in court: People who received a reminder were about 35 percent less likely to fail to appear in court compared to people who did not receive a reminder.²⁵ The meta-analysis also found that reminders appear to be slightly more effective for people with misdemeanor charges compared to those with felony charges,²⁶ likely due to the differing levels of pretrial supervision and support often provided for more serious charges.

From a policy perspective, these findings support the implementation of court-date reminder systems with realistic expectations about the anticipated reduction in failure-to-appear rates. A reminder system is fairly inexpensive and, once in place, requires minimal staffing. It also could decrease reliance on more restrictive pretrial conditions, such as electronic monitoring and even detention, by addressing logistical and informational barriers to court appearances. In this way, reminders align with broader pretrial reform efforts to reduce overreliance on jails through nonintrusive, supportive strategies. Court-date reminders, however, are not likely to reduce nonappearance related to intentional absences or structural challenges that may prevent a person from attending court, such as lack of transportation or childcare. Reminders must be coupled with additional strategies to address the full range of factors that contribute to court nonappearance.

Key Research Finding 4:
Court Navigators Improve Case Processing and Connect People to Behavioral Health Resources in the Community

In recent years, courts have begun to establish court navigator programs to help people successfully traverse their court involvement, connect with needed resources, and reduce their likelihood of further court involvement. What are these programs, and how do they work? To answer these questions, researchers conducted a national scan of programs,²⁷ ultimately identifying 18 court navigator programs in 21 states.²⁸ They then conducted site visits with five programs and completed nearly 100 interviews, including with court personnel, court navigators, service providers, and people who had received services.²⁹

Their findings revealed that court navigator programs are diverse in terms of structure and daily operations. This diversity is largely due to the local development of programs and the great degree of flexibility granted to navigators to respond individually to each person with whom they work. Despite the differences, all court navigator programs generally follow a similar process and provide similar services,³⁰ allowing court-involved people to meet with navigators, ask questions about the court process, and discuss their circumstances and needs. Following this initial discussion, a navigator might provide directions or information related to the courthouse or facilitate a connection to court- or community-based resources. When possible, navigators provide a warm handoff to resource providers.³¹ Some navigators also provide direct services such as crisis management or brief counseling sessions.³²

Evidence from the site visits and interviews suggested that navigators have been well-received.³³ In addition to supporting court-involved people, navigators also support legal-system actors — including judges. Specifically, navigators take over work that legal-system actors sometimes do, such as addressing questions related to social and behavioral health services, even though such work is outside the scope of their roles and responsibilities. Legal-system actors also valued being able to turn to navigators for additional resources, discussions of behavioral health factors in specific cases, and assistance in connecting with service providers.³⁴ Finally, legal actors appreciated the help that navigators provided to court-involved people, believing it improved overall court processing and could reduce future court involvement.³⁵

Key Research Finding 5:
Pretrial Risk Assessment–Informed Decisions Lead to Improved Outcomes

Pretrial risk assessments are designed to improve individual outcomes, public safety, and case processing. Such instruments have been used since the 1960s but have seen rapid uptake by courts across the United States in the past decade. Today, dozens of pretrial risk assessments are used in at least half of the most populous U.S. jurisdictions.³⁶ These instruments use factors legally allowed to inform pretrial decisions in order to provide an evidence-based estimate of a person’s risk of failing to appear in court or being arrested for a new crime during the pretrial period. Judges and other pretrial decision-makers can use these risk estimates as one piece of information to help guide their pretrial decisions, including decisions about bail and other conditions of release.³⁷

Some judges report skepticism about the accuracy of risk assessments³⁸ and choose to disregard or override the findings of risk assessments in their pretrial decisions.³⁹ However, recent evidence suggests that risk assessment scores perform well in predicting the outcomes they were intended to forecast and estimate risk more accurately than other practices that do not rely on risk assessments. Regarding the former, a 2021 meta-analysis of 13 studies found that pretrial risk assessment scores are generally effective in predicting the outcomes they aim to, such as failure to appear in court and new criminal activity during the pretrial period — both overall and across subgroups defined by race and gender.⁴⁰ Regarding the latter, a meta-analysis of 31 studies published in 2024 found that risk assessment instrument results are better at predicting the risk of violence or other offending than are human predictions made without the guidance of an instrument, even by experts.⁴¹ Other research has found that assigned bail amounts are not associated with whether a person fails to appear or is rearrested, but pretrial risk assessment scores are.⁴²

Further, despite concerns espoused by some, research has failed to produce strong evidence that the use of pretrial risk assessment instruments to inform decisions systematically contributes to racial or ethnic disparities within the legal system.⁴³ To demonstrate, a meta-analysis published in 2024 explored whether risk assessment instrument scores were applied differently to inform decisions across racial or ethnic groups to see if the same scores would result in harsher decisions for one group compared to another.⁴⁴ Nearly all of the studies found that the adoption of risk assessments either reduced disparities or at least did not create or worsen preexisting disparities.⁴⁵ Moreover, several rigorous research investigations demonstrate that the use of pretrial risk assessment instruments can aid jurisdictions in meeting reform goals and lead to reduced use of bail and lower rates of pretrial detention.⁴⁶ Notably, pretrial risk assessment instruments are typically very short and simple in nature, with an average of just 10 items that are coded and summed following a transparent process to produce risk scores.⁴⁷ Further, most widely used pretrial risk assessment instruments, including those examined in these studies, have published their item descriptors, rating guidelines, and algorithms online or in their user manuals.⁴⁸

Taken together, research supports using pretrial risk assessment instruments to inform court decision-making. Doing so can contribute to more consistent, transparent, less biased pretrial decision-making. At the same time, judges can champion and support three steps to ensure assessments are being applied appropriately. First, judges can encourage their jurisdictions to employ professionals to empirically validate any pretrial risk assessment instrument they adopt in their own jurisdiction before or shortly after implementation. Such validation would check whether the instrument results predict outcomes as expected and, if not, would suggest revisions to the instrument or alternative instruments that may be more well suited to the population and outcomes of interest. Jurisdictions should revalidate the instrument they use every five to ten years. Second, to increase transparency, judges can provide brief explanations when they override assessment recommendations. Third, they can record their override decisions against tool predictions and outcomes. Over time, knowing their override rates and reasons could help inform discussion about changes to the recommendation matrix that accompanies most pretrial risk assessment instruments. Fourth, and related, judges can lead regular reviews of the recommendations associated with risk assessment results and engage in discussions with other partners to adjust recommendations as needed.

Conclusion

Judges wield a great deal of influence. At the individual level, they can shape the tone of interactions within their courtrooms. At the system level, they can shape the policies and practices that are adopted throughout the courthouse and beyond. Social science research from fields such as psychology or criminology can provide judges with evidence to inform their case decisions and lead the implementation of evidence-based policies or practices. Recent surveys about the public opinion of judges’ roles and responsibilities indicate that people want judges to use research findings and evidence to inform their decisions and practice.⁴⁹ Applying current social science evidence in the courtroom may improve individual outcomes and court practices — and increase the public’s trust in the courts and judges.

Samantha A. Zottola is a senior research associate at Policy Research Associates, Inc. (PRA). She studies court and jail practices, the behavioral health outcomes of people involved with the criminal legal system, and pretrial reform efforts.

Isolynn A. Massey is a research assistant in research and evaluation at PRA. She uses qualitative research to understand the behavioral health needs of people involved in the justice system. 

Sarah L. Desmarais is president & CEO at PRA. Her current research is focused on evidence-based practices for reducing detention rates, especially among people with behavioral health needs.

Elan C. Hope is vice president of research and evaluation at PRA and is an expert in youth development. She uses community-centered quantitative and qualitative research to understand individual and structural factors of behavioral health and well-being, with an emphasis on justice issues. 

1. Mark R. Munetz & Patricia A. Griffin, Use of the Sequential Intercept Model as an Approach to Decriminalization of People With Serious Mental Illness, 67 Psychiatric Servs. 544, 547 (2006). Funding Disclosure: The writing of this article was supported by the John D. and Catherine T. MacArthur Foundation’s Pretrial Risk Management Project. The content is solely the responsibility of the authors.
2. Benes Z. Aldana, When It Comes to Civility in Court, It’s Do or Die, 62 The Judges J. 21, 22 (2023).
3. Jessica Traguetto & Tomas de Aquino Guimaraes, Therapeutic Jurisprudence and Restorative Justice in the United States: The Process of Institutionalization and the Roles of Judges, 63 Int’l J. of Offender Therapy and Compar. Criminology 1971, 1981–82 (2019).
4. Substance Abuse and Mental Health Servs. Admin., Samhsa’s Concept of Trauma and Guidance for a Trauma-Informed Approach 4 (2014).
5. Caroline Wolf Harlow, Defense Counsel in Criminal Cases 1 (2000).
6. Sarah E. Duhart Clarke, et al., Indigent Injustice: A Systematic Review and Meta-analysis of People’s Criminal Legal Outcomes, Critical Criminology (July 15, 2024), https://doi.org/10.1007/s10612-024-09768-2.
7. Id.
8. Id.
9. Id.
10. Id.
11. For additional discussion and suggestions for mitigating bias, see Jeffrey Rachlinski, et al., Getting Explicit About Implicit Bias, 104 Judicature 75 (2020–21).
12. Ross MacDonald, et al., The Rikers Island Hot Spotters: Defining the Needs of the Most Frequently Incarcerated, 105 Am. J. of Pub. Health 2262, 2264 (2015).
13. Understanding the Population of People with Frequent Jail Contact, Pol. Rsch. Assocs., https://new.express.adobe.com/webpage/uFaeVQDv4lKpw (last visited Jan. 2, 2025).
14. Id.
15. Id.
16. Id.
17. Id.
18. Id.
19. Understanding the Population of People with Frequent Jail Contact, supra note 13.
20. Criminal Justice Standards Committee, ABA, ABA Standards for Criminal Justice 6, 55 (3d ed. 2007).
21. Lisa Pilnik, A Framework for Pretrial Justice: Essential Elements of an Effective Pretrial System and Agency 48 (2017).
22. Release Standards Committee, Nat’l Ass’n of Pretrial Servs. Agencies, Standards on Pretrial Release: Revised 2020 73 (2020).
23. Samantha A. Zottola, et al., Court date reminders reduce court nonappearance: A meta-analysis, 22 Criminology & Pub. Pol. 97 (2022).
24. Id. at 97.
25. Id. at 108–09.
26. Id. at 113.
27. Samantha A. Zottola, et al., A National Compendium of Court Navigation Programs: Providing support at the nexus of the legal and behavioral health systems (2023).
28. Id. at 8.
29. Id.
30. Id. at 16–17.
31. Id. at 17.
32. See, e.g., id. at 18 (noting that resource navigators in Arizona receive crisis intervention training).
33. Samantha A. Zottola, et al., A Resource Guide on Court Navigator Programs: Providing Connections and Support Across the Legal and Behavioral Health Systems 108 (2024).
34. Zottola, et al., supra note 27, at 4.
35. See, e.g., Zottola, et al., supra note 33, at 41 (“The clinician reduces court actors’ workload by gathering a wide range of behavioral health information.”).
36. Sarah L. Desmarais, et al., Predictive validity of pretrial risk assessments: A systematic review of the literature, 48 Crim. Just. & Behav. 398, 401 (2021) (citation omitted); see also Pamela K. Lattimore, et al., The Prevalence of Local Criminal Justice Practices, 84 Fed. Prob. J. 28, 31 (2020)
    (“. . . 48 percent of jurisdictions reported the use of pretrial risk assessment.”).
37. See Sarah L. Desmarais & Evan M. Lowder, Pretrial Risk Assessment Tools: A primer for judges, prosecutors, and defense attorneys (2019).
38. See Brandon L. Garrett, et al., Nonviolent risk assessment in Virginia sentencing: Report 2: A survey of circuit court judges 17–18 (2018) (“While risk assessments have weight in matters of sexual crimes, [their] accuracy in larceny and drug crimes is debatable.”).
39. See Jennifer E. Copp, et al., Pretrial risk assessment instruments in practice: The role of judicial discretion in pretrial reform, 21 Criminology & Pub. Pol. 329, 347 (2022) (“[C]ourts’ reliance on money may override the intended outcomes of risk-based systems[.]”).
40. Desmarais, et al., supra note 36, at 413–16.
41. Jodi L. Viljoen, et al., Are risk assessment tools more accurate than unstructured judgments in predicting violent, any, and sexual offending? A meta-analysis of direct comparison studies, 42 Behav. Scis. & The L. (forthcoming 2025) (manuscript at 26) (https://onlinelibrary.wiley.com/doi/full/10.1002/bsl.2698).
42. Samantha A. Zottola & Sarah L. Desmarais, Comparing the relationships between money bail, pretrial risk scores, and pretrial outcomes, 46 L. & Hum. Behav. 277, 285 (2022).
43. Spencer G. Lawson, et al., Disparate impact of risk assessment instruments: A systematic review, 49 L. & Hum. Behav. (forthcoming 2025) (manuscript at 7) (https://doi.org/10.1037/lhb0000582).
44. Id.
45. Id.
46. See, e.g., Evan M. Lowder, et al., Effects of pretrial risk assessments on release decisions and misconduct outcomes relative to practice as usual, 73 J. of Crim. Just. 101754 (2021); Keith Cooprider, Pretrial Risk Assessment and Case Classification: A Case Study Control, 73 Fed. Prob. J. 19 (2009); Chloe Anderson, et al., Pretrial Justice Reform Study: Evolution of Pretrial Justice Reforms That Use the Public Safety Assessment (2019).
47. Sarah L. Desmarais, et al., Predictive Validity of Pretrial Risk Assessments: A Systematic Review of the Literature, 48 Crim. Just. & Behav. 398, 406 (2021).
48. Sarah L. Desmarais, et. al., The Empirical Case for Pretrial Risk Assessment Instruments, 49 Crim. Just. & Behav. 807, 811 (2022).
49. Sarah L. Desmarais, Samantha A. Zottola, & John Monahan, Public opinion about judicial roles and considerations: A latent profiles analysis, L. & Hum. Behav. (forthcoming 2025).