Community Resilience Estimates (Census CRE)

Estimated share of residents with 0, 1-2 or 3+ social-vulnerability components, by county (2016-2024) and tract (2020-2024).

Fuente: U.S. Census Bureau, Community Resilience Estimates, editions 2016-2024 (county and tract files)
Cobertura de años: 2016–2024
Actualización: annual
Versión: 1.0.0
Hash del esquema: 657c579e7ddc

Niveles de detalle disponibles

Condado · predeterminado

Resumen

The U.S. Census Bureau's Community Resilience Estimates (CRE) for Georgia: the estimated number and share of residents with zero, one or two, or three or more of ten components of social vulnerability - household and individual risk factors such as low income, disability, no health insurance, age 65 and over and no broadband - that make it harder to absorb and recover from a disaster or other shock. By county for every edition 2016-2024 and by census tract for 2020-2024. Model-based estimates (American Community Survey 1-year responses combined with population estimates in a small-area model) with 90 percent margins of error; the headline is the share with three or more components.

Granularidad de fila: Una fila por year, county_fips, district_code, tract_geoid.

Métrica clave

pct_three_plus_components — Share With 3+ Vulnerability Components (proportion)

El valor principal que la mayoría de las consultas busca.

Esquema JSON →Contrato de datos (YAML) →

Variables

Todas las columnas de la tabla de hechos. El código es el nombre exacto del campo que devuelve la API.

Columna	Tipo	Unidad	Rango	¿Nulo?	Descripción
`year`Year	integer	—	—	no	Edition year: the year the estimates describe, modeled from that year's American Community Survey 1-year responses and population estimates (the 2024 edition, released January 2026, uses 2024 ACS data). One edition per year, 2016-2024; each is a separately modeled release, not a revision of the others. Tract rows exist for 2020-2024 only. Descripción completa▾ Edition year: the year the estimates describe, modeled from that year's American Community Survey 1-year responses and population estimates (the 2024 edition, released January 2026, uses 2024 ACS data). One edition per year, 2016-2024; each is a separately modeled release, not a revision of the others. Tract rows exist for 2020-2024 only.
`county_fips`County FIPS	string	—	—	yes	5-digit county FIPS code (state 13 + county), from the source GEO_ID. Populated on county rows (every year); NULL on tract rows. Descripción completa▾ 5-digit county FIPS code (state 13 + county), from the source GEO_ID. Populated on county rows (every year); NULL on tract rows. Se une a counties; agrega sus columnas de etiqueta.
`district_code`District Code	string	—	—	yes	Always NULL: the Community Resilience Estimates publish no school-district geography. Kept for the census domain's standard key shape. Descripción completa▾ Always NULL: the Community Resilience Estimates publish no school-district geography. Kept for the census domain's standard key shape. Se une a districts; agrega sus columnas de etiqueta.
`tract_geoid`Tract GEOID	string	—	—	yes	11-digit 2020 census tract GEOID (FK to the tracts dimension), from the source GEO_ID. Populated on tract rows, which exist for 2020-2024 only (the 2016-2019 editions use 2010 tract boundaries and are served at county level only); NULL on county rows. Descripción completa▾ 11-digit 2020 census tract GEOID (FK to the tracts dimension), from the source GEO_ID. Populated on tract rows, which exist for 2020-2024 only (the 2016-2019 editions use 2010 tract boundaries and are served at county level only); NULL on county rows. Se une a tracts; agrega sus columnas de etiqueta.
`pct_three_plus_components`Share With 3+ Vulnerability Components	number	proportion	—	yes	Estimated share of residents with three or more social-vulnerability components. Descripción completa▾ Estimated share of the universe population (num_population) with three or more of the ten CRE components of social vulnerability (0-1). A model-based estimate, served as published (PRED3_PE / 100; the source rounds to two decimals of a percent) and equal to num_three_plus_components / num_population within that rounding. NULL only where num_population is 0 (the published 0, or the 2021 edition's modeled rate, over no residents is set to NULL as known-bad). The headline measure: the Census Bureau's 'most socially vulnerable' group, least able to absorb and recover from a disaster or other shock.
`pct_three_plus_components_moe`Percent Three Plus Components Moe	number	proportion	—	yes	90 percent margin of error for pct_three_plus_components (0-1 scale), as published (PRED3_PM / 100). Equals num_three_plus_components_moe / num_population within rounding (2019 county rows: within 0.03 points). County margins for 2019, 2021 and 2022 come from an older Census variance method and run larger than in other editions; do not compare margins across those editions. Descripción completa▾ 90 percent margin of error for pct_three_plus_components (0-1 scale), as published (PRED3_PM / 100). Equals num_three_plus_components_moe / num_population within rounding (2019 county rows: within 0.03 points). County margins for 2019, 2021 and 2022 come from an older Census variance method and run larger than in other editions; do not compare margins across those editions.
`num_three_plus_components`People With 3+ Vulnerability Components	integer	count	—	yes	Estimated number of people in the universe with three or more of the ten CRE components of social vulnerability (PRED3_E; a model-based estimate). Adds across tracts (a county's tracts sum exactly to its county row) and across counties; with the other two bands it sums exactly to num_population. 0 on tracts with no in-universe residents. The numerator of pct_three_plus_components. Descripción completa▾ Estimated number of people in the universe with three or more of the ten CRE components of social vulnerability (PRED3_E; a model-based estimate). Adds across tracts (a county's tracts sum exactly to its county row) and across counties; with the other two bands it sums exactly to num_population. 0 on tracts with no in-universe residents. The numerator of pct_three_plus_components.
`num_three_plus_components_moe`Number Three Plus Components Moe	number	count	—	yes	90 percent margin of error for num_three_plus_components, as published (PRED3_M; to one decimal in the 2021 and 2022 editions, whole numbers otherwise). 0 on tracts with no in-universe residents. Margins never add across places. Descripción completa▾ 90 percent margin of error for num_three_plus_components, as published (PRED3_M; to one decimal in the 2021 and 2022 editions, whole numbers otherwise). 0 on tracts with no in-universe residents. Margins never add across places.
`pct_one_two_components`Share With 1-2 Vulnerability Components	number	proportion	—	yes	Estimated share of the universe population (num_population) with one or two of the ten CRE components of social vulnerability (0-1). A model-based estimate, served as published (PRED12_PE / 100; the source rounds to two decimals of a percent) and equal to num_one_two_components / num_population within that rounding. NULL only where num_population is 0 (the published 0, or the 2021 edition's modeled rate, over no residents is set to NULL as known-bad). Descripción completa▾ Estimated share of the universe population (num_population) with one or two of the ten CRE components of social vulnerability (0-1). A model-based estimate, served as published (PRED12_PE / 100; the source rounds to two decimals of a percent) and equal to num_one_two_components / num_population within that rounding. NULL only where num_population is 0 (the published 0, or the 2021 edition's modeled rate, over no residents is set to NULL as known-bad).
`pct_one_two_components_moe`Percent One Two Components Moe	number	proportion	—	yes	90 percent margin of error for pct_one_two_components (0-1 scale), as published (PRED12_PM / 100). Equals num_one_two_components_moe / num_population within rounding (2019 county rows: within 0.03 points). County margins for 2019, 2021 and 2022 come from an older Census variance method and run larger than in other editions; do not compare margins across those editions. Descripción completa▾ 90 percent margin of error for pct_one_two_components (0-1 scale), as published (PRED12_PM / 100). Equals num_one_two_components_moe / num_population within rounding (2019 county rows: within 0.03 points). County margins for 2019, 2021 and 2022 come from an older Census variance method and run larger than in other editions; do not compare margins across those editions.
`num_one_two_components`People With 1-2 Vulnerability Components	integer	count	—	yes	Estimated number of people in the universe with one or two of the ten CRE components of social vulnerability (PRED12_E; a model-based estimate). Adds across tracts (a county's tracts sum exactly to its county row) and across counties; with the other two bands it sums exactly to num_population. 0 on tracts with no in-universe residents. Descripción completa▾ Estimated number of people in the universe with one or two of the ten CRE components of social vulnerability (PRED12_E; a model-based estimate). Adds across tracts (a county's tracts sum exactly to its county row) and across counties; with the other two bands it sums exactly to num_population. 0 on tracts with no in-universe residents.
`num_one_two_components_moe`Number One Two Components Moe	number	count	—	yes	90 percent margin of error for num_one_two_components, as published (PRED12_M; to one decimal in the 2021 and 2022 editions, whole numbers otherwise). 0 on tracts with no in-universe residents. Margins never add across places. Descripción completa▾ 90 percent margin of error for num_one_two_components, as published (PRED12_M; to one decimal in the 2021 and 2022 editions, whole numbers otherwise). 0 on tracts with no in-universe residents. Margins never add across places.
`pct_zero_components`Share With No Vulnerability Components	number	proportion	—	yes	Estimated share of the universe population (num_population) with none of the ten CRE components of social vulnerability (0-1). A model-based estimate, served as published (PRED0_PE / 100; the source rounds to two decimals of a percent) and equal to num_zero_components / num_population within that rounding. NULL only where num_population is 0 (the published 0, or the 2021 edition's modeled rate, over no residents is set to NULL as known-bad). Descripción completa▾ Estimated share of the universe population (num_population) with none of the ten CRE components of social vulnerability (0-1). A model-based estimate, served as published (PRED0_PE / 100; the source rounds to two decimals of a percent) and equal to num_zero_components / num_population within that rounding. NULL only where num_population is 0 (the published 0, or the 2021 edition's modeled rate, over no residents is set to NULL as known-bad).
`pct_zero_components_moe`Percent Zero Components Moe	number	proportion	—	yes	90 percent margin of error for pct_zero_components (0-1 scale), as published (PRED0_PM / 100). Equals num_zero_components_moe / num_population within rounding (2019 county rows: within 0.03 points). County margins for 2019, 2021 and 2022 come from an older Census variance method and run larger than in other editions; do not compare margins across those editions. Descripción completa▾ 90 percent margin of error for pct_zero_components (0-1 scale), as published (PRED0_PM / 100). Equals num_zero_components_moe / num_population within rounding (2019 county rows: within 0.03 points). County margins for 2019, 2021 and 2022 come from an older Census variance method and run larger than in other editions; do not compare margins across those editions.
`num_zero_components`People With No Vulnerability Components	integer	count	—	yes	Estimated number of people in the universe with none of the ten CRE components of social vulnerability (PRED0_E; a model-based estimate). Adds across tracts (a county's tracts sum exactly to its county row) and across counties; with the other two bands it sums exactly to num_population. 0 on tracts with no in-universe residents. Descripción completa▾ Estimated number of people in the universe with none of the ten CRE components of social vulnerability (PRED0_E; a model-based estimate). Adds across tracts (a county's tracts sum exactly to its county row) and across counties; with the other two bands it sums exactly to num_population. 0 on tracts with no in-universe residents.
`num_zero_components_moe`Number Zero Components Moe	number	count	—	yes	90 percent margin of error for num_zero_components, as published (PRED0_M; to one decimal in the 2021 and 2022 editions, whole numbers otherwise). 0 on tracts with no in-universe residents. Margins never add across places. Descripción completa▾ 90 percent margin of error for num_zero_components, as published (PRED0_M; to one decimal in the 2021 and 2022 editions, whole numbers otherwise). 0 on tracts with no in-universe residents. Margins never add across places.
`num_population`Universe Population	integer	count	—	yes	The CRE universe population (POPUNI): residents excluding people in adult correctional and juvenile facilities and college dormitories. The Census's own population estimate (decennial census base updated with the Population Estimates Program and treated as exact by the model), so no margin is published. The three band counts sum exactly to it; the denominator of every share. 0 on tracts with no in-universe residents (all-water and special land-use tracts). Descripción completa▾ The CRE universe population (POPUNI): residents excluding people in adult correctional and juvenile facilities and college dormitories. The Census's own population estimate (decennial census base updated with the Population Estimates Program and treated as exact by the model), so no margin is published. The three band counts sum exactly to it; the denominator of every share. 0 on tracts with no in-universe residents (all-water and special land-use tracts).

Filtros

Parámetros de consulta que acepta el endpoint de hechos. Los valores separados por comas son OR dentro de un parámetro; varios parámetros se combinan con AND.

Parámetro	Tipo	Valores permitidos	Notas
`year`	year_exact	cualquier valor	—
`year_min`	year_range	cualquier valor	—
`year_max`	year_range	cualquier valor	—
`detail`	detail	`countiestracts`	enum validado, predeterminado: counties
`district_code`	foreign_key	cualquier valor	valores múltiples
`county_fips`	foreign_key	cualquier valor	valores múltiples
`tract_geoid`	foreign_key	cualquier valor	valores múltiples
`district_type`	dimension_attribute	7 valores permitidos▾ `commission_charter` `resa` `standard` `state_agency` `state_charter` `state_school` `state_special`	valores múltiples, enum validado

Ejemplos de solicitudes

Ejecútalos contra la URL base: devuelven datos en vivo.

All rows (first page), counties level

curl "https://georgiacivicdata.org/api/v1/census/community_resilience"

counties data for 2024

curl "https://georgiacivicdata.org/api/v1/census/community_resilience?year=2024"

One county (FIPS 13121), 2024

curl "https://georgiacivicdata.org/api/v1/census/community_resilience?county_fips=13121&year=2024&detail=counties"

Download 2024 as CSV

curl "https://georgiacivicdata.org/api/v1/census/community_resilience?year=2024&format=csv"

Notas y limitaciones

Manejo de NULL

El cero es un valor real y reportado.

Limitaciones

These are MODEL-BASED ESTIMATES, not counts: the Census Bureau's Community Resilience Estimates (CRE) model one year's American Community Survey responses, so every figure has a 90 percent margin of error (the _moe columns). Each year is a separately modeled edition: compare years with care, and 2020 (pandemic-era survey data) with extra care. Tract rows exist only for 2020-2024 (2020 tract boundaries); the 2016-2019 editions are on 2010 tracts and are served for counties only. A change between years mixes real change, sampling error in each year's ACS sample and method change, so never treat a small change as a trend. The population base is the 2010 Census for 2016-2019 and the 2020 Census from 2020 on. The 2016-2018 and 2020 editions are a historical series first published in May 2026 with the 2024 edition's method; 2019 (released in 2021, the first edition after the experimental CRE of 2018 data was redesigned) and 2021-2023 are earlier releases. The Census urges particular caution with the 2020 edition: the pandemic disrupted 2020 ACS collection, its 1-year weights were experimental and its variance method was adapted, so compare 2020 with other years with extra care. County margins of error in 2019, 2021 and 2022 come from an older Census variance method and run much larger than in the other editions (median county margin for the 3+ count about 26-37 percent of the estimate, versus 18-22 percent), so margins are not comparable across those editions. Tract estimates are much less precise than county ones: the median tract margin for the 3+ count is 55-69 percent of the estimate, so many tract differences are within the margins. A component of social vulnerability is one of ten household or individual risk factors measured in the ACS (including a low income-to-poverty ratio, disability, no health insurance, age 65 and over, crowding, a communication barrier, employment and broadband access); the Census publishes only the 0 / 1-2 / 3+ groups, never the individual components or which ones apply. Universe (num_population): residents excluding people in adult correctional and juvenile facilities and college dormitories; nursing-home and military-barracks residents are included with components assumed by the Census (the 2019 file layout calls it total population). The universe is the Census's own population estimate, treated as exact, so it has no margin. Shares are served as published (two decimals of a percent, divided by 100) and equal count / num_population. Twelve tracts in each 2020-2024 edition (five all-water and seven special land-use tracts) have no in-universe residents: their counts and count margins are 0 and their shares and share margins NULL. There is no state or school-district row: a statewide count is the sum of the 159 county counts (exact, but with no margin of error).