Disparities at Points of Prosecutorial Discretion

A collaboration of the Colorado Evaluation and Action Lab, the Sixth Judicial District Attorney's Office, and the Loyola Chicago Center for Criminal Justice — a project of the Prosecutorial Performance Indicators.

The fair and just treatment of all communities at each stage of the criminal justice process is of significant importance. Central to this discourse is a recognition of the discretionary power that prosecutors wield in shaping the outcomes of criminal cases. This includes, among other things, the decision to prosecute or decline to file charges, adjust the severity of charges, dispose of cases through dismissal, deferral, or plea negotiations, and make sentencing recommendations. It is valuable to examine such points of discretion to understand whether there are differences in outcomes across individuals of different races/ethnicities.

Colorado's 6th District Attorney's (DA) Office data dashboard, released in September 2022, provides a diagnostic tool to identify what is happening. However, if we see differences, for example, between White and Native American individuals, what does that tell us?

Digging Deeper. What do Differences Mean?

To support actionability, it is important to dig deeper, distinguishing between two important concepts — disproportionality and disparity — and considering system drivers of potential differences.

Disproportionality. The DA's Office receives cases after law enforcement agencies make an arrest. Disproportionality exists when more people of a certain race/ethnicity are arrested than we would expect given the population. Race and ethnicity are captured separately in the census and the DA's case management system. The DA's case management system is limited in that individuals cannot be identified as multiracial, and it is rare that both the race and ethnicity for an individual are recorded. To help address this limitation and support an equitable approach, which aims to review and present data in ways that align with how individuals identify, we have combined the race and ethnicity fields.

When looking at cases referred to the 6th Judicial District Attorney's Office between January 2018 and June 2022, we see the following with regard to disproportionality:

With regard to Hispanic individuals:

• If only considering ethnicity, 13.3% of individuals in La Plata, San Juan, and Archuleta counties identify as Hispanic. This compares with 14.8% of individuals referred to the DA's Office.
• When considering race and ethnicity together, data suggest that 7% of individuals in La Plata, San Juan, and Archuleta counties identify primarily as Hispanic (whereas 5.1% identify as Hispanic + multiracial and 0.9% identify as Hispanic + Black, Native American, or Asian; therefore, are represented in those categories). This compares with 14.7% of individuals referred to the DA's Office.

With regard to Native American individuals:

• When considering race and ethnicity together, 5% of individuals in La Plata, San Juan, and Archuleta counties identify as Native American. This compares with 11.5% of individuals referred to the DA's Office.

Any decisions made by the DA's Office are downstream from the first decision in criminal case processing: who is arrested. Where there are racial disproportionalities in the raw number of people arrested, it follows that those disproportionate numbers flow throughout the prosecution process.

Disproportionality may not necessarily be explained by differences in criminal behavior. It can also be due to the behavior of criminal justice actors, like law enforcement practices and resource allocation that result in more people of color being stopped and arrested, or crime trends and enforcement responses in certain neighborhoods.

Disparity. Disparity exists when people who should be treated the same are treated differently. As noted, the current data dashboard presents raw differences. Any differences we see could be the result of differences in cases (e.g., severity of charges), defendants (e.g., criminal history), and/or prosecutorial practices. To support understanding and actionability, it is important to compare similarly situated defendants and similar types of cases. This will help us understand the extent of, and potential reasons for, any disparities.

Systemic Drivers of Disparities. Appropriately, prosecutors evaluate each case on its own merits. While decisions and criteria used to make decisions may seem unrelated to race on a case-by-case basis, they may be influenced by systemic drivers that are correlated with a defendant's race/ethnicity (Figure 1, below). For example, Native American and Hispanic defendants, due to historical inequities, may have unequal access to resources such as educational opportunities, formal medical diagnoses, and steady employment. These circumstances could influence factors that prosecutors consider in their decision-making, such as previous criminal history, ability to pay restitution, or engagement in drug treatment.

Figure 1. Examples of Systemic Drivers of Racial/Ethnic Disparities

Likewise, prosecutors may have different interpretations of defendants' life circumstances and experiences. Defendants may have diverse expressions of concepts such as remorse, respect, or compliance, which may or may not align with prosecutors' expectations. Likewise, prosecutors may differ in their interpretation of a defendant's life circumstances, attitudes, or behaviors. Importantly, our methodology cannot be used to support or refute possible implicit or explicit bias.

We hope this analysis will prompt a conversation about what systemic drivers of racial/ethnic disparities exist and how a DA's Office might work to address them. In reviewing the results we encourage you to take a systemic perspective, considering the variety of potential drivers noted above.

Analysis Focus

This analysis focused on the outcomes of prosecutorial decision making in the DA's Office from January 2018 through the end of June 2022. We assessed the extent of racial and ethnic disparities across the following four decision points: (1) declination of felony referrals; (2) dispositions (dismissal, deferred judgment, and plea agreement) for felony and misdemeanor offenses; (3) charge reduction from filing to disposition for felony and misdemeanor cases; and (4) imposition of an incarceration sentence (Figure 2, below). We prioritized the areas where prosecutors have the greatest direct influence and where we had accurate and reliable data.

Figure 2. Decision Points Analyzed

A more comprehensive overview of how a case moves through the system can be found on the Criminal Justice Process page.

As noted above, for this analysis, we combined the race and ethnicity fields. Due to known challenges in accurately collecting Hispanic ethnicity, we used defendants' last name, linked with census data, to help us better identify Hispanic individuals. See Terms, Methods, and Limitations for more information.

This analysis aims to prompt discussion and raise questions, rather than provide definitive answers. To support this aim, we present results as predicted probabilities: an estimate of the likelihood of the outcome, based on the defendant's race/ethnicity, while taking into account individual and case factors. Information on statistical significance, which is heavily influenced by sample size, can be found in the technical appendix.

Key Takeaways

• Deferred Judgments: While the predicted probability of deferred judgments as an overall category appear to be distributed fairly evenly among White individuals (24.7%), Hispanic individuals (22.6%), and Native American individuals (21.2%), we see a much higher level of variation between White and Native American individuals for DUIs (29.3% vs. 22.4%) and Sex Offenses (27% vs. 20.9%).
• Pleading Guilty: Regarding the predicted probability of defendants pleading guilty to the top charge, Native American individuals (44.2%) are more likely to do so than White individuals (37.5%). We see the same trend when examining the predicted probability of pleading guilty to felony level charges (Native American individuals - 62% vs. White individuals - 54.6%), and to DUIs (Native American individuals - 68.7% vs. White individuals - 60.6%).
• Sentences to Incarceration: We are seeing a clear disparity regarding Hispanic and Native American individuals being more likely than White individuals to receive a sentence to incarceration. For felonies, the predicted probability of incarceration is 57.3% for Hispanic individuals. The predicted probability is 7–8% lower for White individuals (50.0%) and Native American individuals (48.8%). As to misdemeanors, the disparity affects Native American individuals, with the predicted probability of incarceration at 25.9%, while it is lower for Hispanic (21.6%) and White (18.4%) individuals.

Actionability

• In regards to deferred judgments, we intend to re-examine the eligibility criteria for those who receive deferred judgments for DUIs and sex offenses. Having more clear guidelines will likely reduce the disparity we see in the outcome of these crimes.
• The disparity in Native American individuals pleading guilty can be addressed by ensuring these cases receive full process available under the law, encouraging these defendants to seek counsel, and making sure that the offers in these situations are the same that a person with counsel would receive.
• As to the increased likelihood that Hispanic and Native American individuals are sentenced to incarceration, our office will work to offer a broader range of possible sentences to the court and increase the breadth of information we consider prior to making our sentencing recommendations to the court.

Felony Declination

The DA's Office decides whether or not to prosecute felony referrals — formal accusations that a specific person has committed a specific crime. In this section, we consider instances where the DA's Office has declined to file charges following a felony referral.

In the 6th Judicial District, 3,162 felony referrals were made for White, Hispanic, and Native American individuals between January 1, 2018 to June 30, 2022. The DA's Office declined to file charges for 5.5% (173) of these. The Office declined to file charges in 6.6% (142) of felony referrals involving White individuals, 4.7% (25) involving Hispanic individuals, and 1.3% (6) involving Native American individuals. These represent raw rates: any differences we see could be due to differences in individual or case characteristics.

Disposition

The rest of the report considers felonies and misdemeanors filed between January 1, 2018 and June 30, 2022 for individuals identified as White, Hispanic, or Native American. Our sample includes 11,057 cases that were disposed of during that time frame. The racial/ethnic breakdown of defendants in our sample was: 73.1% White (8,083), 14.9% Hispanic (1,650), and 12.0% Native American (1,324).

Case outcomes. Of the cases in our sample, 38.6% were resolved through plea agreements, 24.5% through dismissal, 23.9% through deferred judgments, 11.6% through a global plea, and 1.3% through a trial. Data on other outcomes, such as diversion, are not available.

Differences Among Defendants. Hispanic individuals and Native American individuals were younger than White individuals. Hispanic individuals were more likely to have a criminal history. It took slightly longer to resolve cases for Hispanic individuals.

More on differences among defendants

• Age and gender: In general, Hispanic and Native American individuals were younger than White individuals. For example, 32.6% of Native American individuals and 28.7% of Hispanic individuals were under the age of 26, compared to 23.9% of White individuals. Native American individuals were less likely to be male (65.4%), compared to Hispanic (75.2%) and White (73.0%) individuals.
• Criminal history: Hispanic individuals were more likely to have a criminal history. 57.5% of Hispanic individuals had no prior convictions, compared to 62.0% of Native American and 63.3% of White individuals. A greater proportion of Native American individuals (25.2%) had prior misdemeanor convictions only, followed by Hispanic individuals (24.2%) and White individuals (23.9%). A greater proportion of Hispanic individuals (18.2%) had prior felony convictions, compared to White and Native American individuals (both 12.8%).
• Charge level and charge type: Cases involving Native American and Hispanic individuals were more likely to include a felony charge (21.8% and 20.1%), compared to those involving White individuals (17.8%). In looking at charge type, cases involving Native American individuals had a higher percentage of person or sex charges. Cases involving White and Hispanic individuals had a higher percentage of traffic charges.
• Case Length: It took longer to resolve cases involving Hispanic individuals, with a mean of 4.8 months (standard deviation [SD]: 5.1) compared to 4.5 months for White individuals (SD: 4.9) and 4.3 months for Native American individuals (SD: 4.2).

Dismissal

A case is dismissed when the criminal charges are terminated, either by the court or by the prosecutor. There can be several reasons why a case is dismissed, including: a lack of evidence or unavailability of a witness. Cases may also be noted as dismissed if they are referred to, or successfully complete, a diversion program.

Total Dismissed. Our sample included 2,713 cases that were dismissed. Dismissals accounted for 24.5% of case dispositions in the sample. Overall, 25.3% (2,044) of cases involving White individuals, 24.4% (403) of cases involving Hispanic individuals, and 20.1% (266) involving Native American individuals were dismissed. These represent raw rates: any differences we see could be due to differences in individual or case characteristics.

Deferred Judgment

A deferred judgment is an alternative to traditional prosecution that attempts to address individuals' needs and to offer alternatives such as useful public service, probation, payment of restitution, or counseling or treatment related to their case. The defendant enters a temporary guilty plea, and, if they comply with the terms, their guilty plea is withdrawn and the case is dismissed.

Total Deferred. Our sample included 2,646 cases that were deferred. Deferrals accounted for 23.9% of case dispositions in the sample. Overall, 24.5% (1,977) of cases involving White individuals, 21.8% (359) of cases involving Hispanic individuals, and 23.4% (310) of cases involving Native American individuals were deferred. These represent raw rates: any differences we see could be due to differences in individual or case characteristics.

Plead Guilty

An individual pleads guilty when they admit a factual basis for the plea and acknowledge guilt for a charge, sometimes in exchange for a more lenient sentence.

Total Plead Guilty. Our sample included 4,271 cases that resulted in a guilty plea. Guilty pleas accounted for 38.6% of case dispositions in the sample. Overall, 44.3% (587) of cases involving Native American individuals, 39.0% (644) of cases involving Hispanic individuals, and 37.6% (3,040) of cases involving White individuals resulted in a guilty plea. These represent raw rates: any differences we see could be due to differences in individual or case characteristics.

Charge Reduction

After a prosecutor files a case, the top charge in the case may change over time as some charges are dismissed or amended between filing and disposition. In this section, we considered reductions in the severity of charges from initial filing to disposition for cases that resulted in a guilty plea.

Total Charge Reduction. Of the 4,271 cases that resulted in a guilty plea, 52.2% had no charge reduction, 19.6% had a within charge level reduction (either from a more severe felony to a less severe felony, or from a more severe misdemeanor to a less severe misdemeanor), and 28.2% were reduced across charge levels (from a misdemeanor to a petty offense/infraction or from a felony to misdemeanor or petty offense/infraction). Overall, the breakdown across racial/ethnic groups was:

• No Reduction: 53.0% White (1,611), 52.1% Native American (306), and 48.5% Hispanic (312).
• Within charge level reduction: 23.7% Native American (139), 22.5% Hispanic (145), and 18.3% White (555).
• Across charge level reduction: 29.0% Hispanic (187), 28.8% White (874), and 24.2% Native American (142).

These represent raw rates: any differences we see could be due to differences in individual or case characteristics.

Sentenced to Incarceration

After an individual is found guilty of a crime, a judge imposes a sentence which may include fees, fines, community service, probation, jail, community corrections, or prison. Prosecutors and defense attorneys can negotiate plea bargains or make sentencing recommendations to the judge, who decides on the ultimate sentence. Incarceration includes any jail sentence (with or without probation), community corrections, or prison.

Total Incarceration. Of the 4,271 cases that resulted in a guilty plea, 27.8% had an incarcerative sentence. Overall, 34.9% (205) of cases involving Native American individuals, 30.0% (193) of cases involving Hispanic individuals, and 25.9% (788) of cases involving White individuals resulted in an incarcerative sentence. These represent raw rates: any differences we see could be due to differences in individual or case characteristics.

Conclusion

The fair and just treatment of all individuals at each stage of the criminal justice process is of significant importance. We hope this analysis will prompt a conversation about what systemic drivers of racial/ethnic disparities exist and how a DA's Office might work to address them. We welcome your reflections on the findings and potential next steps.

Terms, Methods, and Limitations

Terms

Case: A collection of charges against a defendant arising out of a single incident. In this analysis, cases are classified by their most serious charge.

Controlling: Taking other factors into account when conducting an analysis. For example, "after controlling for age, gender, criminal history" means that we are taking potential differences in these factors into account when examining the differences between White, Native American, and Hispanic defendants.

Predicted Probability: An estimate of the likelihood of the outcome, based on the defendant's race/ethnicity, while taking into account the control variables.

Standard Deviation: A measure of how dispersed the data are in relation to the mean. Low standard deviation means data are clustered around the mean, and high standard deviation indicates data are more spread out.

Methods

Data Source

All data was drawn from the Action case management system, stored and shared by the Colorado District Attorneys' Council (CDAC).

Variable Construction

Race/Ethnicity: As outlined in the data dashboard, race and ethnicity are determined by law enforcement. For this analysis, we combined the race and ethnicity fields. If an individual was identified as White and Hispanic, we categorized them as Hispanic. If the individual's race was identified as Black, we categorized them as Black, regardless of their ethnicity. If the individual's race was identified as Native American, we categorized them as Native American, regardless of their ethnicity. Because we believe that Hispanic individuals were systematically miscategorized as White in the dataset, we used the defendant's last name to help identify their ethnicity. Based on procedures employed by the Colorado Department of Public Safety in their CLEAR Act reporting, we recategorized any individual as Hispanic who met the following criteria: 1) their race was identified as "White," "other," or their race was missing and 2) the 2010 census file (surnames occurring 100 or more times) identified their surname as having 85% or more individuals with that surname as "Hispanic or Latino". This procedure resulted in the categorization of an additional 1,325 (8.2%) cases with Hispanic defendants.

Criminal History: As outlined in the data dashboard, we calculated criminal history based on convictions (since 2007) within the 8 District Attorneys' Offices participating in the Colorado Prosecutorial Dashboards Project (the 1st, 2nd, 5th, 6th, 7th, 8th, 18th, and 20th). We developed four categories: a) no criminal history, b) prior misdemeanor convictions, c) prior non-violent felony convictions, and d) at least one prior violent violent crime felony conviction. The definition of violent was aligned with the definition used in the dashboard.

Charge Type: As defined in the data dashboard, we classified cases by their top charge into the following categories: person or sex, property, drug, driving under the influence (DUI), traffic, weapons, or other.

Age: We used the following age categories: under 18, 18–25, 26–35, 36–45, and over 45 years old.

Gender: Law enforcement defines gender in the following categories: male, female, other. Due to the small number of individuals identified as "other," we limited our analysis to individuals identified as male or female.

Charge Level: Charge represents the most serious filed charge, categorized as a felony or a misdemeanor.

Charge Class: Charge class represents the most serious filed charge, which we categorized as follows: a) felony 1–4 or drug felony 1–3, b) felony 5–6 or drug felony 4, c) misdemeanor 1–2 or drug misdemeanor 1–2, d) misdemeanor 3, or e) traffic misdemeanor 1–2.

Case Length: We calculated the case length as the number of months to case resolution, using the date the case was filed and the date the case was disposed of. We treated values less than zero or more than five years as missing.

Disposition Quarter: To account for time trends, we constructed a categorical variable representing the calendar year quarter the case was disposed.

Felony Declination: We identified all felony referrals for which the DA's Office declined to file charges.

Case Disposition (Dismissal): We identified all cases that had a disposition of "dismissed" for the most serious filed charge. We did not include plea dismissals (a defendant's case/cases dismissed in exchange for pleading guilty to another case/cases).

Case Disposition (Deferred Judgment): We identified all cases for which a defendant received a deferred judgment to the most serious filed charge.

Case Disposition (Plead Guilty): We identified all cases in which the defendant plead guilty to the most serious filed charge.

Charge Reduction: We developed three ordered categories of charge reduction from filing to disposition: a) no reduction b) within class reduction (for example, a felony 1 to a felony 3 or a misdemeanor 2 to a misdemeanor 4), and c) a charge level reduction (a reduction from a misdemeanor to a petty offense/infraction or a reduction from a felony to a misdemeanor or petty offense/infraction).

Referral Reduced at Charging: Referral charge reduction represents whether the felony referred was reduced to a misdemeanor at point of filing.

Incarceration: As defined in the data dashboard, a sentence to incarceration included any of the following sentences: state prison, youth corrections, community corrections, jail, or any sentence of jail and probation, as well as inmate/outmate programs, work release, in-home detention, and weekenders.

Census Race and Ethnicity

The District's Population was generated from information publicly available through the United States Census Bureau. The 2020 U.S. Census collected race and ethnicity as two questions. In combining the race and ethnicity questions, we used the following logic: If an individual identified as White and Hispanic, we categorized them as Hispanic. If an individual identified as Black, we categorized them as Black, regardless of their ethnicity. If an individual identified as Native American ("American Indian or Alaska Native" or "Native Hawaiian and Other Pacific Islander") we categorized them as Native American, regardless of their ethnicity. Because data suggest that individuals who identify as Hispanic, often report their race as "some other race," (Pew Research Center), we categorized individuals who identified as "Some Other Race" and Hispanic as Hispanic. We categorized individuals that selected two or more races, regardless of their ethnicity, as "Multiracial or Another Race."

Analytic Sample

This analysis focused on cases received by or filed in the DA's Office from January 2018 through the end of June 2022. We focused on that timeframe based on feedback from the Office that this was the period for which the most reliable data on race and ethnicity were available. All analyses were focused on the most serious charge.

We focused on defendants identified as Native American, White, and Hispanic. Due to small sample size, we excluded individuals identified as Asian (65, 0.4%) or Black (318, 1.9%). We excluded individuals whose race/ethnicity was unknown (473, 2.9%). To avoid grouping individuals with diverse identities, we also excluded individuals identified as another race/ethnicity (352, 2.2%).

To conduct a complete case analysis, we limited our analyses to individuals with complete data on all of the variables of interest. For felony declination, we excluded individuals with missing data on gender, age, and criminal history. This excluded 125 individuals (0.8%).

For analyses of disposition, charge reduction, and incarceration, in addition to excluding cases with missing data on gender, age, and criminal history, we excluded felony declination cases and open cases. We limited our analyses to cases that had either a misdemeanor or felony charge at point of filing. We excluded fugitive cases. Lastly we only kept cases with complete data on case length and disposition quarter. This excluded a total of 3,807 cases (25.6%).

Analysis Procedures

We conducted all analyses using Stata 17.0 (Statacorps, 2021). We began by conducting descriptive and bivariate analyses, examining the association between race/ethnicity and all outcomes and covariates.

We used logistic regression to examine the association between outcomes (felony declination, dismissed, deferred judgment, plead guilty, and incarceration) and race/ethnicity. For felony declination models, we included gender, age, and criminal history as covariates. For the three disposition (dismissed, deferred judgment, plead guilty) and incarceration models, we included gender, age, criminal history, case length, disposition quarter, charge type, and charge class as covariates.

We used a generalized ordinal logistic regression model to examine the association between charge reduction and race/ethnicity. We selected a generalized original logistic model because the likelihood-ratio test suggested that the proportional odds assumption was violated in the ordinal model. We included gender, age, criminal history, case length, disposition quarter, charge class, charge type, and whether the referral charge was reduced at filing as covariates.

For all models, we conducted stratified analysis, looking for potential differential impacts by charge class, charge type, the top five most frequent charges, gender, and age. When presenting results for stratified analysis, we considered categories that had the highest proportion of cases within that outcome (e.g., types of cases most frequently resolved by plea agreement), while aiming to avoid small cell sizes (<50).

To support interpretation, we used the margins command to calculate mean predicted probabilities using the sample values of the other predictor variables. Given sample sizes and equity considerations, we have chosen to report p-value in the technical appendix.

Limitations

• We do not have any charge information (charge class or charge type) for cases that are referrals declined at the point of filing (felony declination). Likewise, we do not have any information as to why the decision was made not to file a case.
• We do not have any information on why cases were dismissed or received a deferred judgment. Likewise, it is not possible to tease out whether cases were dismissed because they were referred to or successfully completed a diversion program.
• We are using a proxy measure of criminal history. As noted on the dashboard, we do not have information on convictions outside the 8 pilot DA Offices participating in the Colorado Prosecutorial Dashboard Project or cases from outside the state of Colorado. Likewise, we only have data since 2007. For these reasons, criminal history may be underestimated. However, when benchmarked against data from the Bureau of Justice Statistics, criminal history was found to be of similar magnitude.
• Race and ethnicity is reported to the DA's Office by law enforcement agencies. Law enforcement currently captures this data through various mechanisms: (1) by linking to prior criminal history records, (2) by scanning a Colorado ID or driver's license, (3) through fingerprint technology, or (4) based on the officer's "perceived demographic information of the person contacted" (as required by HB21-1250). Officer assumptions have the potential to lead to inaccurate or inconsistent data. We have attempted to correct potential under-identification of Hispanic individuals using census data. While we were able to test this correction using a small sample of individuals who self-reported race to Jefferson County Pretrial Services, we have no way to assess to what extent this correction is producing accurate results in the full dataset.
• We were not able to examine outcomes for all racial/ethnic groups. We excluded race/ethnicities which represented less than 2% of the overall defendant population. Likewise, in order to calculate reliable percentages and predicted probabilities, we limited our analyses to groups where there were more than 50 individuals of a particular race/ethnicity.
• We were not able to tease apart guilty pleas in relation to other cases; rather, all cases resulting in a guilty plea are counted individually using the top sentence on each unique case. This likely overstates the actual use of different sentences and could impact our results if having multiple cases is not evenly distributed across racial/ethnic groups.
• While we have considered a number of individual and case-level factors in our analyses, it is not possible to consider all unique aspects of the case. For example, examining charge reduction, we have not examined the types of sentences imposed, which may impact whether a defendant received a charge reduction.
• This analysis was limited in its examination of discretion points. For example, due to lack of data, we have not examined bond decisions, which prosecutors provide input on.