Project Name:
Modeling the Criminal Case Process with State and Local Data
Contractor: RTI International
Lessons Learned
Technical Feasibility Is Not the Primary Constraint
Discussions with BJS and NSF confirmed a shared understanding that PPRL techniques are technically mature and reliable. The central question for this project is not whether PPRL works algorithmically, but whether criminal justice agencies have the operational capacity, infrastructure, and institutional readiness to implement these methods.
Impact on Project
This insight shifted the project’s emphasis away from validating technical performance in ideal conditions toward understanding implementation feasibility in realistic agency environments.
Administrative and Governance Processes Shape Timelines and Outcomes
Negotiating data-use agreements, legal review, IRB approval, and internal agency sign-off were identified as major drivers of timelines. These processes vary widely across jurisdictions and can delay or prevent execution even when technical capacity exists.
Impact on Project
Administrative and governance processes are being systematically documented as analytic findings.
Data Readiness and Identifier Quality
Preliminary assessment of candidate datasets suggests substantial heterogeneity in identifier availability, formatting, consistency, and documentation across criminal justice agencies. Variability in identifier persistence over time, cross-system alignment, and governance practices potentially introduces uncertainty into linkage planning that cannot be resolved through privacy-preserving methods alone. These observations reinforce the importance of early data readiness assessments to determine whether PPRL is appropriate, and to identify cases where improvements in data management or documentation would be required prior to any linkage effort.
Impact on Project
Early data readiness assessment has been incorporated into engagement and site vetting activities.
Constraints and Non-Execution Are Informative Outcomes
As the project advances into agency engagement, it is increasingly clear that instances in which data acquisition or linkage cannot proceed—due to administrative delays, governance constraints, limited infrastructure, or data limitations—are themselves informative implementation outcomes. Documenting these non-execution pathways provides critical insight into the conditions under which privacy-preserving record linkage is unlikely to be feasible, directly supporting the government’s interest in understanding practical adoption constraints rather than technical performance alone.
Impact on Project
The project now treats stalled or incomplete engagements as analytic cases to be documented. This insight supports the government’s interest in understanding the practical conditions under which privacy-preserving technologies add value, rather than assuming uniform applicability across criminal justice systems.
Clear Communication Enables Trust and Engagement
As the project moves into active agency engagement, it is clear that structured outreach materials—such as “What to Expect” briefs, readiness assessments, and clear descriptions of data protections—will be essential for establishing trust and facilitating productive conversations with potential partners. Agencies vary widely in their familiarity with privacy-preserving technologies, and early, transparent communication about project goals, expectations, and burdens is likely to shape both willingness to participate and the efficiency of subsequent engagement.
Impact on Project
Development of engagement materials has been prioritized as core project infrastructure. Potential documents needed to assist criminal justice agencies to implement PPRL include:
- Agency Engagement Roadmap for Privacy-Preserving Record Linkage
A step-by-step roadmap describing how criminal justice agencies can progress from initial interest to potential implementation of PPRL. The roadmap will outline key phases (e.g., exploratory discussions, data readiness assessment, governance review, technical preparation), typical decision points, and common challenges observed during engagement.
- “What to Expect” Guide for Participating Agencies
A plain-language document describing project goals, roles and responsibilities, data protections, anticipated staff time commitments, and the types of information agencies may be asked to provide. This document is intended to support informed decision-making by agencies considering participation.
- Agency Readiness and Capacity Assessment Tool
A structured assessment instrument to help agencies and researchers evaluate readiness for privacy-preserving linkage. Topics will include data systems, identifier quality, staffing capacity, governance structures, and secure computing environments.
- Data and Identifier Inventory Template
A standardized worksheet for documenting available datasets, identifier fields, data quality issues, and documentation gaps. This tool will support early feasibility assessment and help determine whether PPRL is appropriate for a given agency context.
- Governance and Approval Pathway Checklist
A checklist identifying common legal, administrative, and institutional approvals required for data sharing and linkage projects. The checklist will document typical sequencing, dependencies, and anticipated timelines.
- Implementation Barriers and Lessons Log
A structured log for documenting challenges encountered during engagement, including delays, constraints, and non-execution pathways. This document will serve as a key input into lessons learned and roadmap development.
Disclaimer: America’s DataHub Consortium (ADC), a public-private partnership, implements research opportunities that support the strategic objectives of the National Center for Science and Engineering Statistics (NCSES) within the U.S. National Science Foundation (NSF). These results document research funded through ADC and is being shared to inform interested parties of ongoing activities and to encourage further discussion. Any opinions, findings, conclusions, or recommendations expressed above do not necessarily reflect the views of NCSES or NSF. Please send questions to [email protected].




