By MIKE SCHNEIDER The Trump administration is eliminating four out of the six locations that had been slated for a practice test to try out new methods for the 2030 census, raising concerns that the U.S. Census Bureau might not learn enough about communities that have been traditionally difficult to count. Related Articles Attorney says Gabbard is holding up a complaint about her actions, which her office denies FACT FOCUS: Images of NYC mayor with Jeffrey Epstein are AI-generated. Here’s how we know Trump’s $10 billion lawsuit against the IRS raises conflict of interest concerns Trump says he won’t tear down the Kennedy Center arts venue but it needs to be closed for repairs Ed Martin removed as head of Justice Department’s ‘Weaponization Working Group’ The test, which started Monday, will be conducted only in Huntsville, Alabama, and Spartanburg, South Carolina, according to a notice submitted by the Commerce Department that will be formally published on Tuesday. The Commerce Department oversees the Census Bureau. Four other sites — Colorado Springs, Colorado, tribal lands in Arizona, western North Carolina and western Texas — originally were included when the Census Bureau announced the locations in 2024. The bureau didn’t respond to an emailed inquiry on Monday about the reasons for the reduced number of sites. In a statement on its website, it said it “remains committed to conducting the most accurate count in history for the 2030 Census and looks forward to the continued partnership with local communities.” Mark Mather, an associate vice president at the Population Reference Bureau, a nonpartisan research group, said limiting the test to just two metro areas in the South would be “a step backward.” “The Census Bureau would be essentially flying blind into communities that need testing most — tribal lands, rural areas with limited connectivity and places with historically low response rates,” Mather said. “You can’t fix what you don’t test.” The test is supposed to give the statistical agency the chance to learn how to better tally populations that were undercounted in the 2020 census and improve methods that will be used in 2030. It also allows it to test its messaging and its ability to process data as it’s being gathered. Among the new methods being tested for 2030 is the use of U.S. Postal Service workers to conduct tasks previously done by census workers. FILE – People walk past posters encouraging participation in the 2020 Census in Seattle’s Capitol Hill neighborhood, April 1, 2020. (AP Photo/Ted S. Warren, File) The original six test sites had been picked for a variety of reasons, including a desire to include rural areas where some residents don’t receive mail or have little or no internet service. Others, including tribal land, fast-growing locations with new construction, and dorms, care facilities and military barracks had been picked because their residents are traditionally hard to count. Ahead of the last census in 2020, the only start-to-finish test of the head count was held in Providence, Rhode Island, in 2018. Plans for other tests were canceled because of a lack of funding from Congress. The once-a-decade head count determines how many congressional seats and Electoral College votes each state gets. It also guides the distribution of $2.8 trillion in annual federal spending. Follow Mike Schneider on the social platform Bluesky: @mikeysid.bsky.social. Follow the AP’s coverage of the U.S. Census Bureau at https://apnews.com/hub/us-census-bureau. ...read more read less