The 2024 American Community Survey (ACS) gives a vacancy estimate for a defined geography. It also gives a margin of error (MOE). Record both values together. A vacancy estimate is housing context, not proof that a particular address is empty, abandoned, or part of an estate.
Start with the table and product
Use the 2024 ACS 1-Year Estimates Data Profile DP04 when the geography and release support a one-year estimate. In the Census Data API, the homeowner vacancy rate is `DP04_0004E` and its MOE is `DP04_0004M`. The rental vacancy rate uses `DP04_0005E` and `DP04_0005M`.
The national DP04 row reports a homeowner vacancy rate estimate of **1.0%** with an MOE of **±0.1 percentage points**. That is the displayed 2024 ACS value for the United States. Keep the percent sign and the unit in the log. Do not treat the MOE as an error rate for a deed, title search, or property file.
The source-log fields
| Field | What to record | Example for the national DP04 row |
|---|---|---|
| Product | ACS release and table | 2024 ACS 1-Year Estimates Data Profiles, DP04 |
| Geography | Census geography and code | United States, `us:1` |
| Variable | Estimate and MOE fields | `DP04_0004E` and `DP04_0004M` |
| Estimate | Value, unit, and label | 1.0%, homeowner vacancy rate |
| MOE | Value, unit, and label | ±0.1 percentage points, 90% MOE |
| Retrieval date | Date the table or API response was obtained | 2026-08-08 |
| Source URL | Stable table or API request | DP04 table link or the exact API URL |
| Limitation | What the row cannot establish | No address, owner, title, or probate status |
For a repeatable API request, use the same product, variables, and geography each time. The request pattern is:
```text https://api.census.gov/data/2024/acs/acs1/profile?get=NAME,DP04_0004E,DP04_0004M&for=us:1 ```
If the API requires a key in your environment, add it according to the Census API instructions and save the response with the research record. Preserve the request URL, response date, table name, and any annotations. A later reviewer should be able to tell whether a changed value came from a different product, geography, variable, or release.
What the margin of error does and does not say
The DP04 table labels its value as a 90% margin of error. The Census Bureau describes the estimate plus or minus the MOE as an approximate 90% confidence interval. That description does not make the estimate a measured count of vacant parcels. It also does not account for every source of nonsampling error.
This workflow records the estimate and the published MOE. It does not manufacture an interval, round a result into a more precise figure, or combine margins of error from unrelated rows. If an analysis needs a confidence interval for a derived statistic, document the formula, inputs, assumptions, and method separately.
Why estate research needs a second evidence trail
ACS vacancy data describes housing units in a survey universe. It does not identify a parcel, recorded owner, decedent, executor, tenant, lien, or court proceeding. A county-level result cannot establish that a named property is vacant.
For a property-specific question, log the parcel identifier, legal description, deed or recorder result, assessor record, tax status, court file, source URL, retrieval date, and unresolved conflicts. The estate-property ACS margin of error research page provides related context. For broader public-record evidence gathering, see investigative research services.
Method and sources
This workflow uses the Census Bureau's 2024 ACS DP04 profile and its published variable definitions, reviewed August 8, 2026. The national example records the displayed estimate and MOE only. It does not infer an address count, an estate-property rate, or a new confidence interval.
- Census Bureau, 2024 ACS DP04 table
- Census Data API, 2024 ACS 1-Year Profile DP04 variables
- Census Bureau, 2024 ACS accuracy documentation
**Professional disclaimer:** This article is general research, not legal, title, tax, appraisal, or housing advice.