Reading the Catalog
Every metric carries a visibility status and a disclosure tier. Together they say: who can see this today, and who should be allowed to.
Visible
Published or publishable today. Aggregate, de-identified, low re-identification risk.
Hidden
Data exists but is fragmented, restricted, or institutionally withheld. Defensible to expose with governance.
Missing
Not tracked by any system. Requires new infrastructure — often user-generated — to measure.
Select a Role
Each stakeholder sees a different slice of the system. Click a role to view its metric catalog. Then click any metric for the full methodology.
Endorse Adoption
The methodology is published for public review. Endorse the metrics you believe should be adopted. Endorsements are aggregated and presented to the Methodology Board and state partners.
State Comparison
How publicly available family court metrics compare across peer states. Missouri is the pilot implementation; peer data is illustrative, drawn from state judiciary reports.
ASSUMPTION: Peer-state data below is representative of public judiciary reporting patterns. Figures should be verified against each state's most recent Office of Court Administration annual report before policy use. This view demonstrates the schema; production data would be refreshed quarterly.
| Metric | Missouri | California | New York | Texas | Illinois | Florida |
|---|
Reform Implications
What the comparison reveals about where national standardization is possible and where state-level innovation must lead.
No state publishes judicial-level outcome data.
Every state ranks zero on individual-judge transparency. This is the single most consistent gap in the country and the single most defensible reform target — if paired with governance protections against adversarial use.
Compliance data does not exist anywhere.
No state tracks whether court-ordered parenting time is actually exercised. This is the "truth layer" — and the space where user-generated data from platforms like CoTrackPro becomes the only viable source.
Disposition time is standardized; quality is not.
Most states report time-to-disposition. Almost none report durability of outcomes — whether the agreements or rulings hold. This creates perverse optimization: fast rulings that return to court repeatedly.
Disparity analysis is politically radioactive.
Outcomes by income, race, and representation status are virtually absent from public reporting. Academic DUAs are the only viable path; legislative mandates face predictable resistance.
Technical Specification
Architecture for a metric publication and endorsement platform. AWS-native, least-privilege, privacy-preserving by default.
Core Components
TypeScript across the board. API Gateway fronts Lambda handlers. DynamoDB single-table for operational data. S3 for published data snapshots and research DUA exports. KMS CMK per tier.
Frontend: React + Vite + Tailwind API: API Gateway + Lambda (TS) Data: DynamoDB single-table Storage: S3 (tiered buckets per class) Auth: Cognito (anon + identity pools) Secrets: SSM SecureString Analytics: CloudWatch + custom audit table
De-identification Pipeline
Every metric publication passes through four gates before release. Nothing skips. Nothing is ever published directly from operational tables.
1. k-anonymity check (k ≥ 10) 2. l-diversity on sensitive attrs 3. Small-cell suppression (n < 10) 4. Differential privacy noise (ε ≤ 1.0) 5. Methodology board sign-off 6. Versioned snapshot to S3
DynamoDB Single-Table Schema
Table: mcc-main PK SK Attributes -- -- ---------- ROLE#<roleId> META# name, summary, version ROLE#<roleId> METRIC#<metricId> status, tier, name, desc METRIC#<metricId> META# full methodology JSON METRIC#<metricId> ENDORSEMENT#<identityId> endorsedAt, source METRIC#<metricId> ENDORSE_COUNT# total (atomic counter) JURISDICTION#<jurId> METRIC#<metricId>#<period> value, n, ci_low, ci_high AUDIT#<ts>#<eventId> DETAIL# actor, action, payload GSI1 (status index): GSI1PK: STATUS#<status> GSI1SK: METRIC#<metricId> GSI2 (tier index): GSI2PK: TIER#<tier> GSI2SK: METRIC#<metricId> GSI3 (endorsements by role): GSI3PK: ROLE#<roleId> GSI3SK: ENDORSE_COUNT#<descCount>
Route Table
Least-Privilege Roles
lambda-public-read: - dynamodb:Query (GSI1, GSI2) - dynamodb:GetItem (ROLE#, METRIC#) - kms:Decrypt (public CMK) lambda-endorse: - dynamodb:UpdateItem (atomic counter) - dynamodb:PutItem (AUDIT#) - cognito-identity:GetId lambda-research-dua: - s3:GetObject (research-exports/*) - kms:Decrypt (research CMK) - STS: short-lived (max 1h) lambda-admin: - dynamodb:* (audit table only) - cloudtrail:LookupEvents
SSM Parameter Namespace
/mcc/prod/db/table-name /mcc/prod/db/audit-table /mcc/prod/kms/public-cmk-arn /mcc/prod/kms/research-cmk-arn /mcc/prod/auth/cognito-pool-id /mcc/prod/auth/cognito-client-id /mcc/prod/alerts/slack-webhook [SecureString] /mcc/prod/research/dua-webhook [SecureString] Never hardcode. Never commit. Rotation: 90 days for SecureString.
Lambda Sample
// src/handlers/endorse.ts
import { APIGatewayProxyHandler } from 'aws-lambda';
import { DynamoDBClient } from '@aws-sdk/client-dynamodb';
import { DynamoDBDocumentClient, UpdateCommand } from '@aws-sdk/lib-dynamodb';
const ddb = DynamoDBDocumentClient.from(new DynamoDBClient({}));
const TABLE = process.env.TABLE_NAME!;
export const handler: APIGatewayProxyHandler = async (evt) => {
const metricId = evt.pathParameters?.metricId;
const identityId = evt.requestContext.identity.cognitoIdentityId;
if (!metricId || !identityId) return { statusCode: 400, body: 'bad request' };
try {
await ddb.send(new UpdateCommand({
TableName: TABLE,
Key: { PK: `METRIC#${metricId}`, SK: `ENDORSEMENT#${identityId}` },
UpdateExpression: 'SET endorsedAt = :ts, #src = :src',
ExpressionAttributeNames: { '#src': 'source' },
ExpressionAttributeValues: { ':ts': Date.now(), ':src': 'web' },
ConditionExpression: 'attribute_not_exists(endorsedAt)'
}));
await ddb.send(new UpdateCommand({
TableName: TABLE,
Key: { PK: `METRIC#${metricId}`, SK: 'ENDORSE_COUNT#' },
UpdateExpression: 'ADD total :one',
ExpressionAttributeValues: { ':one': 1 }
}));
return { statusCode: 200, body: JSON.stringify({ ok: true }) };
} catch (e: any) {
if (e.name === 'ConditionalCheckFailedException') {
return { statusCode: 409, body: 'already endorsed' };
}
return { statusCode: 500, body: 'internal' };
}
};
Rollout Phases
Read-only
Static catalog, no endorsements. Methodology review window opens.
Endorsement
Anon Cognito identities. Endorsements persist. Audit trail active.
MO Pilot
Three counties onboarded. Tier A jurisdiction dashboards live.
Research
DUA endpoint opens. First academic partnership. Peer-reviewed methodology.
Multistate
Schema adopted in 3+ states. JTA model legislation published.
Governance & Risk
A transparency platform without governance is an attack surface. These are the structures that make the catalog defensible.
Methodology Board
No single jurisdiction, firm, or advocacy organization may hold more than 20% of board seats. Public disclosure of funding sources is required. A firewall separates CoTrackPro commercial interests from published metrics.
Academic
Three researchers in family law, child development, or social policy from accredited institutions. Rotating three-year terms.
Judicial
Two retired family court judges with jurisdictional diversity. No active judges to prevent conflicts.
Bar
Two family law practitioners nominated by state bar family law sections. One must represent low-income clients.
Advocacy
One domestic violence survivor advocate and one fathers' rights representative — intentional balance.
Parent
One lived-experience parent member with no active family court case, rotating annually.
Child Welfare
One licensed child welfare or mental health professional with trauma-informed credentials.
Metric Adoption Workflow
Propose
Board member or public submits metric with draft methodology.
Review
Data Ethics Review assesses re-identification risk and harm potential.
Public Comment
60-day open comment period. All comments logged and published.
Board Vote
Two-thirds majority required. Dissents published with the decision.
Publish
Methodology frozen, versioned, and snapshot to immutable storage.
Risk Register
What could go wrong. What we do about it.
Judicial independence challenge
Courts may challenge judge-level metrics as infringing judicial independence doctrines.
Mitigation: Start with court-level aggregates only. Judge-level metrics require methodology board sign-off and state judicial conference coordination before publication.
DV survivor re-identification
Protective order data published at small-county level could expose protected persons.
Mitigation: VAWA-aware pipeline. Suppress counties with fewer than 25 PO cases/quarter. Differential privacy noise mandatory. Never publish location granularity.
Adversarial use by high-conflict litigants
Metrics weaponized to harass judges, attorneys, or opposing parties.
Mitigation: No individual-name rankings. Rate-limit API. Terms of use prohibit use in pending litigation without disclosure. Dispute process for every data point.
Media misinterpretation
Journalists report numbers without methodology context, creating misleading stories.
Mitigation: Methodology link published with every metric. Confidence intervals mandatory on every visualization. Press kit explains limitations.
State bar complaint on attorney data
Attorney-level performance metrics draw ethics complaints.
Mitigation: Attorney-level data lives only in Tier B with DUA. No public attorney rankings. Peer benchmarks return only to the attorney themselves.
Funding capture
A single funder or interest group influences metric selection.
Mitigation: Governance charter caps any funder at 20%. Public funding disclosure. Two-thirds board supermajority for metric adoption.
Perceived CoTrackPro bias
Catalog seen as favoring one parent type due to platform origin.
Mitigation: Every compliance metric is bilateral — measured in both directions. Neutrality audit annually. Spin out to JTA or nonprofit at scale.
Data drift
State data schemas change, breaking comparability over time.
Mitigation: Versioned schemas. Every published metric carries its schema version. Break notifications sent to subscribers of affected metrics.
Missouri First.
Because Reform Starts Where You Can Reach It.
Missouri has the right ingredients: a centralized Office of State Courts Administrator, three diverse pilot counties within a half-day's drive of each other, a cooperative state bar family law section, and active academic partners at Mizzou, Wash U, and SLU law schools. This is the blueprint for a 36-month pilot that proves the catalog works — then scales.
Pilot County Selection
Three counties chosen for jurisdictional diversity. Each validates a different scale and population profile before the framework generalizes statewide.
ASSUMPTION: Filing volumes below are representative of Missouri OSCA annual statistical reports. Production values should be confirmed against the most recent Missouri Judicial Report before finalizing MOUs with each circuit.
St. Louis County
Greene County
Boone County
Proposed Statute Language
Draft amendments to Missouri Revised Statutes enabling metric reporting authority and de-identified data sharing. Written for introduction through the Judiciary Committee.
ASSUMPTION: Section numbering below reflects current RSMo chapters but should be verified with Missouri Legislative Research before drafting bill text. This is educational/policy framing — not legal drafting — and should be reviewed by Missouri-licensed legislative counsel before introduction.
Family Court Metric Reporting Authority
Parenting Plan Compliance Data · Voluntary Reporting
Research Data Use Agreement Authority
Protective Order Data Confidentiality
OSCA Data-Sharing MOU
Template memorandum of understanding between CoTrackPro (or a designated nonprofit fiscal agent) and the Missouri Office of State Courts Administrator.
Memorandum of Understanding
Purpose and Scope
This MOU establishes the terms under which OSCA will provide de-identified aggregate family court data to Platform Partner for publication in the public Metric Catalog, and under which Platform Partner will provide user-contributed compliance data back to OSCA for judicial administration research. Scope is limited to the 13th, 21st, and 31st Judicial Circuits for a 36-month pilot period.
Data Elements Provided by OSCA
- Quarterly aggregate case filings by case type, disaggregated by month and circuit
- Median and P90 time-to-disposition by case type and circuit
- Clearance rates, rolling quarterly
- Mediation referral and completion counts
- Pro se participation rates by case type
- Continuance frequencies per case
- No judge-level, party-level, or individual-case data is provided under this agreement
Data Elements Provided by Platform Partner
- Aggregate parenting time compliance rates, county-level, opt-in user base only
- Aggregate child contact gap distributions
- Aggregate enforcement-request-to-action timelines (user-reported)
- No individual case, party, or identifier data is provided under this agreement
- Minimum k-anonymity threshold of 10 applied to all published cells
Methodology and Governance
All published metrics shall conform to methodology approved by the Methodology Board as constituted in the Catalog Governance Charter (§ 06). OSCA reserves the right to review and comment on any publication involving OSCA-provided data prior to release, with a 15-business-day review window. Changes to published methodology require 60-day public comment before adoption.
Refresh Cadence
OSCA shall provide quarterly data transfers within 45 days of quarter-end. Platform Partner shall publish aggregate user-contributed metrics on the same quarterly cadence with methodology, confidence intervals, and sample size disclosed for each cell.
Audit Rights
OSCA may request an independent audit of Platform Partner's de-identification pipeline once per calendar year at Platform Partner's expense. Platform Partner may request an audit of OSCA data provenance on the same basis. Audit results shall be published publicly.
Prohibited Uses
- No published data may be used to rank or compare named individual judges, attorneys, or other court personnel
- No re-identification attempts are permitted; violations terminate this MOU immediately
- No commercial resale of raw data; published aggregates remain public domain
- No use in active litigation without disclosure to all parties and the court
Term and Termination
Initial term of 36 months commencing on execution. Either party may terminate for cause with 90 days' notice, or without cause with 180 days' notice. Upon termination, Platform Partner shall cease new publication using OSCA data within 30 days; historical publications remain available with clear period-of-coverage labels. Renewal requires affirmative agreement by both parties.
Indemnification and Liability
Each party is responsible for its own acts and omissions. Platform Partner indemnifies OSCA against claims arising from Platform Partner's de-identification failures. OSCA retains sovereign immunity consistent with Missouri law. Neither party is liable for consequential or indirect damages.
Budget · 36-Month Pilot
Three-year cost structure assuming three-county pilot, two-person core team, modest infrastructure, and honoraria for the Methodology Board.
| Category | Line Item | Year 1 | Year 2 | Year 3 |
|---|---|---|---|---|
| Personnel | Program Director (0.75 FTE) | $112,500 | $115,875 | $119,350 |
| Data / Methodology Engineer (1.0 FTE) | $140,000 | $144,200 | $148,525 | |
| Privacy / Audit Contractor (part-time) | $40,000 | $42,000 | $44,100 | |
| Subtotal · Personnel | $292,500 | $302,075 | $311,975 | |
| Infrastructure | AWS (DynamoDB, Lambda, S3, KMS, CloudWatch) | $18,000 | $22,000 | $28,000 |
| Software licenses & tooling | $8,500 | $8,500 | $9,000 | |
| Third-party security audit (annual) | $12,000 | $12,000 | $13,000 | |
| Subtotal · Infrastructure | $38,500 | $42,500 | $50,000 | |
| Governance | Methodology Board honoraria (11 seats, 4 meetings/yr) | $33,000 | $33,000 | $33,000 |
| Data Ethics Review (external) | $18,000 | $18,000 | $18,000 | |
| Subtotal · Governance | $51,000 | $51,000 | $51,000 | |
| Outreach | Stakeholder convenings (4/yr in-state) | $16,000 | $18,000 | $20,000 |
| Publications, website, translation (ES) | $22,000 | $18,000 | $18,000 | |
| Subtotal · Outreach | $38,000 | $36,000 | $38,000 | |
| Reserve | Contingency (10% of operating) | $42,000 | $43,160 | $45,100 |
| Annual Total | $462,000 | $474,735 | $496,075 | |
| Three-Year Total | $1,432,810 | |||
Blended Capital Stack
No single funder exceeds 20% of program costs per the governance charter. Target mix: state appropriation (~30%), philanthropic family justice funders (~35%), research grants via academic partners (~20%), operating subscription from participating circuits (~15%).
Comparable Programs
Scale comparable to the National Center for State Courts' performance measurement initiatives and early Court Statistics Project pilots. Per-county annual cost roughly aligns with existing OSCA data-warehouse operating budgets at similar scope.
Stakeholder Call Sheet
The people whose buy-in determines whether this pilot launches. In approximate order of first outreach.
OSCA Statistical Research Staff
Missouri Bar Family Law Section
Presiding Judges · Pilot Circuits
Judiciary Committee Members
Missouri Coalition Against Domestic & Sexual Violence
Mizzou Law · Family Violence Clinic
Wash U & SLU Law Schools
Missouri Fathers & Families
Communications Timeline
Sequencing for announcements, academic partnerships, and the first public data release. Designed to build credibility before making claims.
Quiet Stakeholder Alignment
No public announcements. All energy into OSCA conversations, bar section briefings, and presiding judge meetings. Draft methodology published for private stakeholder review only. Success metric: three signed circuit letters of participation.
Methodology Board Convened
Public announcement of the board composition and charter. First board meeting publicly noticed. 60-day public comment window opens on draft methodology. Success metric: all 11 board seats filled, first methodology vote completed.
Static Catalog Launch
The Catalog goes live as a read-only publication. No data yet — just the framework, the methodology, the governance, and the MOU template. Position as a civic infrastructure proposal, not a product. Success metric: endorsements from MOCADSV, state bar section, and at least two sitting legislators.
First MOUs Signed
OSCA and circuit MOUs executed. Data pipelines built. First internal-only dashboards for pilot circuits. Academic partnerships formalized with DUAs. Success metric: first OSCA data transfer received and processed.
First Public Data Release
Tier A jurisdiction dashboards published. Accompanied by methodology papers, press kit, stakeholder testimonials, and interactive comparison with pre-existing public OSCA reports. Coordinated announcements across all three circuits on the same day. Success metric: coverage in St. Louis Post-Dispatch, Kansas City Star, and Missouri Lawyers Media.
First User-Contributed Metrics
Tier C opt-in expands. Parenting time compliance rate becomes the first user-generated metric published at county aggregate. The truth-layer metric nobody else has. Expect national coverage. Success metric: 500+ opted-in participants contributing across the three circuits.
Research Partnership Publications
First peer-reviewed papers from Mizzou and Wash U partnerships. Present at AFCC, National Conference on Courts and Technology, and Missouri Bar Annual Meeting. Begin conversations with Illinois, Kansas, and Iowa court administrators about schema adoption. Success metric: three peer-reviewed publications, two conference keynotes.
Legislative Advancement
With 30 months of operational data and published research, return to the Judiciary Committee. Introduce RSMo amendments through a carrying legislator. Simultaneous introduction in one neighboring state demonstrates the replication model. Success metric: bill introduced in Missouri, MOU signed in one additional state.
Success Criteria
What makes the pilot worth doing, measured at the 36-month mark.
The Floor
Three circuits participating. First Tier A dashboards published. Methodology Board operating. One peer-reviewed paper. MOCADSV endorsement retained. No re-identification incidents. Pilot continues into Year 4 with sustainable funding.
The Midpoint
All floor criteria plus first Tier C metric published with 500+ participants. Bill introduced in Missouri. Second state signs MOU. Three peer-reviewed publications. State bar family law section adopts catalog as reference in member practice resources. First attorney-level benchmarks returning privately to practitioners.
The Ceiling
All target criteria plus RSMo amendments passed. Five-state schema consortium formed. Arnold Ventures or Pew Charitable Trusts grant secured for national expansion. National Center for State Courts adopts elements. JTA publishes model legislation. Catalog becomes the standard reference in family court reform literature.