DEADLINING™ Public forensic database · Public access · Updated 2026-11-19 Donate
A research project by Avery Bernard Johns

176 cases.
168 still open.

Black women and girls killed, missing, or found unidentified in Washington, D.C., 1961–2025. Mapped, measured, searchable.

Press / to search · Try “Suitland”, “MD-DC-00002”, or “Still Missing”
Filter
Figure 1. Case density and Black population share by census tract, Washington, D.C.
176cases · 1961–2025
Source: DEADLINING™; U.S. Census 2020 (tract population). Contours are a kernel density estimate of case locations, drawn as terrain. Drag the timeline to filter.

Case map

176 cases with coordinates · satellite basemap · select a pin for the case record

Pins reflect the current search and filters. Imagery © Mapbox, © Maxar. Locations are shown as recorded; 108 records are still under location review.

Mission

DEADLINING™ keeps a sourced and citable record of Black women and girls killed, missing, or found unidentified in Washington, D.C., and studies where and under what conditions these cases occur.

  1. DocumentEvery case we can verify, with its sources, location, and data-quality status.
  2. MeasureTest spatial patterns against city-wide baselines rather than assume them.
  3. TransparentPublish methods and corrections so the work can be checked and cited.
  4. ServeGive families, reporters, and investigators records organized by place and time.

Key figures

Research programAll results

About the databaseMethodology

DEADLINING™ documents 176 cases: 152 homicides, 14 missing-person cases, and 10 sets of unidentified remains. Of the 151 victims with a recorded age, 25 were under 18. 168 cases remain open.

Each record is geocoded, joined to its 2020 census tract, and measured against the city's interstate and parkway network. Cases sit closer to highways than the city's tracts do: 71.7% lie within 2 km, against 58.7% of tracts. In the published tract-level model, each additional kilometre from a highway is associated with about 15% fewer cases (IRR 0.85, 95% CI 0.74–0.97; Mann–Whitney p = 0.014). These two model results come from the analysis files and will update when the analysis is rerun on the current master file.

Results are preliminary. 65 records have verified locations, 108 are under review, and 3 have coordinates outside the District.

Records
176
Mapped
183
Flagged
Period
1961–2025, by date last seen
Spatial unit
2020 census tract (n = 206)
Coordinates
WGS 84 · EPSG:4326
Case registry

Case-level records

Search, filter, and sort every record. Filters here also apply to the spatial analysis. Each record lists its sources, spatial measures, data-quality fields, and a citation.

SelectionAll records
Sort
Spatial analysis

Spatial distribution of cases relative to the highway network

Each case is measured to the nearest interstate or parkway centerline. Census tracts are shaded by Gentrification Risk Index (GRI) tier. Selecting a study site filters the map, the measurements, and the case registry.

Basemap
Loading map…
Cursor —
PointsSolid points are address-level locations. Hollow points are approximate or have no recorded precision.
Dotted linesThe measured straight-line distance from each case to the nearest highway centerline.
ContoursKernel density estimate of case locations in the current selection. Inner rings are denser.
Results

Case locations concentrate near highway corridors

Four results answer the study's research questions (RQ1–RQ4). Each figure is drawn from a published output file named beneath it, so every value can be verified and reproduced.

Data & documentation

Datasets, variables, and access

Every file in the project, with row and column counts measured directly from the files.

All data are © Avery Bernard Johns, all rights reserved. Contact the researcher for permission to use this data.

Data catalog

Programmatic access Planned

# Query cases by ward, period, and corridor GET /api/v1/cases?ward=8&from=2000&corridor=suitland # One record, with sources and quality fields GET /api/v1/cases/MD-DC-00002 # Fixed release for citation GET /api/v1/releases/v1.0/cases.csv

Data dictionary Draft

FieldTypeDefinition
case_idstringPermanent identifier. Never reused or renumbered.
case_typeenumHomicide, Missing, or Unidentified.
outcomeenumRecovered Deceased, Still Missing, or Unidentified.
date_last_seendateISO 8601. Year only when the full date is not documented.
latitude, longitudefloatWGS 84 coordinates of the location of record.
geocode_precisionenumAddress, Address_Level, Place, or Missing. Sets how much weight a location can carry in analysis.
review_needed0 / 11 when a field is disputed or unverified.
source_linkurlPrimary source for the record.
dist_mintegerDerived. Metres to the nearest interstate or parkway centerline.
Methodology

Research design and analytic pipeline

The pipeline runs in order. Each stage writes a file, so any number on the site can be traced back to the stage that produced it.

S1

Source collection

MPD releases, NamUs, Homicide Watch, court and news records.

source_link
S2

Entry and deduplication

One row per person. Duplicates grouped and consolidated.

01_cases_master_cleaned
S3

Geocoding

Location of record converted to coordinates and coded for precision.

latitude, longitude, geocode_precision
S4

Tract join

Cases joined to 2020 tracts, census counts, and GRI.

02_cases_master_GRI_joined
S5

Spatial measures

Distance to corridor, buffers, corridor density.

04_processed_data/spatial_analysis
S6

Models

Poisson and negative binomial models of tract case counts.

04_processed_data/regression_models
S7

Publication

Web exports, figures, and versioned release.

05_web_exports

Definitions

Distance to corridor
Straight-line metres from a case location to the nearest point on any D.C. interstate or parkway centerline.
Buffer share
Percent of case locations, and of tract centroids, within 100, 250, 500, 1,000, and 2,000 m of a corridor.
Corridor density
Cases assigned to a corridor divided by that corridor's length in kilometres.
Standardized ratio (SMR)
Observed cases in a quadrant divided by cases expected from its share of the city's Black population (2010).
Gentrification Risk Index (GRI)
Tract composite of Black population displacement and labor-force change, 2010–2020, grouped into five tiers.

Known limitations

  • Documented sample, not a census. The database holds cases that reached a public record. Undercounting is likely and uneven.
  • Location precision varies. Most records have no recorded geocode precision. Results should be checked on address-level cases alone.
  • Population baselines are from 2010. Quadrant rates use 2010 counts across a 1961–2025 case span.
  • Corridor assignment is geometric. Nearest centerline only. It does not say a highway was involved in a case.
  • Associations, not causes. Models describe where cases cluster. They do not explain why.
Citation & access

Citation, corrections, and support

Cite the dataset

Contact the researcher for permission to use this data.

Johns, A. B. (2026). DEADLINING: Washington D.C. Database for Missing and Murdered Black Women and Girls (MMBDB), 1961–2025 [Data set]. deadlining.org

Corrections and sources

Families, researchers, and reporters can submit a correction or a new source. Every change is logged in the record's history.

Information about a case

DEADLINING™ does not collect tips. Contact the Metropolitan Police Department.

Call(202) 727-9099 Text50411

Support the research

Funds hosting, public-records requests, and research assistants. Donations open once a fiscal sponsor is in place.

About the researcher

Research project by Avery Bernard Johns

DEADLINING™ and the Washington D.C. Database for Missing and Murdered Black Women and Girls (MMBDB) are an independent research project by Avery Bernard Johns, a student at Howard University and a student member of the American Academy of Forensic Sciences (AAFS).

Email
deadlining.org@gmail.com
Instagram
@mmbwdata
Data use
Contact the researcher for permission to use this data.
Affiliations
Howard University (student) · American Academy of Forensic Sciences (student member)