Referring Cases : A Diversion Toolkit for Communities by the Restorative Justice Project
Step 1: Establish a Foundation
A. Youth Criminalization
B. People Harmed
C. Restorative Justice
D. Restorative Justice Diversion
E. The Evidence
F. Interactive Learning
Step 2: Build the Program
A. Program Fit
B. Community Held
C. Community Vision
D. Funding
E. Common Ground
F. Referring Cases
G. Receiving Cases
Step 3: Sign-up for Training
You're 0% Complete

Step 2F: Referring Cases

What Kinds of Cases Should We Receive?

Create the eligibility requirements for cases and establish shared expectations with system partners going forward. Set up pre-charge referrals from the juvenile legal system to your restorative justice diversion program.

Coming off Step 2E: Common Ground, hopefully you’ve begun meeting with folks at the DAO and have some sort of informal buy-in from them for this program and what it could look like in your community. That is amazing and is a feat in its own right! You are building something HUGE from the ground up, so don’t forget to celebrate yourself at every step of the way.

Now that you’ve garnered their interest, we suggest building on this momentum. Up until this point, your conversations on RJD should’ve been more big picture and focused mostly on building relationship and trust. Once you’ve built that solid foundation, it’ll be a lot easier to move into more of the details and minutiae of what this process is going to look and feel like in your community. This section should help you understand how to start identifying and building out the eligibility criteria and referral process to determine what cases can and should be diverted and what that process will look like internally.

Identify Current Data

As you well know, RJD is intended to create accountability to survivors’ self-identified needs, while also ending racial and ethnic disparities in the juvenile legal system. Therefore, the types of cases that are ideal for diversion are those with a clear person harmed and those crimes which most often result in young people of color being incarcerated or placed on probation. To ensure this is the case, we use data to inform the eligibility criteria of the young people who will be referred to RJD. Most likely the DAO or another closely related agency such as probation holds the data you need for this process.

Download Worksheet: Local Youth Justice Landscape – Data

The specific data you need for this process can be found on your Local Youth Justice Landscape Data worksheet from Step 2C: Community Vision. Using your previous research, you should already have an idea of what the county data is going to look like, but it’s always important to get the numbers directly from the county too, if available. Download a clean copy of the worksheet and use the county numbers to fill out:

  • The current population of youth in your county (ages 10-18) broken down by race and sex
  • The top 10 misdemeanors and felonies youth (ages 10-18) are arrested for that have an identifiable person harmed, broken down by race and sex of the youth
  • The top 10 misdemeanors and felonies youth (ages 10-18) are charged with that have an identifiable person harmed, by youth’s race and sex
  • The top 10 misdemeanors and felonies youth (ages 10-18) are convicted/adjudicated delinquent for that have an identifiable person harmed, by youth’s race and sex
  • The top 10 misdemeanors and felonies youth (ages 10-18) are placed on probation for that have an identifiable person harmed, by youth’s race and sex
  • The top 10 misdemeanors and felonies youth (ages 10-18) are detained for that have an identifiable person harmed, by youth’s race and sex
  • The top 10 misdemeanors and felonies that have the highest rate of recidivism for youth (ages 10-18) and have an identifiable person harmed, by youth’s race and sex
  • The top 10 zip codes from where youth (ages 10-18) are:
    • Arrested
    • Charged
    • Convicted/adjudicated delinquent
    • Detained
Download Worksheet: Local Youth Justice Landscape – Programs, Policies, and Boards

Additionally, you’ll want to refer to your Local Youth Justice Landscape – Programs, Policies, & Boards worksheet, downloadable above. At this point, your focus should be on the existing diversion programs in your county, if any exist. Work with your system partners to fill in whatever missing information you may have. Specifically, you want to know:

  • Which of these offenses are diversion eligible for already existing pre-charge diversion and post-charge diversion?
  • If they have pre-filing diversion, what percentage of pre-filing diversion eligible offenses are diverted?
  • What is the referral criteria for existing pre-filing diversion programs (# of priors, types of priors, age of youth, and other limiting criteria)

Obviously, this is a lot of information. We’ve had a wide array of responses from system partners when it comes to their willingness to share their data. Some folks have been very eager to share their data, while some feel neutral about it and others are resistant to the idea. We’ve even done work in jurisdictions that don’t have a centralized place for storing their county-wide data. Additionally, we’ve had a similar array of responses when it comes to analyzing the data.


In the spirit of transparency, sharing a copy of both worksheets (that are blank!) with the folks you are working with can be helpful for them to visually see what you are asking for and where the ask is coming from. Often people are asking for data to sue county entities, or to show that the system has erred or failed. Being overly communicative and collaborative during this process in particular can also be helpful since it can be a sensitive subject. You may want to ask system folks if they want to fill out the Landscape worksheets together or schedule a meeting once you have mapped it out yourself.

If you’re finding yourself in situations where system partners are resistant to share their data, try to figure out where that hesitancy is coming from. Is it because it may be a lot of work for them to compile it all? Or maybe they don’t want to share their data with someone outside the system? The data will most likely show stark disparities across racial and ethnic lines. In some cases, this can be really hard for staff within the DAO to face. Again, it’s important to remember that this is not an opportunity for you to expose the shortcomings of the criminal legal system or to caste blame on any group of people for these disparities. We know the underlying cause is systemic. Rather, you’ll want to find a way to align with your partners and let them know you are not going to use this data for a “gotcha” moment of what’s wrong with the system or with them. Instead, show them you are approaching it with the point of view of “we know your hands are tied because your process isn’t resourced to…” Sometimes it’s just a matter of ‘flipping the script’ on how you present information so that it’s more digestible and relatable. If they had a pretty positive response to the elements of RJD and the 2-Way Expectations of CBO/SP document (downloadable below) you shared in previous meetings, you can also always point back to that shared expectation, and remind them that the data is necessary for everyone to understand whether the program will be successful in fulfilling its principles.

Download Resource: Two-Way Expectations of CBO/SP


All of the juvenile legal data you need for your county may be held in multiple agencies across the system. Being able to identify in advance where all this data is held and how many different agencies you’ll need to contact can save you a lot of time and energy in the long run. This information may come out in one or many of your meetings with different agencies as you’re building trust and relationship around RJD. It’s also a discussion that can come up as you’re reviewing the Two-Way Expectations of CBO/SP template from Step 2E: Common Ground. If not, you can always ask directly. And remember to get a feel for who you will need to spend extra time building relationship with by reviewing the roles and needs from the System and County Leadership Landscape worksheet in Step 2B: Community Held.

Identify Program Eligibility Criteria

With all the information and data in front of you, you can start identifying potential RJD cases! We suggest having some sort of grasp on what the data in front of you means before you meet with the DAO. You might have some idea of what cases you want to be diverted but find that the DAO may not feel comfortable with the severity of crimes you’re focusing on or with the optics of what you have in mind. So make sure you understand what ultimately is and is not a good case for diversion. You also want to be careful about selecting cases that would otherwise be diverted to another existing diversion program. Those are not the cases you want for RJD.

Download Resource: RJD Case and Participant Eligibility Worksheet

From our experience, figuring out the appropriate eligibility criteria and referral process with the DAO has been a sort of informal back-and-forth conversation where we say “here’s what looks ideal” and then the DA pushes back or agrees. These meetings can move at a fast pace and can even be a bit intimidating at first. Our RJD Case and Participant Eligibility worksheet, downloadable above, can be a great visual and accessible way to capture these conversations on paper. The worksheet maps out what the criteria for youth who are diverted will be. It covers:

  • What types of cases and arrest types will be referred (ensuring that it aligns with the data, RJD program elements, and DAO comfort level)
  • How many (if any) prior charges (the number or the type) will disqualify a young person from being referred (remember! studies show that youth who reoffend are most successful in the RJD program. So you should push for this not to be a “first time offender” program)
  • Whether dependency-delinquency youth (aka crossover youth) are eligible for referral
  • What the age range of youth who cause harm will be
  • What zip codes cases will be diverted from (remember! These should be zip codes of high arrest rates and high levels of racial and ethnic disparities)
Download Resource: RJD Case Referral Criteria Checklist

Once you’ve filled out the RJD Case and Participant Eligibility worksheet, feel free to cross-check it with the RJD Case Referral Criteria Checklist, found above, and also to add any and all additional criteria you’ve outlined to that checklist. The checklist can be a helpful resource for both the DAO and you to ensure that the cases that are being identified and referred align with the criteria you all decided on together.

A Note On Discretion

Using data to inform the program eligibility criteria satisfies both RJD core elements and also helps limit the DAO’s discretion around who gets offered diversion and who doesn’t. Studies show that despite the fact that youth of color are overrepresented in the criminal legal system, they are actually underrepresented in diversion programs. We want to make sure RJD doesn’t replicate this dynamic, and one way we do so is through limiting the DAO’s discretion.

Talking about reducing the discretion of the DAO is a very delicate topic. It is our experience that discretion is really important to folks who work in the DAO, regardless of how they use it. Some system partners use their discretion to impose even more punitive outcomes on people and others use it in the opposite way. Either way, the idea of limiting the DAO’s discretion should be navigated delicately.

A perspective we’ve used for a DA who actually wanted to use their discretion to send more serious, direct file cases, was by explaining that if the program is dependant on complete DA discretion, as the program continues to expand and receive a large volume of cases, that is going to result in a ton of work for the DA’s office. It is also helpful to be able to point to the numerous studies that have shown that, whenever there is more discretion, there is always more discrimination, even when policies are implemented in a “race neutral” way. We absolutely want to avoid situations where DAs can look at each case as it comes in and decide whether they think the youth is “good” or “worthy” or “amenable” to diversion based on nothing else but their own impressions off a case file. Implicit bias always comes into play here regardless of how reform minded or anti-racist your DA may be. Relatedly, we also shared with this specific district attorney that as much as we love them and their politics and wish they could be in office forever, there will come a time when they are no longer in office. With that in mind, it’s necessary to implement more standardized protocols and procedures so that the program’s success and the cases referred don’t depend on the character of the current DAO.

Create Case Referral Process

Creating a discretion-less referral process can be hard to do for the reasons listed above, but can be especially hard to do during the early stages of RJD implementation. In order to ensure the sustainability of the program, the number of cases referred should be explicitly based on the capacity of the CBO. In our experience, only 15 cases should be referred to the RJD program if there are two full-time facilitators (the scaling is detailed further in the 6-year RJD Program Growth worksheet, downloadable below). If RED and mass criminalization of youth of color are prevalent in your county, then the eligibility criteria you and the DAO create will more than likely result in diverting exponentially more young people than just 15 a year. Because of this, you want to find a way for the DAO to refer only 15 cases from that larger pool of eligible cases, without giving the DAO complete discretion.

Download Resource: Six-year RJD Program Growth

There are plenty of ways to do this, so feel free to get creative with your system partners in coming up with a process that feels good for everyone. When thinking through a process, here are a few things to consider:

  • You don’t want the DAO to just refer the first 15 eligible cases that come across their desk because that could mean receiving 15 cases in the first month of the program. You want to find a way to limit discretion while also keeping the success and sustainability of the program in mind.
  • Whatever process you come up with shouldn’t be permanent. It should be able to evolve (and potentially disappear altogether!) as your organization’s capacity increases.
Existing Examples

In one of our sites, the DAO uses a free randomization tool to determine which youth get referred and which youth don’t. At this site, the charging juvenile DA receives a case, makes a charging decision and if they decide this is a case they would absolutely charge, and it fits all the county’s eligibility criteria for the program, they send the case over to their legal secretary. The legal secretary uses the randomization tool (which is effectively a piece of paper that says “control, test, control, test, control, control, etc.) to determine the next steps. The randomization tool is adjusted based on capacity of the CBO. Because the CBO can only take a certain number of cases per year, the tool’s algorithm is structured to match that. For example, the CBO can only take 25 cases a year so for every 10 cases that the tool works with, it will randomly select 7 out of those 10 to divert to the CBO.

Using this computer generated randomization tool ensures less discretion and also helps prevent net-widening. Since the charging DA has no idea if the computer will refer the case or not, they have to be absolutely sure this a case they would take to court before sending it to their legal secretary. That way, if the computer decides to charge the case, they are ready to take it to court. Another added bonus of this method has been that at the earlier stages of the program using the random generator has inadvertently given the program a generated match sample of cases to compare the RJD process to, in order to measure for recidivism and other measurements of success.

What we’ve learned through this site’s process, however, is once again the importance of including ending RED in the eligibility criteria process. Because that was not explicitly done in this site, the randomization tool has resulted in RED actually increasing. Further, we’ve learned the importance of creating a plan to move away from this process in the future. There should be no reason to exclude youth who are perfectly eligible for the program once your program has the capacity to receive them.

Another site found a way to limit case referrals in the early pilot stages by creating more narrowly-focused eligibility criteria when it came to case type, which will be expanded as CBO capacity increases. For example, using data, this site picked one case type that had around the same number of charges as the number of cases that could be referred (ie. they saw that there were around 25 burglary charges and they needed 15 referrals), filtered all the young people arrested for that crime type through their remaining eligibility criteria and if it was a match, all those cases were sent to the CBO.

This process works well if you find a case type that is RJD appropriate and statistically relevant. What we’ve learned through this process is that you still need to have a process in place to stagger the case referrals. On the off chance that 15 young people are all charged with burglaries in November and all turn out to be eligible for the program, you don’t want your program receiving all 15 cases at once. Additionally, this option leaves room for the opposite to happen—maybe 25 youth were charged with burglaries last year, but then this year, there are only 10, and out of those 10 only 6 are eligible for RJD. You want to find a process that is dynamic and fluid enough to prevent overwhelming or underwhelming the RJD program.

Sign Legal Documents

Up until this point, your relationship with the DAO and other system partners has been ‘informal’ in the sense that nothing has been legally or contractually agreed upon in writing. Clearly this program is something you are all deeply invested in, but the actual process of referring cases cannot begin until:

  1. A memorandum of understanding (MOU) has been signed by the CBO and all system referring agencies
  2. A standing order has been signed by the presiding juvenile judge in your county
  3. You (the CBO) have officially been trained in all the RJD processes

Those three final steps don’t necessarily have to happen in that order. Some DAs like to sign the MOU and then start referring cases immediately, while others like to sign the MOU and give it a few months before cases actually start being handed off. There is also a chance that the DAO may not feel ready to sign any legal contract or document until you and your staff have all completed the necessary trainings and are officially ready to start receiving cases. If this is the case, it’s still worth going through all the remaining attachments, editing them accordingly, and ensuring that all sides understand what is being asked and required of each other so that once you are trained, the legal documents can be signed, and nothing will come as a surprise to folks later on.

The two legal documents that need to be signed by the DAO before a program can begin are:

Download Resource: Generic RJD DA MOU
  • A memorandum of understanding (downloadable above) is a legal contract signed by the facilitating CBO, the DAO, and any other referring system agencies (i.e. if probation will be directly referring you cases, they must sign this document). Only once this document has been signed, the DAO and other referring agencies can begin referring cases.
Download Resource: Generic RJD Standing Order
  • The standing order (downloadable above) is signed by the presiding juvenile judge of your county. This document allows for unredacted police reports to be sent to the CBO as part of the case referral. It’s important for CBOs to have access to unredacted police reports for various reasons, but most importantly it gives the contact information of all parties that were involved and impacted by the incident so that facilitators can reach out and begin the restorative process.

Both the MOU and the standing order are generic templates and are completely editable and customizable, so you’ll want to go through each document and fill in the specifics of your county. The DAO and other system partners will most likely want to go through and negotiate/change pieces of these documents, especially the MOU, based on their comfort level, politics, etc. We strongly suggest you do not go into any negotiation or modification meetings until you have read through each of these documents extensively and, better yet, gone through them with a lawyer. Using your Youth Justice Landscapes, identify allied youth justice lawyers who can sit down and explain these documents to you and potentially even join you in the negotiations. You want to come into negotiation meetings with a clear understanding of what the document is asking for, what the DA wants to change, and whether or not that interferes with the values and principles of the program.

As you’re reviewing both documents, there are a couple things we’d like to emphasize, especially with regards to the MOU. First, we’re intentional about keeping this document a bit broad and vague. We don’t include the specific zip codes or specific types of cases or referral criteria in the document because we want the program to evolve and grow without having to re-open this contract every time a change is made. The hope is that this document will be signed in perpetuity so that as zip codes and case types expand (in severity and volume) you can do so more easily than reconvening all signatories on the document, making alterations and then resigning the document. The elected district attorney who originally signed the MOU will not be in office forever. You can never know who the future DA of your county will be and what their opinions on the program or certain clauses may be. We don’t want to risk re-opening a document and having certain clauses be up for negotiation again based on the character of the current elected official. Instead, the idea is to have separate internal documents that outline the specifics of eligibility criteria, referral process, etc. so that the overall idea of the diversion program can live in the MOU and will hopefully, over time, become more integrated with the internal structures, procedures, and protocols of the DAO—making it harder for incoming DAs to get rid of it.

Second, other than allowing cases to start being legally diverted to the CBO, the main chunk of the MOU is to ensure and maintain confidentiality of every single participant in the process from the point of referral through the end of the process, regardless of outcome. Maintaining the confidentiality of all participants involved is the cornerstone of the RJD program. Without the protection of knowing that whatever is said throughout the entire process is held in complete confidence, the depth, authenticity, and genuine transformation that comes from this process just won’t happen.


Including many voices will help you get the most accurate assessment. It’s highly recommended that multiple people at your organization contribute to filling out this questionnaire. You could complete it collaboratively or fill it out separately and share your responses together. Also, adding an RJD program to your existing programming could impact your current staff, so it’s a good idea to include everyone early on.

Introducing, reviewing, and negotiating the MOU and Standing Order to your system partners can be a somewhat lengthy process. At the end of it all, there’s a possibility that they may not feel comfortable signing it right away and would instead prefer you be trained and ready to receive cases the second they sign it. Regardless of whether they formally commit to this program via signing the legal documents, or they maintain their informal commitment via not signing the legal documents just yet, head on over to Step 2G: Receiving Cases, to make sure you have everything else in order before you request an RCC training!

What If…?


A Challenging Prerequisite

One organization faced a huge challenge during the initial stages of starting an RJD program when staff found out that the county didn’t collect essential data disaggregated by race, gender, offense type, offense level (felony/misdemeanor) and zip code. This data is necessary so that the system partners and the community-based organization (CBO) can determine which cases are eligible for a pre-charge diversion program, and to be sure that the program is actually being effective and not net-widening. One way this challenge was resolved was through reaching out to each local law enforcement agency in which the data could be collected. The other approach was to reach out to the state attorney general’s office. While they were unable to help directly, they told the organization about their portal that gives access to county statistics. While this information had always been available online, it wasn’t easy to find. The state attorney general’s office offered this information because the organization led their request with the fact that they were partnering with the county district attorney’s office (DAO) on this project. What helped open these doors was the state attorney’s comfort with the organization because of the shared relationship with the DAO’s juvenile legal division contact person. While such connections aren’t always available or possible, when they do exist, they can be leveraged in this way. In this county, as a result, the cross collaboration has led to a community-system partnership in which the DAO provides transparency to the community about critical information such as the impact of this program on communities of color and survivors, which has helped move the implementation plan forward

Tools & Resources

Worksheet: Local Youth Justice Landscape – Data

Worksheet: Local Youth Justice Landscape – Programs, Policies, and Boards

Template: Two-Way Expectations of CBO/SP

Worksheet: Establishing RJD Case & Participant Eligibility

Resource: RJD Case Referral Criteria Checklist

Resource: 6-year RJD Program Growth

Template: MOU Template

Template: Standing Order Template

2F Checklist

RECEIVE and ANALYZE county data

DEVELOP eligibility criteria with DAO using RJD Case Eligibility Setting worksheet

CREATE referral process with DAO

REVIEW the MOU and the standing order with a youth justice lawyer

INTRODUCE the MOU and the standing order to relevant system partners

SIGN the MOU and the standing order

Next 2G:

Receiving Cases

Are We Ready to be Trained to Receive Cases?