Opinion
96-cv-04179-VC
07-26-2024
EMMA C., et al., Plaintiffs, v. TONY THURMOND, et al., Defendants.
ORDER RE STATE'S COMPLIANCE AT PHASE 3B
VINCE CHHABRIA UNITED STATES DISTRICT JUDGE
The consent decree process has reached Phase 3B, which involves a review of the State of California's efforts to help struggling school districts get into compliance (or improve their compliance) with the Individuals with Disabilities Act. As previously noted, “neither excellence nor perfection is necessary.” Emma C. v. Thurmond, 668 F.Supp.3d at 904. Instead, the state must show that its “efforts are adequate.” Id.
The state has demonstrated that its system for intervention with struggling school districts is more than adequate under federal law. It is not a particularly close question. At each step of the intervention process, the state has shown that it provided meaningful support and technical assistance to school districts. The state has also established an escalating set of enforcement mechanisms to ensure that school districts participate in the process. The state is in compliance at Phase 3B.
This order assumes the reader is familiar with the Court's prior orders in this case, especially the order regarding compliance at Phase 3A, which describes the state's intervention plan in more detail.
I
The state has been operating under a consent decree in this case for decades. To get out from under the decree, the state must demonstrate that it has an adequate system for monitoring school districts and enforcing the requirements of the Individuals with Disabilities Education Act (IDEA). The Court established a process for assessing the state's performance. It is broken into four phases: (1) reviewing how the state collects data on school district performance; (2) reviewing how the state selects school districts for intervention based on that data; (3) reviewing the state's monitoring and enforcement activities for school districts in need of intervention; (4) addressing outstanding issues from the prior phases. At each phase, the Court determines whether the school district has demonstrated that it performs that task well enough to satisfy its obligations under federal law.
Earlier in these proceedings, the state decided that it would rather demonstrate an adequate statewide system than merely show that it was dealing adequately with the school district that spawned this litigation (Ravenswood).
In the interest of accessibility and readability, this order uses the term “school district” where the IDEA and the parties use the term “local educational agency,” or LEA. An LEA is very often the same as a school district, though not always. For example, a standalone charter school is an LEA, but would not be thought of as a school district.
The state has made it most of the way through this review process. At Phase 1, the state demonstrated that its annual statewide data collection activities enable it to effectively monitor school districts. Emma C. v. Torlakson, 621 F.Supp.3d 982 (N.D. Cal. 2018). At Phase 2, the state ultimately demonstrated that that its system for using that data to identify school districts in need of intervention is adequate. Emma C. v. Thurmond, 393 F.Supp.3d 885 (N.D. Cal. 2019); Emma C. v. Thurmond, 472 F.Supp.3d 641 (N.D. Cal. 2020). Phase 3 involves a review of the intervention itself, and it was broken into two subparts. At Phase 3A, the state demonstrated that it had a sufficient plan in place to help struggling school districts improve in meeting their obligations under the IDEA. Emma C. v. Thurmond, 668 F.Supp.3d 902 (N.D. Cal. 2023). That brings this process to Phase 3B, where the state must demonstrate that its implementation of that plan is sufficient. The state must enforce the IDEA by providing technical assistance to districts, conditioning or withholding state education funds, and requiring corrective action or improvement plans as appropriate. 34 C.F.R. § 300.600(a)(3).
To assess the state's implementation of its intervention plan, the parties selected eleven school districts to track. The school districts were not informed that their performance was being monitored as part of this litigation. However, the California Department of Education (CDE) employees who partnered with the school districts were aware of the court monitoring because they were asked to be available to testify at hearings as part of the review process. As the school districts worked their way through the intervention plan, the state produced documents and provided hearing testimony to allow the plaintiffs, the court monitor, and the Court to assess the state's performance. Over the course of this phase, the state also amended and further improved the plan itself. This resulted in the production of a tremendous number of documents providing a unique insight into struggling school districts' efforts to meet the IDEA's requirements. The Court held three days of hearings to discuss those submissions. At the hearings, which were conducted over Zoom, state officials answered questions from the plaintiffs, the monitor, and the Court. Those hearings were held on October 17, 2023, February 2, 2024, and April 19, 2024.
II
Using the data it collects on school districts, the state sets a variety of performance targets. If a district falls below a target in a given area, that raises concerns about the district's compliance with the IDEA. So, school districts that the state has determined are not meeting targets must participate in a process that the state calls “Compliance and Improvement Monitoring” (CIM). The CIM process involves a series of activities that struggling school districts must conduct across a three-year period, under the supervision of the CDE. The goal of the process is for school districts to understand the source of the issues and to create action plans for addressing them. The activities are organized into four different steps. Depending on how much a particular school district is struggling, the level of monitoring required and the support available will vary-for example, a school district that has been identified as needing greater intervention will have to complete more activities per step than a school district with a more limited set of issues. But, at a high level, the four steps are the same.
All school districts may access some resources and tools for data collection and performance improvement. The state calls that “Universal Monitoring and Support.” But school districts that are not identified as in need of intervention do not need to go through the steps and activities of the CIM process.
As another example, school districts that are closer to state targets have more limited CDE oversight of and involvement in approval of their submissions through the CIM process.
Step 1 (“Gather and inquire”) gets the process started. The school district must put together the team that will work through the CIM process activities. That includes connecting with a CDE consultant. It may also include working with a Technical Assistance provider (TA provider)-the TA providers contract with the state to offer support to school districts working to improve their provision of education to students with disabilities, and they help with the completion of the CIM process activities. But the core of Step 1 is data collection: the school district must collect both qualitative and quantitative data. The district must then analyze that data to identify problems, and it must also identify whether there is additional data that should be collected to shed further light on these problems. At the end of Step 1, the district must have identified and described its issues.
School districts with greater needs are given access to TA provider support. Other school districts get support with the CIM process through a Special Education Local Plan Area (SELPA) provider. The CIM manual filed by the state and reviewed during these proceedings describes the monitoring tiers, and the corresponding levels of oversight, support, and responsibilities, in great detail. This Court's prior order for Phase 3A also discusses that variation.
Step 2 (“Investigation”) requires the school district to move from finding problems to identifying solutions. The activities are all designed with that goal in mind. The district starts by performing a “root cause analysis” to determine the sources of the issues identified at Step 1. The district must then prioritize those root causes. Finally, the school district must create a “theory of action”-it must identify strategies that it will take to address its issues and explain why it chose those strategies, how they will help students with disabilities, and how they were informed by the data collection and root cause analysis. At the end of Step 2, the district should have a viable “theory of action.”
Step 3 (“Planning”) moves from theory to practice. The school district must determine what initiatives it will implement over the next two years to address its problems. As the state officials convincingly explained, it is impossible for struggling districts to work on every problem at the same time, and attempting to do so tends to prevent meaningful progress in any areas. Thus, the CIM process is meant to help districts identify “high leverage” actions-in other words, actions that will have the greatest impact on improving outcomes for students with disabilities. Those actions will generally be ones that can be performed at scale. And those actions will ideally be ones that target multiple issues at once, as often there will be overlap. At the end of Step 3, the district must have an “action plan” that includes timelines and milestones for improvement. Steps 1-3 should be completed within a single year.
Step 4 (“Implementation”) is when the school district gets to work at implementing its initiatives. The CDE monitors the implementation for a two-year period. The CDE is also charged with providing support and feedback during implementation. The CDE will assess whether the action plan needs revision, which may happen if the school district does not meet its timeline or milestones. After two years of monitoring, the CIM process is complete. But it is worth emphasizing that the CDE does not simply disengage with a school district at the end of a single CIM process. The whole point of these proceedings is that now the state has a system up and running for collecting data on IDEA compliance and identifying school districts in need of intervention. If a school district completes the three-year CIM process and continues to be identified, then the CDE will be able to intervene further, with a focus on new or unresolved issues.
As the CDE worked through the CIM process with the first set of school districts, it identified areas where the CIM process could be improved. That does not in any way suggest that its prior CIM process was deficient. Making changes in response to challenges in implementation is exactly what policymakers should be doing. An updated CIM Manual was filed at the end of Phase 3B. Dkt. No. 2796-2, Ex. 471.
III
The CIM process will inevitably seem abstract to someone who does not participate in it. But as mentioned, the parties and the Court followed eleven school districts through the process. In an effort to bring more concreteness to the discussion, this section will describe the journey of one school district.
School District 9 is an elementary school that was selected for intervention because of “significant disproportionality”-the district overidentified Hispanic students with specific learning disabilities, and the problem had been getting worse for the last three years. At first, School District 9 was confused about why it had to participate in the CIM process. In response, the CDE consultant assigned to that district provided greater detail about the process and explained the activities that would be required at Step 1. The CDE consultant also ensured that the TA provider was working with the district to provide support. Several trainings and virtual meetings were held with School District 9's team. There was also an in-person workshop. As the district worked on its Step 1 activity, it gave the CDE consultant access to a Google document so that he could see the progress in real-time.
School District 9's Step 1 submission demonstrated its thoughtful consideration of a substantial amount of quantitative and qualitative data. For example, the district sought parent input using both a survey and a group meeting. And the district's data drill-down revealed gaps between male Hispanic students and other students not only in special education identification but also in discipline and absenteeism. After some back and forth to ensure that the Step 1 submission involved sufficient qualitative data, the CDE consultant approved School District 9's Step 1 submission. It is worth noting that School District 9 started on Step 2 before final approval-this is not an aberration. Testifying CDE officials explained that they did not want school districts tied up while exchanging documents at the early stages. Thus, sometimes, CDE consultants would approve satisfactory but imperfect submissions, or continue working with the school districts on improving prior submissions even as the next step proceeded. The state has demonstrated that this approach is adequate.
At Step 2, a few problems arose for School District 9. Early in Step 2, the district stopped working with its TA provider. The CDE consultant learned about this not long after the relationship broke down. The CDE consultant was informed that the district was looking for more hands-on, involved support from its TA provider-that is understandable, as this was the district's first time facing significant disproportionality issues and working through the CIM process. The CDE consultant worked with his colleagues to identify other TA providers who might better fit School District 9's needs, and he made some recommendations. And the district ended up selecting a new provider.
This does raise some concerns: What would have happened if School District 9 had been less effective at advocating for itself? How effective are school districts at selecting their TA provider when they do not fully understand the CIM process or their needs? How much insight does the CDE consultant have into the relationship between the school district and the TA provider? But those are interpersonal and inter-entity dynamics that can be worked out. And the story of School District 9 helps demonstrate that. Within two months, the CDE consultant learned that the district was dissatisfied with its TA provider, the consultant asked questions to understand the mismatch and looked into options that would be a better fit, and the school district selected a new TA provider that it seems to have worked well with through the rest of the CIM process steps.
The next problem School District 9 faced was more substantive. The CDE consultant initially returned the district's Step 2 submission without approval. This was in part because a problem that was called out at Step 1 was left unaddressed by the Step 2 plan. That reveals thoughtful evaluation by a CDE consultant-the review of the submissions is about more than just box checking, it is about tying the steps together. The return without approval was also in part because the work that School District 9 had done with its new TA provider was not included in the submission. Then, the district provided all of that work to the CDE consultant in full. While the information was rich and detailed, it did not provide the high-level analysis and conclusions that Step 2 drives toward. The CDE consultant provided feedback to the district and the TA provider on how to document the Step 2 work “in a concise, straightforward manner that provides narratives for the activities they have completed and information they discovered along the way.” In the meantime, the district was permitted to get started on Step 3. When the Step 2 submission was approved, the district's theory of action called for implementing a multi-tiered system of supports for students, improving its English language development curriculum and discipline matrix, and providing additional coaching to staff. The monitor described the approved theory of action as “thoughtful and robust.”
Step 3 then went well for School District 9. It developed an action plan with three strategies that follow from its theory of action: the multitiered support system, a new behavior matrix, and an equity initiative aimed at creating a “culture of belonging” for students and parents. The plaintiffs noted that the plan “includes qualitative and quantitative data for each root cause, as well as planned activities that are clearly connected to the articulated goals.” The monitor described the strategy associated with each activity as “reasonable.”
To be fair, School District 9 seems to have had more success than some other districts did in the CIM process. But its journey is a useful example. It shows that the support offered by the state is meaningful. It also shows that the evaluations performed by the state are thoughtful and help the districts get the most out of the process-the state is not just rubber stamping these submissions. It also provides insight into what happens when the system doesn't work precisely as planned, like when a school district falls behind or a TA provider relationship falls apart. The state has shown that it is capable of adapting and responding to those challenges.
IV
The plaintiffs and monitor provided valuable feedback to the CDE as it implemented and amended its CIM process throughout Phase 3B (and, for that matter, Phase 3A). Many of the issues that the plaintiffs and monitor flagged were resolved by later submissions or testimony. However, at the end of Phase 3B, the plaintiffs still ask the Court to withhold a compliance finding. Ultimately, none of the issues that the plaintiffs raise justify a conclusion that the state's monitoring and enforcement system is inadequate under the IDEA and its regulations.
As discussed, many of the issues raised in Phase 3B were resolved by the state either amending its system or providing further information. For example, one recurring issue was whether and how the CDE would hold school districts accountable if they failed to participate, or failed to participate meaningfully, in the CIM process. In response to these concerns, the CDE provided an accountability matrix that specifies an escalating series of responses and consequences. The CDE officials also thoughtfully testified about the causes they have seen for a lack of cooperation and how they are working to address those causes.
Another issue was the role of the TA providers, who offer support to many school districts during the CIM process. The plaintiffs asked important questions about their training, their capacity, the process for selecting and overseeing them, and how they work with school districts. This resulted in the production of records from the TA providers who worked with the school districts tracked through this litigation. Two directors of TA providers also testified and answered questions about their work generally and their involvement in the CIM process. In the end, the Court was left with no serious concerns about the TA providers.
Additionally, the plaintiffs were dissatisfied with the CIM process for small school districts. Technically, that concern has not yet been addressed-but the state is currently revising its CIM process for small school districts, and the parties have agreed to consider that revised process at Phase 4.
A couple of issues were not fully addressed by the state, but they don't come close to putting the state out of compliance at Phase 3B. First, there were issues with inconsistency in when the CDE approved submissions versus returning them for further work. At times, the rubric offered by the state that the CDE consultant would have used to determine whether approval was appropriate seemed to counsel against the conclusion that the CDE consultant reached. That is a real issue. The CIM process is far less likely to lead to beneficial results for students without active engagement from CDE consultants with relatively high standards, and that has to be true across the board. However, as the experience of School District 9 revealed, too much gatekeeping at each stage can stymie a district's progress through the CIM for limited benefit, and so CDE consultants may make different determinations about when to let a district proceed or when to hold it back. As the hearing testimony revealed, the CDE consultants are trained to make these determinations and they do so in collaboration with their supervisors and colleagues. Moreover, the CDE has created rubrics that specify the criteria that must be satisfied for approval at each stage. Those rubrics were not in use for the entirety of the Phase 3B process, but they will be moving forward.
Precisely how much flexibility CDE consultants have when approving a submission or allowing a school district to proceed to the next step remains unclear. That is in part because the CDE has been changing the CIM process in response to these proceedings and its experience. But it is also, very likely, in part because sometimes CDE consultants will fail to perfectly implement the CIM process. That is inevitable. But considering the eleven school districts evaluated at Phase 3B, the state has shown that it is adequately (if imperfectly) implementing a process to intervene with struggling school districts and bring them into IDEA compliance. And that is enough.
Second, many of the school districts in the CIM process are struggling in multiple areas. For example, School District 9 was identified for intervention because of its significant disproportionality, but it also failed to meet state-level targets for educating students in the “least restrictive environment” (LRE). The monitor expressed concerns throughout Phase 3B about how submissions in the CIM process focused only on some of the areas that the school district was deficient but not on others. This is also a serious issue. It is important that the takeaways from the state's monitoring and selection of school districts translate into the intervention and enforcement process. And the interaction between multiple deficient metrics can supercharge an issue. Take again School District 9-the data suggests that Hispanic students are being overidentified as students with disabilities, and then also being educated in overly restrictive environments. Nonetheless, School District 9 focused on disproportionality during the CIM process, not the LRE issue. As another example, take School District 1, which did not meet statelevel targets for “child find.” When School District 1 did its data drill-down, it determined that its child find numbers were low largely because of its initiatives for early intervention and provision of resources. Given that, School District 1 decided to focus on activities to target other issues revealed through the CIM process, including implementing a multi-tiered system of support. The CDE did not force School District 1 to focus on child find. Instead, there was a discussion that included the CDE consultant and the Special Education director for School District 1, and they agreed that School District 1's approach was appropriate. That does underscore, to an extent, the concern: a school can enter the monitoring and improvement system because of failure to meet a particular target, and then work through the monitoring and improvement system without addressing that target.
The CDE responded to this concern by emphasizing that prioritization is a key part of the CIM process. When it comes to significantly disproportionate school districts, the CDE has determined that the disproportionality is the priority-although it hopes that the kinds of interventions that address disproportionality, like early intervention in the general education classroom, will positively affect students and impact other indicators. Generally, though, when it comes to school districts facing multiple issues, the choice will be up to the school districts through the CIM process to prioritize and identify high-leverage activities. And if the CDE consultant does not understand or agree with the decisions made by the school districts about root causes or what to focus on, then the CDE consultant can decide whether to make the school district change course. But the state has reasonably explained that it would rather give school districts that discretion within this system than impose a requirement across all school districts about what must be targeted. Moreover, if a school district implements its action plan and then is identified again for the same failure to meet state-level targets, then further intervention on that target will follow. These decisions about how to run the CIM process are reasonable decisions for the state to make to enforce the requirements of the IDEA.
V
As initially conceived, Phase 4 would involve the documentation and formalization of the processes developed in Phases 1-3. But, as discussed at the April hearing, that no longer seems wise. The state has already produced a substantial amount of documentation, both for internal use and for use by school districts. It would be mere busy work to require further formalization just for this litigation. With the agreement of all parties, Phase 4 will have a different focus. A few issues remain outstanding, from this phase and prior phases, and those issues will be addressed in Phase 4. The stipulated list provided by the parties is appropriate. Dkt. No. 2841. The subjects of Phase 4 will thus be (1) the State's plan for the inclusion of IEP implementation data in selection for, and the process of, CIM; (2) the State's plan for the inclusion of restraint and seclusion data in selection for, and the process of, CIM; and (3) information regarding small LEAs' selection for monitoring and the updated rubrics for small LEAs.
The state has made striking improvements to its system for monitoring and enforcing IDEA compliance. While its system is not perfect, it is a world away from what the state was doing before working with the Court, the plaintiffs, and the monitor to get into compliance with the IDEA. Students with disabilities are served well by this system. And the state has demonstrated its capacity to adapt and respond as issues arise, to continue to better serve its students. Every consent decree process must have an exit door. In this case, the state appears to be barreling towards it.
IT IS SO ORDERED.