demographics_single_parent_county_year
County × year share of families headed by a single parent, from ACS 5-year 2023 (2019–2023, table B11003). Single parent = householder no spouse present with own children under 18 (male + female householder types combined). Two shares are exposed: of all families, and of families raising own children (the standard single-parent rate). Margin of error is carried, and single_parent_coverage flags small-county estimates whose coefficient of variation exceeds 0.30 as LOW_PRECISION. Part of the P1 social-context (SDOH) cluster — a household-structure / social-isolation indicator, homed here as household demographics. Tier B (ACS sample estimate).
overview
All ~3,222 counties covered (ACS is a full-county survey). single_parent_coverage marks counties whose single-parent count has a coefficient of variation above 0.30 as LOW_PRECISION; the rest are SUPPORTED. No confidentiality suppression and no non-reporting gaps apply to ACS, so UNSUPPORTED occurs only for the degenerate zero-families case.
current vintage — ACS 2019–2023 5-year (end year 2023)
history — Single vintage loaded (2019–2023). ACS publishes B11003 back to the 2005–2009 5-year window; earlier vintages not yet ingested.
source & licensing
fields
| name | type | definition |
|---|---|---|
| country_idkey | string | ISO alpha-2 country code (always 'US' for domestic tables).Part of primary key. |
| county_idkey | string | 5-character FIPS code identifying the county.Part of primary key. Joins dim.counties on county_id. |
| county_name | string | Human-readable county name corresponding to county_id. |
| families_with_own_children | int64 | Families with own children under 18 (married-with-children + both single-parent types). Denominator for pct_single_parent_of_with_children. |
| female_single_parent | int64 | Female householder, no spouse present, with own children under 18 (B11003_016). |
| male_single_parent | int64 | Male householder, no spouse present, with own children under 18 (B11003_010). |
| married_couple_families | int64 | Married-couple families (B11003_002). |
| pct_single_parent_of_families | float64 | single_parent_families / total_families (0–1). |
| pct_single_parent_of_with_children | float64 | single_parent_families / families_with_own_children (0–1). The standard single-parent rate — share of families raising children headed by one parent. |
| single_parent_coverage | string | Per-cell coverage/precision flag (context/guides/coverage-flags.md). SUPPORTED = usable; LOW_PRECISION = CV > 0.30, wide CI; UNSUPPORTED = no families (degenerate). Filter on this for small-area work. |
| single_parent_families | int64 | Families headed by a single parent: householder no spouse present with own children under 18 (B11003_010 male + B11003_016 female). |
| single_parent_families_moe | int64 | 90% margin of error for single_parent_families, root-sum-of-squares of the male and female component MOEs. |
| state_idkey | string | 2-character FIPS code identifying the state.Part of primary key. Joins dim.states on state_id. |
| total_families | int64 | Total families in the county (B11003_001, the universe). |
| yearkey | int64 | Reference year of the observation.Part of primary key. |
joins
how to use this table
single_parent_families = B11003_010 (male householder, no spouse, with own children <18) + B11003_016 (female householder, no spouse, with own children <18). pct_single_parent_of_families = single_parent_families / B11003_001 (total families). pct_single_parent_of_with_children = single_parent_families / (B11003_003 married-with-children + B11003_010 + B11003_016). Combined MOE = sqrt(MOE_010^2 + MOE_016^2). Coverage flag from CV = (MOE/1.645)/estimate: CV > 0.30 → LOW_PRECISION. Computed in dbt from raw.acs_county_single_parent_2023.
Small-area point estimates without checking single_parent_coverage / the MOE (the estimate is unstable in small counties). Causal interpretation of family structure (this is a descriptive area-level share, not an effect). Treating consecutive ACS 5-year vintages as independent observations — their windows overlap.
Standard ACS sampling error — MOE widens in small counties; the single_parent_coverage flag uses a CV > 0.30 threshold to mark LOW_PRECISION cells. ACS 5-year overlap means consecutive vintages are not statistically independent. 'Single parent' here is a family-type measure (householder with no spouse present and own children), not a count of children living with one parent (that is B23008 / the CHR children-in-single-parent measure).