Data · Legal need
Who may need a lawyer, county by county
The people in each county who are most likely to need a lawyer for a specific kind of problem: older residents who plan estates, noncitizens and people who speak limited English, divorced residents, veterans with a disability rating, people with a disability or on SSI, renters stretched by their rent, uninsured residents and commuters on the road. Every number comes from the Census Bureau's American Community Survey for 2020 to 2024, with its margin of error, for every county, every state and the country.
What is in this dataset?
The people in every U.S. county who may need a lawyer, counted from the Census Bureau's American Community Survey 2020 to 2024 5-year estimates: residents aged 65 and over, residents who are not U.S. citizens, foreign-born residents who arrived since 2010, residents who speak English less than very well (and Spanish speakers among them), divorced residents (and, for states, people divorced in the last year), veterans and veterans with a VA disability rating, residents with a disability, households on SSI, renter households and renters paying 30 or 50 percent of income on rent, residents without health insurance, workers who drive to work and median household income. One record per county (3,144, in the 50 states and DC), per state and for the US, each measure with its 90% margin of error and its share of the population it is counted in.
Use it as the denominator of a local market: how many people in a county could need an immigration, family, elder-law, estate-planning, veterans' disability, Social Security disability, landlord-tenant or injury lawyer, and how that compares with the state. Pair it with the law-offices, deserts, fatal-crashes or immigration-courts datasets by county FIPS code. Agents can look up a state's counties, compare states, or search a county by name and measure.
- Records
- 3,196
- Look up by
- 52
- state
- Primary source
- 3,196
The letters are explained in what P, S, D and U mean.
Where does the data come from?
Read from the Census Bureau's ACS 2020-2024 5-year table-based Summary File (one .dat file per table): each measure is one table line or the sum of several lines, with the margin of error of a sum approximated as the root of the summed squares; shares divide by the table's own universe. Every state's counties sum exactly to the state row and the states to the US row; a gate re-checks every total and re-reads sample values from the raw files on each release.
- Period covered
- January 1, 2020 to December 31, 2024
- Places
- The whole US, states and counties
- How often we refresh it
- Once a year
- The source's next release
- December 2026
- Records read
- October 28, 2026
Official sources
- U.S. Census Bureau, American Community Survey, by bulk download. Census Bureau data products are works of the U.S. government: public domain (17 U.S.C. 105)..
- U.S. Census Bureau, American Community Survey (table shells: the line labels), by bulk download. Public domain (17 U.S.C. 105)..
The sources the records cite most
What does a record look like?
-
Texas, American Community Survey 2020-2024 (5-year estimates): 30,188,424 residents; 4,056,252 aged 65 and over; 3,080,983 not U.S. citizens; 1,868,090 foreign-born arrivals since 2010; 3,666,736 who speak English less than very well (3,034,861 of them Spanish speakers); 2,394,590 divorced; 176,848 divorced in the last 12 months; 1,402,357 veterans, 511,053 of them with a VA disability rating; 3,652,642 with a disability; 488,694 households on SSI; 4,113,561 renter households, 1,994,428 paying 30% or more of income on rent; 5,085,788 without health insurance; 11,631,408 workers commuting by car, truck or van; median household income $78,476.
Status P · checked October 28, 2026 · source: U.S. Census Bureau, American Community Survey 2020-2024 5-year estimates, table-based Summary File (tables B01001, B05001, B05005, B08301, B12001, B12503, B16005, B18101, B19013, B19056, B21001, B21100, B25003, B25070, B27010)
-
California, American Community Survey 2020-2024 (5-year estimates): 39,287,377 residents; 6,158,371 aged 65 and over; 4,841,377 not U.S. citizens; 2,463,057 foreign-born arrivals since 2010; 6,499,255 who speak English less than very well (4,107,223 of them Spanish speakers); 2,860,355 divorced; 183,341 divorced in the last 12 months; 1,305,453 veterans, 375,735 of them with a VA disability rating; 4,497,552 with a disability; 799,159 households on SSI; 5,980,523 renter households, 3,138,319 paying 30% or more of income on rent; 2,557,944 without health insurance; 13,817,095 workers commuting by car, truck or van; median household income $99,122.
Status P · checked October 28, 2026 · source: U.S. Census Bureau, American Community Survey 2020-2024 5-year estimates, table-based Summary File (tables B01001, B05001, B05005, B08301, B12001, B12503, B16005, B18101, B19013, B19056, B21001, B21100, B25003, B25070, B27010)
-
Florida, American Community Survey 2020-2024 (5-year estimates): 22,416,077 residents; 4,781,978 aged 65 and over; 2,088,458 not U.S. citizens; 1,795,552 foreign-born arrivals since 2010; 2,663,509 who speak English less than very well (2,076,728 of them Spanish speakers); 2,355,322 divorced; 135,105 divorced in the last 12 months; 1,327,071 veterans, 412,681 of them with a VA disability rating; 3,016,072 with a disability; 429,728 households on SSI; 2,834,163 renter households, 1,594,400 paying 30% or more of income on rent; 2,548,782 without health insurance; 8,214,861 workers commuting by car, truck or van; median household income $74,568.
Status P · checked October 28, 2026 · source: U.S. Census Bureau, American Community Survey 2020-2024 5-year estimates, table-based Summary File (tables B01001, B05001, B05005, B08301, B12001, B12503, B16005, B18101, B19013, B19056, B21001, B21100, B25003, B25070, B27010)
-
New York, American Community Survey 2020-2024 (5-year estimates): 19,852,366 residents; 3,554,028 aged 65 and over; 1,832,210 not U.S. citizens; 1,325,284 foreign-born arrivals since 2010; 2,550,611 who speak English less than very well (1,235,249 of them Spanish speakers); 1,480,795 divorced; 90,556 divorced in the last 12 months; 581,776 veterans, 123,504 of them with a VA disability rating; 2,482,006 with a disability; 465,828 households on SSI; 3,525,661 renter households, 1,733,205 paying 30% or more of income on rent; 993,274 without health insurance; 5,128,551 workers commuting by car, truck or van; median household income $85,974.
Status P · checked October 28, 2026 · source: U.S. Census Bureau, American Community Survey 2020-2024 5-year estimates, table-based Summary File (tables B01001, B05001, B05005, B08301, B12001, B12503, B16005, B18101, B19013, B19056, B21001, B21100, B25003, B25070, B27010)
The full records add the dates, the quoted source text and every field in the CSV and JSON files below.
What each of the 73 main columns means
jurisdiction- Two-letter state code (50 states and DC), or US for the national row
county- 5-digit county FIPS code (2025 county list, as in the crosswalks dataset); empty on state and US rows
level- county, state or us
name- County and state, state name, or United States
value- Total residents (ACS 2020-2024 5-year estimate, table B01001)
period- The survey years the 5-year estimate pools (2020-2024)
pop_total- Total residents (B01001 line 1)
pop_total_moe90- 90% margin of error of pop_total (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
age_65plus- Residents aged 65 and over (B01001 lines 20-25 and 44-49); share of total residents
age_65plus_moe90- 90% margin of error of age_65plus (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
age_65plus_pct- age_65plus as a percentage of its universe (pop_total), one decimal
noncitizens- Residents who are not U.S. citizens (B05001 line 6); share of total residents
noncitizens_moe90- 90% margin of error of noncitizens (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
noncitizens_pct- noncitizens as a percentage of its universe (pop_total), one decimal
recent_arrivals- Foreign-born residents who entered the U.S. in 2010 or later (B05005 line 4); share of residents born outside the U.S.
recent_arrivals_moe90- 90% margin of error of recent_arrivals (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
recent_arrivals_pct- recent_arrivals as a percentage of its universe (born_abroad), one decimal
lep- Residents aged 5+ who speak English less than "very well", any language (B16005, the well / not well / not at all lines); share of residents aged 5+
lep_moe90- 90% margin of error of lep (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
lep_pct- lep as a percentage of its universe (pop_5plus), one decimal
spanish_lep- Spanish speakers aged 5+ who speak English less than "very well" (B16005 lines 6-8 and 28-30); share of residents aged 5+
spanish_lep_moe90- 90% margin of error of spanish_lep (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
spanish_lep_pct- spanish_lep as a percentage of its universe (pop_5plus), one decimal
divorced- Residents aged 15+ whose current marital status is divorced (B12001 lines 10 and 19); share of residents aged 15+
divorced_moe90- 90% margin of error of divorced (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
divorced_pct- divorced as a percentage of its universe (pop_15plus), one decimal
divorced_last_year- Residents aged 15+ divorced in the last 12 months (B12503 lines 5 and 10); the Census Bureau publishes this table for states and the US only, so county rows are empty
divorced_last_year_moe90- 90% margin of error of divorced_last_year (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
divorced_last_year_pct- divorced_last_year as a percentage of its universe (pop_15plus), one decimal
veterans- Civilian veterans aged 18+ (B21001 line 2); share of civilians aged 18+
veterans_moe90- 90% margin of error of veterans (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
veterans_pct- veterans as a percentage of its universe (civ_pop_18plus), one decimal
va_rated- Civilian veterans with a VA service-connected disability rating, any percentage (B21100 line 3); share of civilian veterans
va_rated_moe90- 90% margin of error of va_rated (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
va_rated_pct- va_rated as a percentage of its universe (civ_veterans), one decimal
va_rated_70plus- Civilian veterans rated 70 percent or higher (B21100 line 8); share of civilian veterans
va_rated_70plus_moe90- 90% margin of error of va_rated_70plus (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
va_rated_70plus_pct- va_rated_70plus as a percentage of its universe (civ_veterans), one decimal
disability- Civilian noninstitutionalized residents with a disability (B18101, the 12 'with a disability' lines); share of that population
disability_moe90- 90% margin of error of disability (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
disability_pct- disability as a percentage of its universe (civ_noninst_pop), one decimal
ssi_households- Households with Supplemental Security Income (B19056 line 2); share of households
ssi_households_moe90- 90% margin of error of ssi_households (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
ssi_households_pct- ssi_households as a percentage of its universe (households), one decimal
renter_households- Renter-occupied housing units (B25003 line 3); share of occupied housing units
renter_households_moe90- 90% margin of error of renter_households (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
renter_households_pct- renter_households as a percentage of its universe (occupied_units), one decimal
rent_burdened- Renter households paying 30 percent or more of household income on gross rent (B25070 lines 7-10); share of renter households whose rent share is computed
rent_burdened_moe90- 90% margin of error of rent_burdened (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
rent_burdened_pct- rent_burdened as a percentage of its universe (renter_units), one decimal
rent_severe- Renter households paying 50 percent or more (B25070 line 10); same universe
rent_severe_moe90- 90% margin of error of rent_severe (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
rent_severe_pct- rent_severe as a percentage of its universe (renter_units), one decimal
uninsured- Civilian noninstitutionalized residents with no health insurance coverage (B27010 lines 17, 33, 50 and 66); share of that population (B27010 line 1)
uninsured_moe90- 90% margin of error of uninsured (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
uninsured_pct- uninsured as a percentage of its universe (B27010 line 1), one decimal
median_household_income- Median household income in the past 12 months, in 2024 inflation-adjusted dollars (B19013 line 1); not a count, so not summed
median_household_income_moe90- 90% margin of error of median_household_income (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
commute_vehicle- Workers aged 16+ who commute by car, truck or van, driving alone or carpooling (B08301 line 2); share of workers aged 16+
commute_vehicle_moe90- 90% margin of error of commute_vehicle (a sum of lines: root of the summed squares, the Census Bureau's approximation); 0 where the estimate is controlled
commute_vehicle_pct- commute_vehicle as a percentage of its universe (workers_16plus), one decimal
born_abroad- Universe: residents born outside the United States (B05005 line 1)
pop_5plus- Universe: residents aged 5 and over (B16005 line 1)
pop_15plus- Universe: residents aged 15 and over (B12001 line 1)
civ_pop_18plus- Universe: civilians aged 18 and over (B21001 line 1)
civ_veterans- Universe: civilian veterans aged 18 and over (B21100 line 1)
civ_noninst_pop- Universe: civilian noninstitutionalized residents (B18101 line 1)
households- Universe: households (B19056 line 1)
occupied_units- Universe: occupied housing units (B25003 line 1)
renter_units- Universe: renter-occupied units whose rent share of income is computed (B25070 line 1 minus line 11)
workers_16plus- Universe: workers aged 16 and over (B08301 line 1)
noisy- Measures whose 90% margin of error is more than half the estimate (or whose estimate is 0 with a margin above 0): read them as rough
suppressedMeasures- Measures the Census Bureau did not publish for this place (sample too small); their value is empty
How do I use it in my own tools?
legalneed_lookup- For one two-letter state code (or US): the state row and every county in it, with residents aged 65+, noncitizens, recent arrivals, residents who speak English less than very well, divorced residents, veterans and VA disability ratings, disability, SSI households, renters and rent burden, uninsured residents, drivers to work and median household income, each with its margin of error and share.
legalneed_compare- The same records for up to 10 two-letter state codes, to set one state's counties against another's.
legalneed_search- Records whose text matches the words given, such as a county and a measure ("Harris County divorced"), a county FIPS code ("48201") or a state name.
How to cite this dataset
You can reuse, adapt and republish this data, including commercially, under the Creative Commons Attribution 4.0 license (CC BY 4.0). The only condition is that you credit Probative and link to the source.
- Version
- 1.0.0, released October 28, 2026
- Every version
- 10.5281/zenodo.23741885 (always points to the newest)
- License
- CC BY 4.0
Download the data
- CSV: legal-need.csv 4627 KB · one row per record (columns in README.txt)
- JSON: legal-need.json 10697 KB · every record with its sources
Free, no email or login. The files are hosted on this site and Zenodo, a public research archive.
Suggested citation
Probative. (2026). Legal need by county: who may need a lawyer, from the American Community Survey (Version 1.0.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.23741886
Credit line for a chart, table or article
Source: Probative, Legal need by county: who may need a lawyer, from the American Community Survey, version 1.0.0, https://doi.org/10.5281/zenodo.23741886, licensed CC BY 4.0.
Add to a reference manager (Zotero, Mendeley, EndNote)
Not legal advice. These are survey estimates of people with a characteristic, not counts of legal problems or cases: a renter, a veteran or a noncitizen does not necessarily need a lawyer. Each estimate has a margin of error, and small counties' estimates are rough.
What changed recently?
- 2026-10-28 · 3,144 counties, 51 states and DC, and the US from the ACS 2020-2024 5-year estimates: 334,922,499 residents, 22,674,672 not U.S. citizens, 27,139,643 who speak English less than very well, 16,185,883 veterans and 45,017,354 renter households. (source)
The full log is in changes.json.
What are the limits?
- The American Community Survey is a survey, not a full count. Each number is an estimate pooled over five years of interviews and comes with a ninety percent margin of error. A measure whose margin is more than half the estimate is listed under noisy in its record; read it as rough.
- A person counted here is not a case. A renter, a veteran or a noncitizen may never need a lawyer; the measures size a market, they do not count legal problems, filings or clients.
- Ability to speak English is self-reported by the household, and limited English here means speaking English less than very well, the Census Bureau's own line.
- Small counties are noisy. In counties with a few thousand residents the margins are wide, and the Census Bureau withholds a value when the sample is too small; a withheld value is left empty and marked suppressed, never estimated.
- Divorces in the last twelve months are published by the Census Bureau for states and the country only, so county rows carry the number of divorced residents instead. Puerto Rico is not included.
- The estimates describe the years 2020 to 2024 as a whole, not one year. Places that grew fast since then, such as some suburban counties in the Sun Belt, have more residents today than these numbers show.
- Records older than 400 days since their last check are marked
stalein the files until we re-check them. Every confirmed error we fix is listed in the data corrections log.
Found a wrong value? Report an error with the record id and the source that shows the right value.
Running a firm? You don’t need the raw data: we handle your state’s ad rules for you. See where your firm stands first.
Free
Get your free review.
Send us your firm’s website. Probative Growth records a 90-second video of what clients see when they look you up, and emails it with a one-page summary within 2 business days. No logins, no call needed.
Not sure where to start? Answer 7 questions Ready to talk? Talk to us Or see what AI says about your firm
Your free review: It covers your Google listing, website, phone and ads, from public information.