[ { "id": "manifest_ed_campus_safety_dataset", "source_family": "ed_campus_safety_dataset", "legal_risk_class": "low_official_structured", "bulk_import_eligible": true, "default_review_tier": "imported_public_source", "default_record_lane": "aggregate_safety_stat", "source_urls": [ "https://ope.ed.gov/campussafety/", "https://ope.ed.gov/campussafety/api/dataFiles/file?fileName=Crime2024EXCEL.zip", "https://ope.ed.gov/campussafety/api/dataFiles/file?fileName=Crime2023EXCEL.zip", "https://ope.ed.gov/campussafety/api/dataFiles/file?fileName=Crime2025EXCEL.zip" ], "acquisition_date": "2026-07-06", "importer_command": "npm run import:ed-campus-safety-scoped", "field_map": { "school": "institution identifier and institution name", "date": "reported calendar year", "date_precision": "year", "category": "offense or bias category mapped to project category", "affected_communities": "bias category when supplied by the workbook", "source_locator": "workbook, sheet, row, column, and cell" }, "duplicate_strategy": "Use stable deterministic event IDs and source-locator keys; reject duplicate school, year, source family, source row, category, and affected-community combinations.", "sampling_plan": "Review importer code before each wave, inspect deterministic samples from the first 50 imported rows, and inspect at least 50 additional rows selected across source sheets and affected-community labels.", "known_limits": [ "Official structured source records may describe reported data rather than adjudicated facts.", "Publication at import tier does not imply individual human certification of each row.", "Counts reflect public-source availability and workbook structure, not a prevalence estimate or campus safety score." ], "publishable_fields": [ "school", "year", "state", "category", "affected_communities", "source_locator", "source_url" ], "prohibited_fields": [ "private names", "private contact details", "direct messages", "private screenshots", "unpublished narrative allegations" ], "exclusion_rules": [ "Exclude rows that require private-person credibility judgments.", "Exclude rows that cannot be linked to a public source locator.", "Exclude rows that would require adding sensitive personal information beyond the source file." ] }, { "id": "manifest_ocr_open_investigation", "source_family": "ocr_open_investigation", "legal_risk_class": "medium_official_document", "bulk_import_eligible": true, "default_review_tier": "imported_public_source", "default_record_lane": "civil_rights_case", "source_urls": [ "https://ocrcas.ed.gov/open-investigations", "https://www.ed.gov/laws-and-policy/civil-rights-laws/office-civil-rights-ocr-reading-room" ], "acquisition_date": "2026-07-07", "importer_command": "node scripts/discover-ocr-open-investigations.mjs && node scripts/export-ocr-open-investigation-wave-candidates.mjs", "field_map": { "school": "OCR institution name resolved to a known postsecondary school when possible", "date": "open investigation date", "date_precision": "day", "category": "OCR type of discrimination mapped to a non-finding project category", "affected_communities": "civil-rights class indicated by OCR type of discrimination", "source_locator": "OCR table row with state, institution, institution type, discrimination type, and open date" }, "duplicate_strategy": "Use stable row hashes across state, institution, institution type, discrimination type, and open investigation date; duplicate rows are quarantined deterministically.", "sampling_plan": "Review discovery counts, inspect first and last page samples, inspect at least 50 accepted PSE rows, inspect all quarantine reason counts, and confirm public wording says open investigation rather than finding.", "known_limits": [ "The OCR open-investigations table identifies investigations OCR has opened; it does not mean OCR has made a decision about the case.", "Elementary-secondary institution rows are preserved as excluded source rows and are not published in university accountability waves.", "Imported public-source records from this family are not individual human certification, legal findings, rankings, prevalence estimates, safety scores, or severity scores." ], "publishable_fields": [ "school", "state", "open investigation date", "type of discrimination", "institution type", "source_url", "source_locator" ], "prohibited_fields": [ "private names", "private contact details", "complainant names", "respondent names", "unsupported liability conclusions", "claims that OCR made a finding when the source only lists an open investigation" ], "exclusion_rules": [ "Exclude non-postsecondary institution rows from university accountability waves.", "Quarantine postsecondary rows whose institution cannot be resolved to a known school before publication.", "Quarantine rows whose public text implies a violation, legal finding, ranking, safety score, severity score, or prevalence estimate." ] }, { "id": "manifest_ocr_resolution_document", "source_family": "ocr_resolution_document", "legal_risk_class": "medium_official_document", "bulk_import_eligible": true, "default_review_tier": "imported_public_source", "default_record_lane": "civil_rights_case", "source_urls": [ "https://ocrcas.ed.gov/ocr-search", "https://www.ed.gov/laws-and-policy/civil-rights-laws/office-civil-rights-ocr-reading-room" ], "acquisition_date": "2026-07-07", "importer_command": "node scripts/discover-ocr-resolution-search.mjs && node scripts/export-ocr-resolution-wave-candidates.mjs", "field_map": { "school": "Postsecondary institution named in OCR Recent Resolution Search and resolved to a known school record", "date": "resolved date listed in OCR search", "date_precision": "day", "category": "OCR resolution-document record mapped to a non-finding project category", "affected_communities": "civil-rights domain indicated by the OCR source family when the search row does not expose a narrower issue facet", "source_locator": "OCR reference number, resolved date, institution, state, and public letter/agreement document links" }, "duplicate_strategy": "Use OCR reference number as the primary stable key; duplicate candidate rows with the same school, resolved date, category, and OCR reference are quarantined deterministically.", "sampling_plan": "Review OCR result counts, inspect first and last page samples, inspect document links for at least 50 accepted rows, inspect all mapping-quarantine reason counts, and confirm public wording says resolution document rather than finding or violation.", "known_limits": [ "OCR resolution documents describe agency case resolution materials; they do not by themselves mean Campus Evidence Lab has made a legal finding.", "A resolution agreement may expressly avoid admission of liability, and imported public-source records must not be presented as violations, rankings, prevalence estimates, safety scores, or severity scores.", "Imported public-source records from this family are not individual human certification of each row." ], "publishable_fields": [ "school", "state", "resolved date", "OCR reference number", "document type", "source_url", "source_locator" ], "prohibited_fields": [ "private names", "private contact details", "complainant names", "respondent names", "unsupported liability conclusions", "claims that a resolution document proves a violation when the source does not say that" ], "exclusion_rules": [ "Quarantine postsecondary rows whose institution cannot be resolved to a known school before publication.", "Quarantine rows without at least one public OCR document link.", "Quarantine rows whose public text implies a violation, legal finding, ranking, safety score, severity score, or prevalence estimate." ] }, { "id": "manifest_ocr_or_ed_release", "source_family": "ocr_or_ed_release", "legal_risk_class": "medium_official_document", "bulk_import_eligible": false, "default_review_tier": "imported_public_source", "source_urls": [ "https://www.ed.gov/about/news/press-release/us-department-of-education-probes-cases-of-antisemitism-five-universities", "https://www.ed.gov/about/news/press-release/us-department-of-educations-office-civil-rights-sends-letters-60-universities-under-investigation-antisemitic-discrimination-and-harassment", "https://www.ed.gov/about/news/press-release/ed-hhs-and-gsa-initiate-review-of-anti-semitic-activity-university-of-washington", "https://www.ed.gov/about/news/press-release/us-department-of-education-initiates-title-vi-investigation-haverford-college-allegedly-tolerating-anti-semitic-harassment", "https://www.ed.gov/about/news/press-release/us-department-of-education-notifies-columbia-universitys-accreditor-of-columbias-title-vi-violation", "https://www.ed.gov/about/news/press-release/us-departments-of-education-and-health-and-human-services-notify-harvard-universitys-accreditor-of-harvards-title-vi-violation", "https://www.ed.gov/preview-link/node/5060/e58bc9e0-17b9-483a-845a-0df68624a768" ], "acquisition_date": "2026-07-06", "importer_command": "npm run import:ed-letters", "field_map": { "school": "named institution in official release or listed letter", "date": "release date or source item date", "category": "official OCR, Title VI, or Title IX action mapped to project category", "affected_communities": "community named in the official release", "source_locator": "source item label, release URL, or document section" }, "duplicate_strategy": "Use source URL, named institution, release date, and action type; merge later updates into changelog instead of creating duplicate records.", "sampling_plan": "Review every source template before reuse and manually inspect any new release format before importing more than one record from it.", "known_limits": [ "Official releases can describe allegations, investigations, or agency positions without resolving all underlying facts.", "The source family is not treated as a legal finding by Campus Evidence Lab.", "Bulk import is disabled until each release format has source-item locators and a source-family review gate." ], "publishable_fields": [ "school", "date", "category", "affected_communities", "agency action", "source_url", "source_locator" ], "prohibited_fields": [ "private names not needed for the public record", "private contact details", "unsupported motive or liability conclusions", "claims beyond the source release" ], "exclusion_rules": [ "Exclude releases without a stable public URL or source item label.", "Exclude extracted claims that are not directly supported by the release text.", "Escalate records that require legal interpretation beyond neutral source attribution." ] }, { "id": "manifest_annual_security_report", "source_family": "annual_security_report", "legal_risk_class": "medium_official_document", "bulk_import_eligible": false, "default_review_tier": "imported_public_source", "source_urls": [ "https://clery.arizona.edu/asr", "https://www.montana.edu/clery/reports/asr/2025/index.html", "https://www.qcc.cuny.edu/publicSafety/docs/2025-QCC-ASR.pdf", "https://www.umass.edu/police/media/175/download", "https://www.villanova.edu/university/public-safety/reports-policies/annual-security-fire-safety-report-2025.html", "https://wit.edu/sites/default/files/2025-09/2025%20Annual%20Security%20and%20Fire%20Safety%20Report%20CY%202022-2023-2024%20%281%29.pdf" ], "acquisition_date": "2026-07-06", "importer_command": "Manual source-family import only until annual-security-report table locators are normalized.", "field_map": { "school": "reporting institution", "date": "reported calendar year or report publication date", "date_precision": "year unless a report row supplies a more precise date", "category": "reported Clery or hate-crime category mapped to project category", "affected_communities": "bias category named in report table", "source_locator": "report URL, page, table, row, or document section" }, "duplicate_strategy": "Use school, calendar year, report URL, page or table locator, category, and affected-community label.", "sampling_plan": "Inspect report-specific tables before import and require page or table locators for every imported row.", "known_limits": [ "Annual security reports vary by institution format and table structure.", "Some reports publish aggregate statistics rather than item-level incident narratives.", "Records from this family must not be treated as prevalence comparisons between schools." ], "publishable_fields": [ "school", "year", "category", "affected_communities", "report title", "source_url", "page or table locator" ], "prohibited_fields": [ "private victim names", "private accused-person names", "private contact details", "unpublished incident details" ], "exclusion_rules": [ "Exclude tables that cannot be located in the public report.", "Exclude rows whose school, year, or bias category cannot be mapped without guesswork.", "Escalate reports with unusual table structures before import." ] }, { "id": "manifest_university_statement", "source_family": "university_statement", "legal_risk_class": "medium_institutional_public_statement", "bulk_import_eligible": false, "default_review_tier": "imported_public_source", "source_urls": [ "https://www.brown.edu/news/2024-07-08/resolution-agreement", "https://drexel.edu/president/communications/message/2024/August/Drexel-Reaches-Voluntary-Resolution-Agreement-on-Title-VI-Investigation", "https://news.lafayette.edu/2024/06/20/lafayette-resolves-ocr-review/", "https://www.sjsu.edu/diversity/office/comms/archive/20220617_anti-asian_hate_crime_on_sjsu_campus.php", "https://www.ohio.edu/news/2023/12/reiterating-ohios-commitment-fostering-community-respect", "https://www.towson.edu/news/2016/hatebiasupdate.html" ], "acquisition_date": "2026-07-06", "importer_command": "Manual source-family import only.", "field_map": { "school": "publishing institution", "date": "statement publication date", "category": "institution-described event, response, or policy action", "affected_communities": "community named by the institution", "source_locator": "statement URL and relevant paragraph or section" }, "duplicate_strategy": "Use statement URL, school, publication date, and described event or response; merge follow-up statements into changelog when they concern the same record.", "sampling_plan": "Individually inspect each statement before publication at any tier above imported public source.", "known_limits": [ "University statements can be partial institutional accounts and may deny, qualify, or omit disputed facts.", "The record should attribute claims to the institution unless independently supported by another source.", "Bulk import is disabled for this family because statements vary in scope and risk." ], "publishable_fields": [ "school", "date", "category", "affected_communities", "institutional_response", "source_url", "source_locator" ], "prohibited_fields": [ "private names not necessary to understand the public statement", "private contact details", "unsupported allegations beyond the statement", "legal conclusions not stated in the source" ], "exclusion_rules": [ "Exclude statements that cannot be attributed to an official institutional page.", "Exclude claims that require resolving a disputed factual issue beyond the statement.", "Escalate records naming private individuals before publication." ] }, { "id": "manifest_campus_public_safety_notice", "source_family": "campus_public_safety_notice", "legal_risk_class": "medium_institutional_public_statement", "bulk_import_eligible": false, "default_review_tier": "imported_public_source", "source_urls": [ "https://www.farmingdale.edu/university-police/hate-crime-res.shtml" ], "acquisition_date": "2026-07-06", "importer_command": "Manual source-family import only.", "field_map": { "school": "campus public safety office or institution", "date": "notice publication date or reported event date", "category": "public safety category mapped to project category", "affected_communities": "community named by the notice", "source_locator": "notice URL and section" }, "duplicate_strategy": "Use notice URL, school, date, and described incident; merge updates from the same notice into the existing record changelog.", "sampling_plan": "Inspect each notice for private names, safety-sensitive details, and attribution before publication.", "known_limits": [ "Public safety notices may be preliminary and can contain safety-sensitive context.", "Notices may describe alleged conduct before investigation is complete.", "This family requires manual review before higher-tier publication." ], "publishable_fields": [ "school", "date", "category", "affected_communities", "source_url", "source_locator", "public safety attribution" ], "prohibited_fields": [ "private names", "private contact details", "precise private residence details", "investigative details not needed for the public record" ], "exclusion_rules": [ "Exclude notices that would expose sensitive personal information.", "Exclude claims that are not clearly present in the notice.", "Escalate any notice involving minors, private addresses, or active safety risk." ] }, { "id": "manifest_government_case_or_letter", "source_family": "government_case_or_letter", "legal_risk_class": "medium_official_document", "bulk_import_eligible": false, "default_review_tier": "imported_public_source", "source_urls": [ "https://www.justice.gov/crt/case-summaries", "https://pres.uky.edu/sites/default/files/2025-10/2025-10-01_legal_ocr_finding_signedagreement.pdf" ], "acquisition_date": "2026-07-06", "importer_command": "Manual source-family import only.", "field_map": { "school": "named institution in official case summary or letter", "date": "letter date, case date, or publication date", "category": "official action mapped to project category", "affected_communities": "community named in the official document", "source_locator": "case summary section, PDF page, document section, or paragraph" }, "duplicate_strategy": "Use agency, case or letter identifier, school, document date, and source locator.", "sampling_plan": "Inspect each official document and preserve page or section locators before publication.", "known_limits": [ "Government case summaries and letters may use legal terms that should be attributed to the source, not to Campus Evidence Lab.", "Some documents combine findings, agreements, allegations, and required actions in one file.", "Records in this family should avoid legal interpretation beyond neutral source description." ], "publishable_fields": [ "school", "date", "agency", "category", "affected_communities", "source_url", "source_locator" ], "prohibited_fields": [ "private names not necessary to understand the official document", "private contact details", "legal conclusions not attributed to the source", "claims beyond the official document" ], "exclusion_rules": [ "Exclude documents without a stable public source URL.", "Escalate records where the document names private individuals.", "Escalate records that require legal analysis beyond source attribution." ] }, { "id": "manifest_government_guidance", "source_family": "government_guidance", "legal_risk_class": "medium_official_document", "bulk_import_eligible": false, "default_review_tier": "imported_public_source", "source_urls": [ "https://www.ed.gov/media/document/resolving-hostile-environment-under-title-vi-discrimination-based-race-color-or-national-origin-including-shared-ancestry-or-ethnic-characteristics-2025" ], "acquisition_date": "2026-07-06", "importer_command": "Manual source-family import only.", "field_map": { "school": "named institution when the guidance references one", "date": "guidance publication date", "category": "policy or guidance action mapped to project category", "affected_communities": "community named in the guidance", "source_locator": "guidance URL, page, and section" }, "duplicate_strategy": "Use agency, guidance title, publication date, school when any, and document section.", "sampling_plan": "Inspect the guidance context before creating or updating any event record.", "known_limits": [ "Government guidance may state agency policy rather than an event-specific factual finding.", "Guidance records should not be used as evidence that a school violated law unless the source itself states an official finding and the record attributes that statement.", "This family is not bulk-import eligible." ], "publishable_fields": [ "agency", "date", "guidance title", "category", "affected_communities", "source_url", "source_locator" ], "prohibited_fields": [ "private names", "private contact details", "unsupported school-specific conclusions", "legal advice" ], "exclusion_rules": [ "Exclude guidance that does not support a campus-specific record or policy-development record.", "Escalate any school-specific interpretation before publication.", "Do not import generalized policy language as if it were an incident record." ] } ]