Updated at 9:05 p.m. ET
If the Trump administration had been allowed to add the now-blocked citizenship question to the 2020 census, it likely would not have had a significant effect on self-response rates, the Census Bureau said Thursday.
Preliminary analysis of a national experiment the Census Bureau conducted earlier this year with two versions of a test census form — one with a citizenship question and one without — suggests that question could lower self-response rates in some parts of the country and for some populations. In a blog post released Thursday, the bureau highlighted a 0.3% difference in the share of participants identifying as Hispanic.
Still, the differences overall were "small," according to Victoria Velkoff, the bureau's associate director for demographic programs who wrote the post.
The bureau's early findings could temper some concerns that including the question would deter households, especially those with noncitizens, from taking part in the constitutionally mandated head count of every person living in the U.S.
The Census Bureau randomly selected approximately 480,000 households across the country, except in remote Alaska and Puerto Rico, to take part in what it has called the "2019 Census Test." Half of those households were asked to complete test forms with the question, "Is this person a citizen of the United States?"
The bureau scrambled to put together the test earlier this year in response to the administration's push for the question. It's not clear when the bureau plans to release a final report on the experiment.
Some critics of the citizenship question are holding their judgment of the bureau's early findings. An earlier study by researchers at the bureau suggested the question would have deterred at least 9 million people from self-responding to the census.
"All other research to date by the Census Bureau has indicated that adding a citizenship question to the census would depress responses among noncitizens and Hispanics," Dale Ho, an ACLU attorney who is helping to represent plaintiffs in lawsuits over the question, said in a written statement. "We look forward to seeing whether the full results of this latest study are consistent with the bureau's previous findings in this regard."
But in a statement, Commerce Secretary Wilbur Ross — who oversees the bureau and approved adding the question — called the preliminary results "gratifying news to those who supported its inclusion."
Throughout the legal battle over the question, opponents raised concerns that adding a citizenship question would force the Census Bureau into spending more time and money to gather responses from reluctant households.
However, the bureau's preliminary analysis of its field test suggests it would not have needed more door knockers to follow up with people in households who did not fill out a form themselves, Velkoff wrote in the blog post.
Velkoff added that it's unclear from these test results how the question could have impacted the "completeness and accuracy" of the 2020 census overall.
The bureau's early findings come more than a year and a half after Ross announced his decision in 2018 to add the hotly contested question. This summer, three federal courts permanently blocked the question from being added — in part because the bureau had not conducted required testing of public reaction to including a citizenship question on 2020 census forms.
The nine-week test took place in the midst of a heated legal battle over the question. By early July, it sparked confusion around the country about why the bureau was continuing to use census forms to ask about people's U.S. citizenship status after a majority of the U.S. Supreme Court ruled to keep the question off.
The bureau has said the test results could be "valuable" to any officials considering adding such a question to future census forms.
After backing down from efforts to use the 2020 census to ask about citizenship status, the Trump administration is now moving forward with compiling government records to produce detailed citizenship data.
In an executive order released in July, President Trump said that he wants the data to be available for state redistricting officials to use when redrawing voting districts after the national head count. A prominent GOP redistricting strategist, Thomas Hofeller, has concluded that this kind of citizenship data could give Republicans and non-Hispanic white people a political advantage.
In his executive order, Trump also left open the possibility of a resurrected political fight over a census citizenship question. The president directed the commerce secretary, who oversees the Census Bureau, to "consider initiating any administrative process necessary to include a citizenship question on the 2030 decennial census."