2026 Census Test — the first of two field tests before the 2030 Census

The **2026 Census Test** is, in the Census Bureau's own words, **"the first of two major field tests planned to help the Census Bureau prepare for the 2030 Census."** It is a rehearsal, not a count: nothing from it becomes official population data.^[1]

The Bureau splits the exercise into two approaches. **Viability testing** means "trying out new methods or processes to see if they work well enough to use in the 2030 Census." The page pairs this with a second approach under the same heading, framing the test as a way to decide **what carries over** to 2030.^[1]

This is the part most likely to mislead. As checked on **18 September 2026**, the page still says the test **"will be conducted in six sites"** — future tense — while the schedule on the same page places **Census Day on April 1, 2026** and says **"Summer 2026: 2026 Census Test concludes."**^[1]

**Both of those dates are in the past.** The straightforward reading is that **the page wording has not been updated**, not that the test failed to happen. **This page does not report results**, and **whether findings have been published is not something this source establishes.**^[1]

The Bureau lists the sequence as: **Fall 2025** — outreach raises awareness and recruiting begins for temporary census jobs in the test sites; **Spring 2026** — the public begins responding; **Census Day, April 1, 2026** — participants are asked to respond based on where they live as of that date; **Summer 2026** — the test concludes.^[1]

★**Census Day works the same way in the real decennial census.** It is a reference date, not a deadline: responses arriving later still describe the household as it stood on that date. Carrying that convention into the test is itself part of the rehearsal.^[1]

The Bureau names **five enhancement areas identified from lessons learned in the 2020 Census**: **Simplify Data Collection**; **Modernize Group Quarters Enumeration**; **Integrate Data Processing with Data Collection in Near Real-Time**; **Optimize Operational Support Infrastructure**; and **Build on Alternative Data Sources and Methods for Use Throughout the Decade**. It calls these **"our top research priorities."**^[1]

★**"Group quarters" is the term worth knowing.** It covers places where people live in group arrangements rather than households — the category that has historically been hardest to count accurately. Listing it among five priorities signals where the 2020 difficulties were concentrated.^[1]

The test runs **in six sites and also includes a nationally representative sample**. The Bureau says the sites "provide optimal opportunities to implement all the changes and enhancements planned for the test, under real conditions, and on a larger scale than research simulations allow," with the aim of reaching **"hard-to-count and historically undercounted populations."**^[1]

One site is described in detail on the page: **Western Texas**, which **includes Brewster, Jeff Davis, Pecos and Presidio counties**. ★**The other five site names are not quoted here** because the page presents them through expandable sections and downloadable maps rather than as a plain list.^[1]

Separately from the six sites, the Bureau **mails materials to a nationally representative sample of households, inviting them to respond online, by phone or by mail.** It says this will help improve the **2030 Census online questionnaire and invitation materials** and give insight into **self-response rate trends.**^[1]

★**Everything above comes from a single Census Bureau programme page.** That page is the primary source for the test's own design, but **no second institution is cited here**, and **no independent assessment of the test is included.**^[1]

If you need results rather than design, **the Bureau's later publications are the place to look** — this page describes what was planned, and **was still written in the future tense when checked.**^[1]