Glossary

Introduction

Between January and December 2021, the U.S. DOJ’s Office of Justice Programs, Bureau of Justice Assistance (BJA) provided funding for The Council of State Governments (CSG) Justice Center to conduct an analysis of racial equity in Vermont’s criminal justice system to support the ongoing efforts of the state’s Justice Reinvestment (JR) II Working Group. Earlier data analyses by CSG Justice Center staff, completed in 2019 during the first phase of Vermont’s Justice Reinvestment II Initiative, found preliminary evidence of Black-White racial disparities in the court and Department of Corrections (DOC) systems; however, an in-depth analysis was not possible at that stage due to time constraints and data availability. In light of these initial findings, the JR II Working Group requested further study of racial disparity across the state’s criminal justice system.

The purpose of this new analysis was to investigate patterns of racial disparity over time and provide insight into the drivers behind them, with the goal of providing the JR II Working Group with actionable recommendations to address racial disparities moving forward. Consequently, this effort focused strategically on disparities in sentencing, which can be addressed by policy and practice changes within the criminal justice system.

The CSG Justice Center’s racial equity analysis included the following key components: (a) high-level analysis of sentencing and corrections patterns across racial groups; (b) in-depth analysis of sentencing patterns across racial groups and of factors that contribute to any observed disparities, and (c) an assessment of existing data to identify ways to improve future analysis and monitoring of racial disparity. This project was designed to complement and build upon Vermont’s ongoing work to address racial disparities in the state, particularly those of the Racial Disparities in the Criminal and Juvenile Justice Advisory Panel (RDAP) and efforts to institute a new Office of Racial Justice Statistics.¹²

This report describes results from analysis of criminal justice system data for the six-year period between January 2014 and December 2019 and is organized into the following sections:

• A background section that reviews findings from the CSG Justice Center’s 2019 analysis of racial disparity in Vermont’s courts and corrections systems and provides context from the broader research literature on racial disparity in the criminal justice system in the U.S.
• An analytic approach section, which provides an overview of the data sources and quantitative methods used in this analysis
• A summary of results addressing the extent to which Black-White racial disparities exist in criminal case volume, case processing, and sentencing decisions
• Data-driven policy recommendations that outline a path forward to advance racial equity in the state

Background

Throughout the United States, Black people face incarceration at rates that are grossly disproportionate to their representation in the general population despite progress reducing such racial disparities during the last two decades.¹³ A large body of research documents the role of historical and ongoing forms of systemic racism in creating and perpetuating these racial disparities.¹⁴

Importantly, national research also shows that Black-White criminal justice system disparities are especially pronounced for drug offenses: Black people are three to four times more likely to be arrested and nine times more likely to face state prison incarceration for drug offenses relative to White people, yet these differences are not explained by differences in drug use or sales.¹⁵

Vermont has low average crime and incarceration rates compared to other states;¹⁶ however, despite those achievements, the state’s criminal justice system is rife with the same types of racial disparities that are shown in national research.¹⁷ In fact, for Black-White incarceration disparities, Vermont fares worse than the national average.¹⁸

– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –

The CSG Justice Center’s FY2019 analysis of Vermont’s corrections population included the following key findings:¹⁹

• Black Vermonters were overrepresented in all corrections populations relative to their representation in the Vermont general population; these disparities were most pronounced among sentenced and detained incarcerated populations (see Figure 1).
• Relative to the White incarcerated population, a greater share of Black incarcerated people were convicted of drug offenses; at the same time, a lesser share of the Black incarcerated population was convicted of property offenses.

– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –

Additionally, CSG Justice Center staff’s previous analysis of racial disparities in the FY2019 (July 2018–June 2019) corrections population included the following key findings:

• Without controls for crime type or criminal history, Black people appeared to be incarcerated for misdemeanors and felonies more often than White people.
• Among those incarcerated for felony offenses, average sentence lengths did not vary between Black and White people.

Figure 1. Relative Rate Indices: Vermont DOC Snapshot Population, by Type and Race (FY2019)²¹
In FY2019, Black people were six times more likely to be part of the sentenced incarcerated population relative to White people. Black people were disproportionately represented in all other corrections populations as well.

As part of the CSG Justice Center’s qualitative engagement work during Justice Reinvestment II, Vermont stakeholders reported perceptions that racial disparities in Vermont’s criminal justice system—particularly for drug offenses—were due to crimes committed by people from out of state, specifically individuals who traveled to or through Vermont for the express purpose of trafficking drugs and who often had longer criminal histories that became factors in sentencing decisions.²² The implication or direct reasoning of this thinking was that nothing inherent to Vermont’s actors or systems was responsible for disparate treatment of Black people. As a result, CSG Justice Center staff designed the current study to account for defendants’ state of residence. Importantly, this perception by some stakeholders was not supported by empirical results of this analysis.

Analytic Approach

It is important to note that racial disparity in incarceration populations and rates can be caused by a multitude of factors that originate both within and outside of the criminal justice system²³ (as depicted in Figure 2 below), and research shows that disparities in sentencing are one important contributing factor.²⁴

Figure 2. Sources of Racial Disparity in the Criminal Justice System

This analysis focuses strategically on disparities in court case processing and sentencing, which can be addressed by policy and practice changes within the criminal justice system and are within the scope of the JRI activities. Specifically, this project investigates racial disparities at three key points within the court system: the inflow of criminal cases, the likelihood of conviction in a case, and sentencing decisions related to incarceration.

CSG Justice Center staff used quantitative research methods to conduct an analysis of racial equity and develop data-driven policy recommendations to address observed racial disparities in Vermont.²⁵ Data for this project were obtained from the Vermont Judiciary, the Vermont Crime Information Center (housed in the Department of Public Safety), and the U.S. Census, and datasets were cleaned and analyzed using standard statistical methods, including relative rate index calculations and regression analysis.26 Additionally, CSG Justice Center staff engaged with JR II Working Group and community members to refine and finalize policy recommendations. Through these activities, CSG Justice Center staff aimed to answer the following key analysis questions (see Figure 3):

(1) Are there racial disproportionalities in the volume of felony and misdemeanor cases filed?

(2) Are there racial disparities in conviction across offense types after accounting for key case and defendant characteristics, including Vermont residency and criminal history?

(3) Are there racial disparities in incarceration sentences across offense types after accounting for key case and defendant characteristics, including Vermont residency and criminal history?

Figure 3. Analysis Questions

Quantitative Data Sources and Measures

Vermont Judiciary: The main dataset analyzed for this project was obtained via a data use agreement between the CSG Justice Center and the Vermont Judiciary. Court staff shared data on all criminal felony and misdemeanor court cases disposed between January 1, 2014, and December 31, 2019.27 After excluding cases that were out of the scope of study (e.g., fish and game violations, cases transferred to juvenile court, or cases missing key information), the primary analysis sample included a total of 79,570 cases. A secondary set of analyses were also conducted, using a subset of the main analysis restricted only to people who were Vermont in-state residents at the time a case was filed. This secondary analysis sample consisted of 68,471 cases in total. Findings from the primary analysis are presented in the Results section of this report; however, conclusions are based on results from both the primary and secondary analyses. Details on the development of the analytic sample and secondary analysis results are available in the Technical Appendix.

Vermont Department of Public Safety Crime Information Center: In addition to court data, CSG Justice Center staff obtained in-state criminal history information via a data use agreement with the Vermont Department of Public Safety Crime Information Center (VCIC). VCIC maintains a statewide repository of criminal history data, including information from law enforcement agencies and the court system.

Court cases: CSG Justice Center staff analyzed information at the case level. To identify unique criminal cases, charges that were filed on the same day with the same case number listed were grouped into a single case. Cases were classified by offense level (misdemeanor or felony) as well as offense category. Cases that included at least one felony charge were categorized as a felony case, and cases with exclusively misdemeanor charges were classified as a misdemeanor case. Additionally, each case was classified according to the most severe charge filed in a case (e.g., drug offense, property offense) using a severity index from Vermont’s Crime Research Group, Inc. (CRG). CRG is a nonprofit agency that contracts with the Vermont Department of Public Safety to carry out state Statistical Analysis Center activities,²⁸ such as criminal justice analysis to support policy development.

Key measures and quantitative methods: To understand whether there are racial disproportionalities in cases coming into the court system (Analysis Question 1), relative rate indices were calculated.²⁹ Additionally, regression analysis was employed to examine three main case processing and sentencing outcomes, including the following: whether a defendant was convicted of any charges in a case (Analysis Question 2); among cases with a conviction, whether the sentence included incarceration in state prison, and among cases with an incarceration sentence, the length of incarceration time imposed (Analysis Question 3). Notably, in Vermont, incarceration sentences are served at correctional facilities run by the DOC.³⁰

CSG Justice Center staff collected and coded two types of additional information to isolate potential sources of disparities: case characteristics and defendant characteristics. Case characteristics³¹ included the offense level and category, the total number of charges filed in a case, the year in which the case was disposed, and the county in which the case was processed. Defendant characteristics included race (Black or White), gender, age at the time of case filing, and Vermont or out-of-state residency at the time of case filing. The primary analysis (described above) accounted for each of the case and defendant characteristics just described; the secondary analysis, which was restricted only to defendants who were Vermont residents, additionally accounted for in-state criminal history. Average characteristics of defendants in the primary analysis are summarized in Table 1 below.

Table 1. Characteristics of Defendants in the Primary Analytic Sample³²

Notes: N=79,570

As with any analysis project, there were some factors that could not be accounted for due to limited data availability. For example, the judiciary’s administrative system does not currently track details about decisions made by the prosecuting attorneys or whether the case included a victim’s statement; therefore, such details were not available for this analysis.

Table 2 below summarizes key details of the analytic approach, including information on research questions, outcomes examined, case counts, data sources, and analysis conducted.

Table 2. Summary of Analytic Approach and Results

*Weapons cases were excluded from regression analysis because there were too few to analyze. For details, see Technical Appendix.
**CSG Justice Center staff used in-state criminal history information for Vermont residents only.

Results

Case Flow

To understand the context in which the court system is operating, CSG Justice Center staff examined racial differences in the volume of misdemeanor and felony cases coming to the courts. To analyze racial differences in incoming cases, CSG Justice Center staff used an approach called a relative rate index (RRI).³³

An RRI is useful for comparing the rates of an event or outcome between two groups. Here, when the RRI is above 1, it indicates that Black people are defendants in a misdemeanor case at a higher rate than White people in Vermont. During this six-year period, RRIs for misdemeanor cases are all above 1; this indicates that Black people were disproportionately represented in such cases, relative to White people in the state. For example, in 2019, Black people were 3.5 times more likely than White people to be defendants in a misdemeanor case (See Figure 4). A similar story emerges when RRIs are examined for felony cases (See Figure 5).

Figure 4. Black-White Relative Rate Indices, Misdemeanor Cases (2014–2019)³⁴
In 2019, Black people were 3.5 times more likely than White people to be defendants in a misdemeanor case.

Figure 5. Black-White Relative Rate Indices, Felony Cases (2014–2019)³⁵
In 2019, Black people were 5.9 times more likely than White people to be defendants in a felony case.

As shown in Figure 5, between 2014 and 2019, Black people are disproportionately represented in felony cases relative to White people. However, this racial disparity is more pronounced for felony cases than for misdemeanor cases. For example, in 2019, Black people were 5.9 times as likely as White people to be a defendant in a felony case. Given the level of disparity seen in felony cases, CSG Justice Center staff additionally examined RRIs for felony cases by type of offense (see Figure 6).

Figure 6. Average Annual Black-White Relative Rate Indices, Felony Cases (2014–2019)³⁶
Between 2014 and 2019, Black people were 14.6 times more likely than White people to be defendants in a felony drug case.

As shown in Figure 6, between 2014 and 2019, Black people were disproportionately likely to be defendants in felony cases across all offense categories, including person, property, drug, motor vehicle, and public order crimes. These disparities are especially dramatic for certain types of offenses. During that period, Black people were 14.6 times more likely to be defendants in a felony drug case compared to White people.

In sum, the RRIs for misdemeanor and felony cases indicate that Black people are defendants in criminal cases in Vermont at rates that exceed rates experienced by White people. Importantly, RRI results indicate that there are substantial differences that emerge even before a judge is involved in a case. There are several decision-making points within the criminal justice system that could potentially contribute to these disparities. For example, law enforcement personnel have some discretion in deciding when to make an arrest or issue a citation. Additionally, state’s attorneys, who are the prosecuting attorneys in Vermont, determine whether to move forward with a case and decide which charges to file.

While this project was not designed to examine factors that contribute to these observed disproportionalities, future work in Vermont should investigate the role of community factors, policing, and prosecutorial decision-making, as national research indicates that each may play a role in causing racial disparity in justice system involvement.37 In the sections that follow, results from more rigorous regression analyses are presented to reveal specific drivers of disparity that policymakers can act upon.

Conviction

To examine racial disparities in case processing and sentencing outcomes, CSG Justice Center staff employed a statistical method called regression analysis.38 Regression analysis is a common approach for comparing differences in outcomes between two groups, particularly when there is an interest in making an “apples to apples” comparison between those groups. To examine Black-White differences in the likelihood of being convicted of a crime, CSG Justice Center staff used regression to account for baseline differences between groups (e.g., severity of offense, criminal history, demographic characteristics).

Table 3. Offense Categories Examined in Regression Analysis

Additionally, outcomes were assessed for 10 different combinations of offense types (e.g., person, property) and offense level (i.e., misdemeanor, felony), as shown in Table 3. In the figures below, results are reported only for offense categories for which there was a statistically significant Black-White racial difference. In other words, if an offense category does not appear in a graphic below, it is because there was no statistically significant difference identified.

The first outcome examined was the likelihood of conviction in a case (see Figure 7). In the primary analysis sample for this project, over 99 percent of convictions were made via a plea bargain, and less than 1 percent of convictions were made via a trial. The alternative dispositions here include having a case dismissed or being acquitted. Over 99 percent of the time, a case that does not result in a conviction is dismissed (not acquitted), meaning that a determination was made that the case could not proceed due to limitations such as a lack of sufficient evidence or improperly collected evidence.

Figure 7. Adjusted Probability of Conviction, by Offense Category and Race (2014–2019)³⁹
Results here are mixed: Black people are more likely to face conviction for misdemeanor drug offenses but less likely to face conviction for four other offense categories.

The results from conviction analysis are mixed. After accounting for key case and defendant characteristics, Black people are less likely to be convicted in cases where the top charge is one of the following four offense categories: misdemeanor property, misdemeanor motor vehicle, misdemeanor public order, and felony drug. For example, for misdemeanor property crimes, on average and after accounting for other factors, Black people face conviction about 48 percent of the time, whereas White people face conviction about 59 percent of the time. However, the results reverse when we examine misdemeanor drug cases. Here, Black people face conviction more frequently, about 60 percent of the time, while White people face conviction 48 percent of the time. These results indicate that conviction decisions are not a major driver of racial disparity in Vermont’s criminal justice system and are unlikely to contribute to the disproportionate representation of Black people in Vermont’s prison system.

Incarceration

CSG Justice Center staff additionally assessed the likelihood of receiving a sentence to incarceration among people who had been convicted of a crime (see Figure 8). Specifically, staff examined the likelihood of being sentenced to “straight” incarceration, i.e., prison time without a probation term and without any lenient provisions that allow time to be served in the community, such as a deferred sentence. Alternative sentences included probation, split sentences, deferred or suspended sentences, and pre-approved furlough.

Figure 8. Adjusted Probability of Incarceration, by Offense Category and Race (2014–2019)⁴⁰
Black people are 18 percentage points more likely to be incarcerated for felony drug and property offenses.

As shown in Figure 8, results from regression analysis indicate that Black people are consistently more likely to be sentenced to incarceration relative to their White counterparts even after adjusting for key case and defendant characteristics described earlier on page 9. Specifically, this is true for four offense categories: misdemeanor person, felony property, felony drug, and felony public order. The most dramatic racial disparities are seen for felony property and felony drug offenses, where Black people are 18 percentage points more likely to receive an incarceration sentence relative to comparable White people.

Table 4. Most Common Felony Drug Offenses, by Race (2014–2019)⁴¹
For Cases Resulting in Incarceration

Notes: N=565

To provide more detail on the types of offenses that lead to incarceration, CSG Justice Center staff analyzed the most common specific felony drug offenses that resulted in incarceration sentences during the study period. As shown in Table 4, results indicate that for White people convicted of a felony drug offense and sentenced to incarceration, cases most frequently involve possession or sales of heroin. In contrast, for Black people sentenced to incarceration for a felony drug conviction, cases most often involve cocaine possession or sales. This finding suggests that any policy response to racial disparities in the Vermont criminal justice system will need to account for multiple types of drugs to be effective.

Incarceration—Sentence Length

Finally, CSG Justice Center staff used regression to examine differences in incarceration sentence length imposed by race. After examining differences in minimum incarceration sentence length across the 10 offense categories detailed in Table 3, there were no consistent statistically significant Black-White differences. The results were similar when maximum sentence length was examined as an outcome. Overall, differences in sentence lengths do not appear to be a driver of incarceration disparities in Vermont.

Policy Recommendations

Informed by the results of this analysis, Vermont can enact targeted reforms that aim to reduce racial disparities at sentencing. The following recommendations focus on actionable policy changes that specifically target decision-making points in the court process where racial disparities were found to be most pronounced.

Recommendation 1: Apply a race equity lens to the reclassification of drug possession offenses.

Over the past half century, the targeted disproportionate enforcement of drug policies in Black communities following desegregation and escalation of the War on Drugs has resulted in pronounced racial disparities across criminal justice systems.⁴² Nationally, Black people are more likely to be arrested and incarcerated for drug offenses despite the fact that Black and White people use and sell drugs at similar rates.⁴³

The results of this analysis show that many of these troubling trends are also evident in Vermont. There are significant disparities in how Black people in Vermont are represented and sentenced in felony drug cases compared to White people. Black people are overrepresented in cases coming before the court; they are 14 times more likely to be a defendant in a felony drug case relative to White people. Additionally, Black people convicted of a felony drug offense are 18 percentage points more likely to be sentenced to incarceration than comparable White people.

The Vermont Sentencing Commission is currently considering recommendations for a standardized offense classification system, including for drug offenses. The Sentencing Commission, as well as the legislature, should use the results of this analysis to better understand opportunities within the drug offense classification process for acknowledging and addressing racial disparities. Specifically, the state can use analysis findings to apply a racial equity lens to classification by

• Reclassifying lower- to mid-level felony drug possession offenses to misdemeanors; and
• Reevaluating the threshold of the highest level of possession and sales to better reflect significant amounts of drugs intended for distribution.

As part of the classification process, the legislature directed the Sentencing Commission to examine penalty reductions for the possession of opioids.⁴⁶ Based on the results of this analysis, it is important to note that just focusing on opioid-related possession offenses would potentially increase racial disparities in incarceration in Vermont. This analysis found that while heroin possession is the most common felony drug offense for which White people are sentenced to incarceration, cocaine possession is the most common felony drug offense for which Black people are sentenced to incarceration. To apply a race equity lens to the classification process and avoid perpetuating the historical systemic inequities related to drug policy, Vermont should consider penalty reductions across substances, not just opioids.

Several states have already reduced drug possession from a felony to a misdemeanor; five states have done so regardless of drug quantity up to the third conviction.⁴⁷Unlike Vermont, states that have fully de-felonized drug possession have a mechanism for charging someone with intent to distribute in cases where there are indicators of potential sale. Should the highest level of drug possession in Vermont remain a felony for the purpose of functioning as a de facto intent to distribute mechanism, it is important that the weight threshold be raised to avoid overcriminalization, particularly of people with significant substance use disorders.

Recommendation 2: Establish non-binding sentencing guidance or presumptive probation for certain drug and property offenses.

A primary finding in this analysis is that Black people in Vermont are more likely to face incarceration for drug and property offenses than White people even after controlling for key case variables including in-state criminal history.⁴⁸ While the analysis did not identify consistent racial disparities in likelihood of conviction, this finding does indicate there is a statistically significant difference in the “in/out” incarceration sentencing decision by race for similarly situated defendants.⁴⁹ This means that, once convicted, a Black person is more likely to receive a sentence to incarceration than a White person with similar case characteristics, including criminal history and offense type.

Nationally, state sentencing guidelines or guidance to support the use of judicial discretion has been shown to effectively reduce racial disparities in sentencing outcomes.⁵⁰ Similarly, previous CSG Justice Center analysis found no racial and geographic disparities in maximum misdemeanor and felony probation term lengths for which Vermont has non-binding statutory sentencing guidance.⁵¹

To address racial disparities in incarceration for felony drug and property offenses for similarly situated defendants, Vermont should pursue either non-binding sentencing guidance or presumptive probation for offenses where racial disparities are most pronounced including misdemeanor drug offenses to support the use of discretion in determining whether a person should receive incarceration or community supervision. Guidance or presumptive probation should focus on offenses where racial disparities are most pronounced, as well as where there is an opportunity to support the use of probation rather than incarceration without compromising public safety.

Recommendation 3: Examine racial disparities in diversion and pretrial services.

National research shows that BIPOC are less likely to receive diversion than White people.⁵² However, the data to understand if these same disparities exist in Vermont are not readily available. The legislature should require the collection and public reporting of race and ethnicity data in the Attorney General’s Pretrial Services and Court Diversion Report, which provides the legislature with an annual review of pretrial and diversion service usage and outcomes. The report should include the race and ethnicity of individuals who are eligible for, receiving, and declining services.

In 2020, the Vermont Attorney General’s office began collecting race and ethnicity information for pretrial and diversion programming and is in the process of considering next steps related to this recommendation.⁵³ Collecting and publicly reporting this information would help Vermont determine whether racial disparities in program access exist and if those disparities are pronounced in specific counties.

Recommendation 4: Develop internal guidance to support consistency in charging and plea-bargaining decisions within state’s attorneys’ offices.

State’s attorneys play a critical, if sometimes less visible, role in sentencing outcomes. Prosecutorial data is rarely available to fully understand the impact of charging and plea-bargaining decisions on sentencing outcomes, including racial disparities.⁵⁴ However, national studies have found that Black people are less likely to receive charges or offers for non-custodial sentences during the plea-bargaining process than White people.⁵⁵ Given that 99 percent of cases in Vermont are resolved by plea bargains, the role of state’s attorneys’ offices in the charging and plea-bargaining process is an important area of focus for understanding and addressing racial disparities in sentencing outcomes.⁵⁶

Prosecutors in other jurisdictions have proactively sought to reduce racial disparities by adopting internal guidance to provide structure for decision-making during the charging and plea-bargaining process; Vermont’s state’s attorneys should consider adopting this type of approach. Like non-binding sentencing guidance for judges, this type of guidance for state’s attorneys maintains the use of discretion and can be limited to specific offenses where racial disparities are most pronounced, to have the most targeted impact.

To monitor implementation of internal guidance, Vermont state’s attorneys’ offices should build on existing efforts to standardize collection of prosecutorial data, aim to regularly collect and examine charging and plea-bargaining data, and consider establishing a process for internal charge review prior to filing. Several jurisdictions in other states have successfully adopted internal guidance to guide prosecutorial decision-making for the purpose of achieving policy goals, including reducing racial disparities. For example, the Milwaukee District Attorney’s Office eliminated disparities in drug paraphernalia charges by establishing guidance that prioritized diversion or dismissal.⁵⁷ They coupled this guidance with a limited internal charge review process as well as regular data collection and monitoring to benchmark progress. The Philadelphia District Attorney’s Office implemented sentencing policies governing charging and plea-bargaining decisions that resulted in decreased disparities in community supervision sentencing.⁵⁸

Recommendation 5: Improve the collection, analysis, and availability of race and ethnicity data to inform ongoing training and decision-making.

Vermont’s Justice Reinvestment II legislation, Act 148 (2020), tasked the Racial Disparities in the Adult and Juvenile Justice System Advisory Panel (RDAP) and other stakeholders with studying and making recommendations regarding gaps and challenges in race and ethnicity data collection. RDAP has since delivered two reports to the legislature recommending the creation of an Office of Racial Justice Statistics (Office) to manage the collection and analysis of criminal justice-related race and ethnicity data.⁵⁹

Several of the following recommendations support key components of the RDAP proposal for the Office as described in its November 2021 report to the Vermont legislature. If sufficiently resourced, the Office would be a comprehensive, first-of-its-kind entity that will provide Vermont with a more complete understanding of how disparities compound as a person moves through the system, as well as the information to develop targeted, data-driven policy reforms.

Expand availability of Hispanic ethnicity data to law enforcement and the courts. The data that the Vermont Judiciary receive from law enforcement do not currently include Hispanic ethnicity. When ethnicity data are not collected in addition to race it can result in significant undercounting of the Hispanic population.60 For example, in Vermont, it was only after the DOC began collecting ethnicity data separate from race in August 2020 that it became apparent that Hispanic people were overrepresented in certain corrections populations.⁶¹ Specifically, in January 2022, the proportion of the DOC corrections population incarcerated or supervised in the community who identified as Hispanic was 10.4 percent and 7 percent, respectively. The percentage of Hispanic people represented in each of these corrections populations is notably higher than Vermont’s general Hispanic population, which is approximately 2 percent.⁶² It is possible that once the Vermont Judiciary begins to collect Hispanic ethnicity data, additional disparities will come to light. To better understand and address disparities at sentencing for people who identify as Hispanic, Vermont must ensure that accurate ethnicity data are available in the court’s data system.

Invest in staffing and system improvements necessary to increase future data collection and analysis capacity. Whether through the Office of Racial Justice Statistics or a similar mechanism, addressing gaps in racial and ethnic disparity data will require targeted, long-term investments in the people and infrastructure necessary to collect, share, and analyze quality information. Additionally, individual agencies and organizations may also require funding for system upgrades to improve initial data collection and make information sharing possible.

Collect and analyze sentencing data statewide and by judicial district. Vermont should collect race and ethnicity data for each key decision-making point in the court process, including case inflow, conviction, incarceration, and sentence length. Comprehensive data collection should also include information on charging and plea-bargaining decisions.

Identify opportunities to publish racial disparity data, including an annual report to benchmark and monitor progress. Racial disparity data should regularly be made available to the public to promote transparency and accountability. Data should be in an accessible format that includes critical context for the public to understand and engage with the information.

Engage impacted communities in collecting quantitative and qualitative data as well as in developing and implementing racial disparity-related policy changes. RDAP’s November 2021 report emphasized the importance of building relationships with impacted communities and ensuring they are meaningful partners at each stage of the data collection and analysis process to avoid perpetuating practices that reflect existing systemic racism.⁶³ To provide critical context for quantitative data, Vermont should also consider collecting qualitative data to better understand the lived experiences of BIPOC impacted by the criminal justice system.

Use data and community engagement to inform judicial training to support consistent decision-making. Even when cases are resolved by plea bargain, judges still have significant authority during the sentencing process, including the ability to question or refuse a plea. Quantitative and qualitative data can help the Vermont Judiciary identify training opportunities for judges to increase their understanding of how to identify and address racial disparities from the bench, as well as generally support consistency in decision-making across the state.

Conclusion

Over the past two decades, Vermont has worked to actively understand and address complex and persistent challenges in its criminal justice system, including undertaking two separate JRI processes. Produced at the request of the JR II Working Group, this report demonstrates Vermont’s ongoing commitment to using a data-driven approach to achieve a safer, more equitable, and more just system statewide. It summarizes important findings that highlight the extent to which disparities exist at key decision-making points in the court system, with an eye toward identifying potential drivers of racial disparities in the state’s incarcerated population.

Results indicate that Black people are more likely than White people to be defendants in both misdemeanor and felony cases, without adjusting for other factors. This suggests that prior to cases coming to the court, there are differences in community factors and/or criminal justice system decision-making that contribute to disparities in the volume of cases coming into the court system. Additionally, after accounting for key case and defendant characteristics, evidence did not suggest that Black people are more likely to be convicted of most offenses, or that incarceration sentence length differs between Black and White people. However, when similarly situated Black and White defendants were compared, there was consistent evidence of racial disparities that disadvantage Black people in the decision to incarcerate; this was true even when the analysis was restricted to defendants from Vermont.

The disparities identified in this analysis underscore the need for actionable policy change. In this vein, the five recommendations in this report offer data-driven strategies Vermont can employ to create a more equitable justice system, and in November 2021, JR II Working Group members voted to move forward with all of them. However, it is important to note that sentencing is only one driver of the racial disparities in incarceration identified in Vermont and there is more work to be done. Initiatives such as RDAP’s Office of Racial Justice Statistics can provide Vermont with the data analysis capacity to examine other potential drivers at key decision-making points in the system, including arrest, pretrial, charging, and plea bargaining. This information will be critical as Vermont continues to understand and address the compounding impacts of racial disparity within its criminal justice system.

Technical Appendix

Quantitative Data Sources

CSG Justice Center staff obtained data from two main sources for this project: the Vermont Judiciary, which provided court records for criminal cases, and the Vermont Crime Information Center, which houses the state’s criminal history repository. Additionally, Census data were used in RRI calculations.

Vermont Judiciary: The main courts dataset analyzed for this project was obtained via a data use agreement between the CSG Justice Center and the Vermont Judiciary. Court staff shared data on 160,079 criminal felony and misdemeanor court charges disposed between January 1, 2014, and December 31, 2019. By grouping together charges filed on the same day with the same case number, CSG Justice Center staff identified a total of 90,142 cases that were used to construct an analytic sample (see section below on Primary and Secondary Analytic Samples for details).

Vermont Department of Public Safety Crime Information Center (VCIC): In addition to court data, CSG Justice Center staff obtained in-state criminal history information via a data use agreement with the VCIC. CSG Justice Center staff securely submitted information on defendants in the racial equity analysis to VCIC and obtained a matching in-state criminal record for 93.4 percent of cases involving a defendant who resided in Vermont at the time of case filing (some cases with matching criminal history data were not ultimately used in the analysis due to other exclusion criteria detailed in tables A1–2 below). Although matching criminal record information could include arrests, criminal cases, and convictions, only conviction information was used to develop a criminal history score. Criminal history data were used in the secondary analysis only, as they are relevant only for in-state residents and were unlikely to be complete for out-of-state residents.

U.S. Census Data: CSG Justice Center staff also obtained publicly available Census data.⁶⁴ Specifically, information on the number of Black and White Vermont residents, ages 17–85 years, was obtained for relative rate index calculations. This information was not, however, needed for regression analyses.

Primary and Secondary Analytic Samples

Primary Analytic Sample: Table A1 below provides details on how the primary analytic sample was constructed, starting from the initial 90,142 cases that were identified in the data shared by the Vermont Judiciary. The final primary analytic sample includes 79,514 cases.

Table A1. Development of Primary Analytic Sample: Vermont and Out-of-State Residents

Subsamples: The final primary analytic sample used to analyze the first outcome, conviction, included 79,514 cases. As summarized in Table A2, this group was then narrowed to the subset of 49,594 cases where there was a conviction to analyze the second outcome, incarceration. A final subsample was developed, consisting of the 10,754 cases with a sentence to incarceration and no missing information for minimum sentence length, to analyze the third regression outcome, incarceration sentence length.

Table A2. Primary Analytic Sample and Subsamples

Secondary Analytic Sample: Table A3 provides details on how the secondary analytic sample was constructed, starting from the 79,514 cases used in the primary analyses.

Table A3. Development of Secondary Analytic Sample: Vermont Residents Only

Subsamples: The final secondary analytic sample used to analyze the first outcome, conviction, included 68,471 cases. As indicated in Table A4, this group was then narrowed to the subset of 43,601 cases where there was a conviction to analyze the second outcome, incarceration. Lastly, a final subsample was developed, consisting of the 10,062 cases with a sentence to incarceration and no missing information for minimum sentence length, to analyze the third regression outcome, incarceration sentence length.

Table A4. Secondary Analytic Sample and Subsamples

Relative Rate Indices

Starting with the primary analytic sample consisting of 79,514 cases disposed between 2014 and 2019, Black-White relative rate indices (RRIs) were calculated for misdemeanor and felony cases, respectively, per year. Figure A1 below presents an example of an RRI calculation for misdemeanor cases that were disposed in Vermont in 2019.

Figure A1. Black-White Relative Rate Index Calculation for Misdemeanor Cases (2019)⁶⁵

Calculating Relative Rate Indices

As shown in Figure A1, to calculate an RRI for 2019 misdemeanor cases, the number of misdemeanor cases in which a Black person was a defendant is divided by the number of Black people, ages 17–85 years, who were living in Vermont that year; this provides a misdemeanor case rate of .0689. Stated differently, about 68.9 per 1,000 Black people were defendants in a misdemeanor case in 2019.

When the same calculation is performed for White people, the misdemeanor case rate is .0197, meaning that 19.7 per 1,000 White people were defendants in a misdemeanor case in 2019. Finally, the Black rate is divided by the White rate to obtain an RRI of 3.5. An RRI over one indicates that Black people are disproportionately represented in criminal cases relative to White people in Vermont. In 2019, Black people were 3.5 times more likely than White people to be defendants in a misdemeanor case.

Regression Analysis: Measures

To complete regression analyses, a series of relevant dependent (outcome) variables, as well as independent and control variables, were constructed using Vermont Judiciary case disposition data.

Dependent Variables

The first dependent variable is conviction, a binary variable indicating if there was a conviction—via plea agreement or jury trial—for any charge filed in the case. The alternative values here included case dismissal or acquittal. The second dependent variable is incarceration, a binary variable indicating whether a sentence to “straight” incarceration was imposed in the case. Alternatives to straight incarceration include probation, a split sentence (which is very uncommon in Vermont), a deferred sentence, suspended sentence, or a sentence to pre-approved furlough, which may include a work or treatment program—rather than imprisonment—administered by the
Department of Corrections.

The third and final dependent variable is incarceration sentence length, specifically, the minimum number of prison days imposed. To construct incarceration sentence length, sentencing information was summarized across the case. If there was only one conviction that included a prison sentence, then the incarceration sentence length outcome corresponds to that particular sentence only.

However, in many cases, there are several charges for which the defendant is convicted, and each conviction is associated with a prison term, to be served either concurrently or consecutively. In cases where the prison terms are to be served concurrently, the incarceration sentence length outcome corresponds to the longest sentence in the case (e.g., if someone is sentenced to serve 100 days and 200 days concurrently, the effective sentence length is 200 days). In cases where the prison terms are to be served consecutively, then the incarceration sentence length outcome was constructed by adding all terms together (e.g., if a person is sentenced to serve 100 days and 200 days consecutively, the effective sentence length is 300 days). Among the 12,576 cases in the data provided by the judiciary that had information for minimum sentence length available, 1,447 cases (about 11.5 percent) had more than one prison term listed (across multiple convictions), but notation about whether those terms were to be served concurrently versus consecutively was not recorded in a consistent manner. As a result, these cases were excluded from the incarceration sentence length analysis because a reliable estimate of minimum sentence length imposed could not be calculated. Decisions to construct the sentence length variable in the manner just described were based on conversations with Vermont Judiciary court administrative staff, who provided information on how sentencing information typically works in practice.⁶⁶

Independent and Control Variables

The main independent variable is defendant race, a binary variable indicating whether the defendant identified as Black or White. The Vermont Judiciary obtains race information from law enforcement; law enforcement practice is to ask individuals to report their racial identity.⁶⁷ In addition, two sets of control variables were constructed to account both for key case and defendant characteristics, respectively.

In terms of case characteristics, the first control variable is the offense level and category. Offense level (misdemeanor or felony) is recorded for every charge in a case; a case is considered a felony case if one or more felony charges are filed in the case. Additionally, a severity index specific to Vermont⁶⁸ was used to rank offense categories from most to least severe. Using both offense level and category, each case was classified into one of 10 possible categories to reflect the most severe charge filed in the case (e.g., misdemeanor-public order, felony-person; for a full list, see Table 1 on p. 8). A category for misdemeanor weapons and felony weapons cases was initially constructed but later dropped from the analyses because there were too few cases to reliably compare Black and White defendants. Additionally, a count variable for the total charges filed in a case was constructed; this variable was logged to account for a skewed distribution, a standard statistical practice. Finally, controls for the year in which the case was disposed and an indicator for the county in which the case was processed were included.

Regarding defendant characteristics, in addition to race, demographic information included a binary variable for gender (female or male) and a count variable for age at the time of case filing. A variable indicating whether the defendant was a Vermont or out-of-state resident was constructed using court records on the defendant’s address at the time of case filing.

The primary analysis (detailed above) accounted for each of the case and defendant characteristics just described; the secondary analysis, which was restricted only to defendants who were Vermont residents, additionally accounted for in-state criminal history. To construct a criminal history score, conviction records from the Vermont Crime Information Center were used. Prior felony convictions were counted as a “1,” and prior misdemeanor convictions were counted as “.5”; the two added together yielded a total prior criminal history score.⁶⁹ Because this criminal history variable included zero values and had a skewed distribution, it was adjusted using a log(x+1) transformation. Means and proportions for the dependent variables are shown in Table A5; means and proportions for the independent and controls variables are summarized in Table A6.

Summary Statistics: Primary and Secondary Analysis Groups

Table A5. Outcome Means/Proportions for Primary and Secondary Analytic Samples, by Race (2014–2019)

Table A6. Summary Statistics for Primary and Secondary Analytic Samples, 2014–2019

Notes: Proportions may not add up to 1 due to rounding.

Regression Analysis Results

Regression analysis was used to examine the extent to which there are racial disparities at key decision-making points in Vermont’s court system after accounting for case and defendant characteristics. For each outcome—including (1) likelihood of conviction, (2) likelihood of a sentence to straight incarceration, and (3) incarceration sentence length—regressions were run first on the primary analysis sample and then on the secondary analysis sample; results from both sets of regression were compared to develop a conclusion. For binary outcomes—likelihood of conviction and likelihood of a sentence to straight incarceration—logistic regression was employed. For incarceration sentence length, a count variable, negative binomial regression was used because sentence length is an over-dispersed count variable, meaning that there is a high level of variation, or spread, in the distribution of this variable.

Table A7 presents a summary of results from all regression analyses. Table A7 explains that analysis results from both the primary and secondary analyses were used in combination to determine whether there was consistent evidence of a racial disparity for each decision-making point. Tables A8–10 present results from regression models for each outcome across the primary and secondary analytic samples. Recall that the size of the analytic sample varies across primary and secondary analyses, and between outcomes, due to data availability (as summarized in Tables A1–4 in the appendix).

To obtain the results that are depicted in Figures 7–9 (in the main body of the report), fully adjusted regression models were used to obtain predicted outcomes for each offense category (e.g., M-Person, F-Public Order), but results were only reported in graphical form when the Black-White difference in the predicted outcomes was statistically significant.

Table A7. Summary of Results from Primary and Secondary Analyses

Table A8. Odds Ratios from Logistic Regression Predicting Conviction, 2014–2019.

Notes: The omitted category for race is “White”; the omitted category for offense category is “M-Person”; the omitted category for gender is “female”; the omitted category for residence at case filing date is “out-of-state”; the omitted category for disposition year is “2014”; and the omitted category for county is “Addison.”
*** p<0.001, ** p<0.01, * p<0.05

Table A9. Odds Ratios from Logistic Regression Predicting Incarceration, 2014–2019

Notes: The omitted category for race is “White”; the omitted category for offense category is “M-Person”; the omitted category for gender is “female”; the omitted category for residence at case filing date is “out-of-state”; the omitted category for disposition year is “2014”; and the omitted category for county is “Addison.”
*** p<0.001, ** p<0.01, * p<0.05

Table A10. Incident Rate Ratios (IRRs) from Negative Binomial Regression Predicting Minimum Sentenced Incarceration Length, 2014–2019

Notes: The omitted category for race is “White”; the omitted category for offense category is “M-Person”; the omitted category for gender is “female”; the omitted category for residence at case filing date is “out-of-state”; the omitted category for disposition year is “2014”; and the omitted category for county is “Addison.”
*** p<0.001, ** p<0.01, * p<0.05

Endnotes

1. Rowena Fong, Ruth McRoy, and Alan Dettlaff, “Disproportionality and Disparities,” Encyclopedia of Social Work, June 2, 2014; accessed June 15, 2020, https://oxfordre.com/socialwork/view/10.1093/acrefore/9780199975839.001.0001/acrefore-9780199975839-e-899.

2. Matthew Clair and Jeffrey S. Denis, “Sociology of Racism” in The International Encyclopedia of the Social and Behavioral Sciences,
2nd edition, Volume 19, ed. James D. Wright (Elsevier Ltd., 2015), 857–863.

3. “Vermont Statutes Online: 13 VSA § 1 Felonies and misdemeanors defined,” Vermont General Assembly, 2021, accessed June 14, 2021, https://legislature.vermont.gov/statutes/section/13/001/00001.

4. “FindLaw Legal Dictionary: Disposition,” FindLaw by Thomas Reuters, 2021, accessed August 10, 2021, https://dictionary.findlaw.com/definition/disposition.html.

5. “FindLaw Legal Dictionary: Conviction,” FindLaw by Thomas Reuters, 2021, accessed August 10, 2021, https://dictionary.findlaw.com/definition/disposition.html; Lindsey Devers, Plea and Charge Bargaining: Research Summary (Washington, DC: Bureau of Justice Assistance, 2011).

6. “Vermont Statutes Online: 13 VSA § 7041 Deferred sentence.”

7. “FindLaw Legal Dictionary: What is a Suspended Sentence?” FindLaw by Thomas Reuters, 2021, accessed August 10, 2021, https://www.findlaw.com/criminal/criminal-procedure/suspended-sentences.html; meeting between The Council of State Governments Justice Center and the Vermont Judiciary, July 15, 2021.

8. Jacqueline Salvi et al., “Vermont Justice Reinvestment II Working Group Meeting” (PowerPoint presentation, Vermont Justice Reinvestment Initiative II, December 16, 2019).

9. Development Services Group, Inc., Disproportionate Minority Contact (Washington, DC: Office of Juvenile Justice and Delinquency Prevention, 2014), https://www.ojjdp.gov/mpg/litreviews/Disproportionate_Minority_Contact.pdf.

10. Roxy Peck, Chris Olsen,and Jay L.Devore, Introduction to Statistics and Data Analysis, 5th Edition (Boston: Cengage Learning, 2016).

11. William J. Sabol, Thaddeus L. Johnson, and Alexander Caccavale, Trends in Correctional Control by Race and Sex (Washington, DC: Council on Criminal Justice, 2019).

12. Racial Disparities in the Criminal and Juvenile Justice Advisory Panel, Report of the Racial Disparities in the Criminal and Juvenile Justice Advisory Panel Systems Advisory Panel concerning Act 65 – the creation of a data entity to aggregate and analyze data related to systemic racial bias and disparities within the criminal and juvenile justice systems (Vermont: Racial Disparities in the Criminal and Juvenile Justice Advisory Panel, 2021), https://legislature.vermont.gov/assets/Legislative-Reports/ACT-65-RDAP-REPORT-FINAL.pdf.

13. E. Ann Carson, Prisoners in 2019 (Washington, DC: Bureau of Justice Statistics, 2020); Ashley Nellis, The Color of Justice: Racial and Ethnic Disparity in State Prisons (The Sentencing Project, 2021); Sabol, Johnson, and Caccavale, Trends in Correctional Control by Race and Sex. This discussion focuses on Black-White disparities, which are the most relevant to the analysis presented in this report. However, as noted in this report, other BIPOC groups also face disparities in the criminal justice system that are equally important and urgent to address.

14. Fong, McRoy, and Dettlaff, “Disproportionality and Disparities”; Travis, Western, and Redburn, “The Growth of Incarceration in the United States: Exploring Causes and Consequences.” For an overview of this topic, see Susan Nembhard and Lily Robin, Racial and Ethnic Disparities throughout the Criminal Legal System: A Result of Racist Policies and Discretionary Practices (Washington, DC: Urban Institute, 2021). For a discussion of the recent literature on this topic, see “There’s Overwhelming Evidence that the Criminal Justice System is Racist,” editorial, Washington Post, June 10, 2020,
https://www.washingtonpost.com/graphics/2020/opinions/systemic-racism-police-evidence-criminal-justice-system/.

15. The Council of State Governments Justice Center, September 22, 2020, Vermont Justice Reinvestment Working Group Racial Disparity Analysis, Presentation to the Racial Disparities in the Criminal and Juvenile Justice System Advisory Panel Subcommittee; Mitchell and Caudy, “Examining Racial Disparities in Drug Arrests”; Rothwell, “Drug Offenders in American Prisons: The Critical Difference Between Stock and Flow”; Travis, Western, and Redburn, “The Growth of Incarceration in the United States: Exploring Causes and Consequences.”

16. Cassondra Warney, Ed Weckerly, and Ellen Whelan-Wuest, “Vermont Justice Reinvestment II Working Group Meeting” (PowerPoint presentation, Vermont Justice Reinvestment Initiative II, August 26, 2019); Carson,  Prisoners in 2019.

17. The Council of State Governments Justice Center, “Vermont Justice Reinvestment II Working Group Racial Disparity Analysis,” (PowerPoint presentation, Racial Disparities in the Criminal and Juvenile Justice System Advisory Panel Subcommittee, September 22, 2020); Nellis, The Color of Justice: Racial and Ethnic Disparity in State Prisons.

18. Ibid.

19. The Council of State Governments Justice Center, September 22, 2020, Vermont Justice Reinvestment Working Group Racial Disparity Analysis, Presentation to the Racial Disparities in the Criminal and Juvenile Justice System Advisory Panel Subcommittee.

20. Meetings between The Council of State Governments Justice Center and Vermont Justice Reinvestment Working Group members and stakeholders, August 2019–January 2020.

21. Analysis of FY2019 Vermont Department of Corrections data and 2010–2019 U.S. Census Bureau State Population by Characteristics data conducted by The Council of State Governments (CSG) Justice Center, November 2021.

22. Development Services Group, Inc., Disproportionate Minority Contact (Washington, DC: Office of Juvenile Justice and Delinquency Prevention, 2014).

23. National Academies of Sciences, Engineering, and Medicine, Addressing the Drivers of Criminal Justice Involvement to Advance Racial Equity: Proceedings of a Workshop in Brief (Washington, DC: The National Academies Press, 2021); Stan Orchowsky, “Administering ‘Race Neutral’ Justice: What Works?” Presentation to the National Criminal Justice Association, Cambiare Consulting, Sep. 17, 2019, https://www.cambiareconsulting.com/resources; Travis, Western, and Redburn, “The Growth of Incarceration in the United States: Exploring Causes and Consequences.”

24. Travis W. Franklin, “The State of Race and Punishment in America: Is Justice Really Blind?” Journal of Criminal Justice 59 (2018): 18–28; Ryan D. King and Michael T. Light, “Have Racial and Ethnic Disparities in Sentencing Declined?” Crime and Justice ⁴⁸, no. 1 (2019): 365–437; Michael T. Light, “The Declining Significance of Race in Criminal Sentencing: Evidence from US Federal Courts” Social Forces (2021), https://doi.org/10.1093/sf/soab018; Ojmarrh Mitchell, “A Meta-Analysis of Race and Sentencing Research: Explaining the Inconsistencies,” Journal of Quantitative Criminology 21, no. 4 (2005): 439–466.

25. During the project’s planning phase, CSG Justice Center staff submitted a research plan and obtained Institutional Review Board approval to conduct this work from CSG Justice Center’s partner, the University of Southern Maine National Review Board.

26. These methods are described in brief on pp. 7–8. Additionally, a detailed description of methods used is available in the Technical Appendix.

27. “Guide for New Legislators,” Vermont Judiciary, 2019, accessed June 14, 2021, https://www.vermontjudiciary.org/sites/default/files/documents/Guide%20for%20New%20Legislators%20011619.pdf. Vermont has a Unified Court System, and all criminal cases are filed in the criminal division of the Superior Court, which has one unit located in each of the state’s 14 counties.

28. “Vermont’s Statistical Analysis Center and Crime Research Group, Frequently Asked Questions,” Crime Research Group, accessed November 15, 2021, https://www.crgvt.org/uploads/5/2/2/2/52222091/sac_faqs.pdf; “Statistical Analysis Centers,” Justice Research and Statistics Association, accessed November 15, 2021, https://www.jrsa.org/sac/index.html.

29. See Technical Appendix for details.

30. “Department of Corrections,” Vermont Agency of Human Services Department of Corrections, accessed June 14, 2021, https://doc.vermont.gov/.

31. For detailed summary statistics for the analytic sample, see Technical Appendix.

32. Analysis of 2014–2019 Vermont Judiciary data conducted by The Council of State Governments (CSG) Justice Center, November 2021.

33. For details on how RRIs were calculated, see Technical Appendix.

34. Analysis of 2014-2019 Vermont Judiciary disposition data and 2010–2019 U.S. Census Bureau State Population by Characteristics data conducted by The Council of State Governments (CSG) Justice Center, November 2021.

35. Ibid.

36. Analysis of 2014–2019 Vermont Judiciary data conducted by The Council of State Governments (CSG) Justice Center, November 2021.

37. National Academies of Sciences, Engineering, and Medicine, Addressing the Drivers of Criminal Justice Involvement to Advance Racial Equity: Proceedings of a Workshop—in Brief (Washington, DC: The National Academies Press, 2021), https://doi.org/10.17226/26151; National Academies of Sciences, Engineering, and Medicine, Proactive Policing: Effects on Crime and Communities (Washington, DC: The National Academies Press, 2018), https://doi.org/10.17226/24928; Christi Metcalfe and Ted Chiricos, “Race, plea, and charge reduction: An assessment of racial disparities in the plea process,” Justice Quarterly 35, no. 2 (2018): 223–253.

38. Roxy Peck, Tom Short, and Chris Olsen, Introduction to Statistics and Data Analysis, 6th edition (Cengage Learning, 2019).

39. Analysis of 2014–2019 Vermont Judiciary data conducted by The Council of State Governments (CSG) Justice Center, November 2021.

40. Ibid.

41. Ibid.

42. Susan Nembhard and Lily Robin, Racial and Ethnic Disparities throughout the Criminal Legal System, A Result of Racist Policies and Discretionary Practices (Washington, DC: Urban Institute, 2021).

43. Mitchell and Caudy, “Examining Racial Disparities in Drug Arrests”; Rothwell, “Drug Offenders in American Prisons: The Critical Difference Between Stock and Flow”; Travis, Western, and Redburn, “The Growth of Incarceration in the United States: Exploring Causes and Consequences.”

44. Magnus Lofstrom, Brandon Martin, and Steven Raphael, “Effect of Sentencing Reform on Racial and Ethnic Disparities in Involvement with the Criminal Justice System: The Case of California’s Proposition 47,” Criminology and Public Policy 19 (2020): 1165–1208.

45. Oregon Criminal Justice Commission, “Update to Possession of Controlled Substances Report” 2019, https://www.oregon.gov/cjc/CJC%20Document%20Library/2019PCSReport.pdf.

46. Act 142 General Assembly of the State of Vermont, Regular Session (2018) https://legislature.vermont.gov/Documents/2018/Docs/ACTS/ACT142/ACT142%20As%20Enacted.pdf.

47. Brian Elderbroom and Julia Durnan, Reclassified, State Drug Law Reforms to Reduce Felony Convictions and Increase Second Chances (Washington, DC: Urban Institute, 2018). Alaska later repealed reclassification of drug possession offenses.

48. Analysis of 2014–2019 Vermont Judiciary data conducted by The Council of State Governments (CSG) Justice Center, November 2021.

49. Ibid.

50. Kelly Lyn Mitchell, “State Sentencing Guidelines: A Garden Full of Variety,” Federal Probation 81 (2017); Dale Parent, et al., “The Impact of Sentencing Guidelines” (Washington, DC: National Institute of Justice, 1996); Michael Tonry, “Fifty Years of American Sentencing Reform: Nine Lessons,” Crime and Justice 48 (2019).

51. Analysis of FY2015-FY2019 Vermont Judiciary disposition data by The Council of State Governments Justice Center, September 2020.

52. Mauer and Ghandnoosh, Incorporating Racial Equity in Criminal Justice Reform.

53. Charity Clark (Chief of Staff, Vermont Attorney General’s Office), in discussion with the authors, November 2021.

54. John F. Pfaff, “Prosecutorial Guidelines” in Reforming Criminal Justice, ed. Erik Luna (Phoenix: The Academy for Justice, 2017), 101–120.

55. Ram Subramanian et al., In the Shadows: A Review of the Research on Plea Bargaining (New York: The Vera Institute of Justice, 2020).

56. Analysis of FY2015-FY2019 Vermont Judiciary data by The Council of State Governments Justice Center, November 2019.

57. Mauer and Nazgol, Incorporating Racial Equity into Criminal Justice Reform.

58. The Philadelphia District Attorney’s Office, Ending Mass Supervision: Evaluating Reforms (Philadelphia: The Philadelphia District Attorney’s Office, 2021), https://www.docdroid.net/rWxCb15/dao-supervision-report-pdf.

59. Racial Disparities in the Criminal and Juvenile Justice Systems Advisory Panel, Report of the Racial Disparities in the Criminal and Juvenile Justice Systems Advisory Panel concerning Act 65.

60. Nellis, The Color of Justice: Racial and Ethnic Disparity in State Prisons.

61. Analysis of Vermont Department of Corrections October 2021 population snapshot data by The Council of State Governments Justice Center, December 2021.

62. Analysis of Vermont Department of Corrections data January 2022 and 2021 U.S. Census Bureau State Population by Characteristics data conducted by The Council of State Governments (CSG) Justice Center, February 2022.

63. Racial Disparities in the Criminal and Juvenile Justice Systems Advisory Panel, Report of the Racial Disparities in the Criminal and Juvenile Justice Systems Advisory Panel concerning Act 65.

64. “State Population by Characteristics, 2010–2019,” U.S. Census Bureau.

65. Analysis of 2014–2019 Vermont Judiciary disposition data and 2010–2019 U.S. Census Bureau State Population by Characteristics data conducted by The Council of State Governments (CSG) Justice Center, November 2021.

66. Meeting between The Council of State Governments Justice Center and the Vermont Judiciary, July 15, 2021.

67. Email between The Council of State Governments Justice Center and the Vermont Crime Research Group, June 15, 2021.

68. Email between The Council of State Governments Justice Center and the Vermont Crime Research Group, May 26, 2021. Severity index provided by Vermont Crime Research Group.

69. See Robin Weber, Race and Sentencing in Vermont (Montpelier, VT: Crime Research Group, 2015). This method for constructing a total prior criminal history score was developed by the Vermont Crime Research Group.

Project Credits

Writing and Research: Sara Bastomski, PhD; Madeleine Dardeau, MA; and Lorretta Sackey, MA, CSG Justice Center

Advising: David D’Amora, MS, LPC, CFC; Sara Friedman, MPA; Angela Gunter, MPA; Elizabeth K. Lyon, BA; Jessica Saunders, PhD; and Ellen Whelan-Wuest, MA, CSG Justice Center

Editing: Leslie Griffin, CSG Justice Center

Public Affairs: Brenna Callahan, CSG Justice Center

Design: Michael Bierman

Web Development: Eleventy Group