Baltimore, MD · 2022 to 2024
In Baltimore, Black women's reported assault rate is about 2.5 to 3 times White women's
Age, police district and neighborhood conditions don't explain it. The comparison with Hispanic women can't be settled, because ethnicity is missing for 42% of women victims.
From 2022 through 2024, the Baltimore Police Department recorded 47,588 victims of common and aggravated assault, including 1,739 people wounded in non-fatal shootings. Per 100,000 residents a year, Black women's reported assault rate was 3,605, against 1,397 for White women: 2.6 times as high.
Reported rates are high for every group in Baltimore, and the gap between groups is narrower than in DC or Los Angeles. Black women's rate is still higher than the rate for men in any group, including Black men (3,218).
Hispanic women's rate is uncertain because ethnicity is unknown for 41.6% of women victims (see below). Source: Baltimore Police Department records; ACS 2024 5-year population.
The usual explanations don't shrink it
The analysis compared women of the same age, in the same year, in the same police district and in census tracts with similar poverty, income, unemployment and housing. None of these narrowed the gap. Against all other women, it was 2.6 times before any controls and 2.8 times after all of them.
Neighborhood conditions are census-tract poverty, income, unemployment and housing. Source: tract-level model in the full analysis.
Place doesn't account for it either. If Black women had other women's rate in each police district, their rate would be 1,664 per 100,000, not 3,605. Fully adjusted, Black women's rate is 3.1 times White women's.
Neighborhood conditions are shaped by segregation and disinvestment, so they are not neutral controls. Here they don't narrow the gap at all.
The Hispanic comparison can't be settled
Baltimore records ethnicity apart from race, and it is unknown for 41.6% of women victims. As recorded, Black women's rate is 1.6 times Hispanic women's. But 41.9% of women recorded as White have unknown ethnicity, and where it is known, 18.1% of them are Hispanic. If all the women with unknown ethnicity who were recorded as White were in fact Hispanic, Black women's rate would be 0.8 times Hispanic women's and 4.9 times White women's. The data can't tell which is closer.
The White comparison holds either way, at between 2.5 and 4.9 times.
A second records system
Baltimore police moved to the FBI's NIBRS format in January 2025. In the new system's first 20 months, through August 2026, Black women's rate was 2.9 times White women's, against 2.6 times in 2022 to 2024.
What this does not show
- Why. The data shows a difference in reported rates. It does not measure causes, offenders or circumstances.
- Every assault. Only assaults that reached the police are counted. What the assault-victim gap measures compares police records with the national victimization survey and hospital records, including a test of how Baltimore police record 911 calls. Police data also reflects where officers patrol and who calls them, so it can't separate more policing or more reporting from more assaults.
- Exact group membership. Victim race is recorded by officers; the Census counts residents by their own answers. If multiracial victims were recorded as Black, the ratio to White women would be 2.5 times instead of 2.6.
- Partner assault and assaults on officers. No field marks assaults by a partner, so they can't be separated. Assaults on police officers are counted with everyone else's.
- People. Rates count reports, not people: someone assaulted twice counts twice. Rates divide by residents, not by where people spend their time.
Part of the series Assault victims by race and sex, 6 reports.