Transparency about how we collect, process, and analyze New Jersey court data to ensure accuracy and reproducibility.
71,577
Court Cases
6,951
GRC Complaints
886
OPRA Cases
84,900
Entities Tracked
Civil case records from the New Jersey Judiciary's eCourts system. Data includes docket numbers, case titles, parties, filing dates, disposition information, and venue assignments.
Denial of Access Complaints filed with the NJ GRC under OPRA. Data includes complaint numbers, complainants, custodians, filing dates, and outcomes.
Raw case data is collected from public court records and GRC complaint databases. Records are validated for completeness and standardized to a common schema.
Party names are parsed and normalized to create unique entity records. Government agencies are identified using pattern matching against known entity types (Township, City, County, Police Department, Board of Education, etc.).
Cases are classified as OPRA-related using text analysis of case titles and party names. Keywords include "OPRA," "Open Public Records," "Denial of Access," and related terms. Classification is validated against known OPRA case lists.
Entities are deduplicated across court cases and GRC complaints using fuzzy name matching to identify the same agency or individual appearing under different name variations.
Aggregate statistics are computed including case counts by venue, year, and case type; average days to disposition; top litigants; and trend analysis.
Pattern-based classification identifies government entities using regex patterns for:
Entities are deduplicated using probabilistic matching:
The NJ Judiciary's ACMS system exports data in a fixed-width format (PAB0231), which truncates fields at specific character positions. This means:
We display names exactly as provided in the source data. Where possible, we maintain a mapping of truncated names to complete names, but coverage is not exhaustive.
Complainant and public entity names in GRC complaints are extracted from HTML decision pages. Variations occur due to:
Party names in Office of Administrative Law cases may be intentionally abbreviated for privacy, particularly in education, health, and child welfare matters. Names like "J.S. o/b/o M.S." are common in these jurisdictions.
For detailed information about data quality issues and how we handle them:
View Data Quality GuideSome court records may be sealed, confidential, or not yet digitized. Our database represents publicly available records and may not include all cases filed in NJ courts.
OPRA case classification is based on text analysis and may include false positives or miss cases that don't use standard OPRA terminology in their titles.
Entity deduplication is probabilistic. Some entities may appear as duplicates due to name variations, and some distinct entities may be incorrectly merged if they have similar names.
This database is updated periodically and may not reflect the most recent court filings or GRC decisions. Check official sources for real-time information.
99.2%
Entity Extraction Accuracy
94.7%
OPRA Classification Precision
96.8%
Entity Match Accuracy
If you notice errors in the data, have questions about our methodology, or would like to contribute to this project, please contact NJFOG.