WASHINGTON – The Trump government's justifications for adding a citizenship question to the 2020 census are not particularly strong.
Trade Secretary Wilbur Ross said his decision to add the question was based on three things. First, he wrote in a memo, while the American Community Survey (ACS) already asks the census for citizenship. The data collected are not specific enough for the Ministry of Justice to enforce the electoral law. Second, he said, he had seen "limited empirical evidence" that a citizenship question would lower response rates. And thirdly, a citizenship question was already well tested because the ten-year census in the nineteenth and twentieth centuries had asked for it (it was discontinued after the 1
The justifications of the Trump administration, which have already been postponed, are being scrutinized as civil rights groups and Democrats file lawsuits to challenge the decision. The challengers say the decision to add a citizenship question is politically motivated and will lead to a shortage of immigrants and minorities who will be afraid to respond to the poll.
The ACS has been asking for citizenship since 2005, but John Thompson, director of the Census Bureau from 2013 to 2017, noted that the ACS is a completely different form than the ten-year census. The ACS consists of several dozen personal questions on things like income, rent, disability and transportation to work, in addition to citizenship. Every year, it reaches 2.6 percent of the population. The 10-year census of 2010 was only 10 questions long and asked very basic questions, including how many people lived in a particular household, phone number, race or ethnicity, and housing status. Thompson noted that much attention is paid to the short-form census, which should be answered by everyone every ten years.
Justin Levitt, who was Deputy Attorney General of Justice at the Ministry of Justice from 2015 to 2015, noted that the ACS has lower stakes than the 10-year census, which he said is the first responsibility that the federal government's constitution gives , If there is a low response rate on the ACS, there are ways for officials to correct their data. But the constitution stipulates that the federal government counts all people. An inaccurate number could disproportionately affect certain states. For example, if undocumented immigrants are not fully counted in an immigrant state because many are afraid to answer a citizenship question, that state could lose congressional seats and a fair distribution of hundreds of billions of dollars in federal funds.
If we get the percentage of the population who happens to have indoor lavatories or outside toilets wrong, it does not have the constitutional extent, "said Levitt." If you ask people for a lot of information and they have already signed up for this other information deliver, then ask them for information on whether they are citizens, feels – as many people have said – less aggressive and less worrying. "
Two former top officials of the Ministry of Justice entrusted with enforcing the 1965 Electoral Law to protect minority voting power, say the Trump administration that existing ACS data is not good enough to enforce the law is untrue.
"I oversaw the enforcement of the Voting Act for nearly two and a half years in the previous administration and can tell you clearly that the strict enforcement of the Voting Law has never required the introduction of a citizenship question in the census," said Vanita Gupta, president and CEO of Leadership Conference on Civil and Human Rights, which led the Justice Department's Civil Rights Division from 2014 to 2017. The Leadership Conference has opposed the adoption of citizenship issues.
Levitt said that most jurisdictions are big enough to cover any mistakes in the ACS estimate to elimi The only places where the ACS did not provide reliable citizenship estimates were tiny towns and communities. According to Levitt, increasing the accuracy of the question of citizenship would be outweighed by the potential for error caused by the addition of an unchecked question.
"By attempting to increase this accuracy from 9.99999 to 10, they cause such deep chaos and such deliberate and predictable underestimation that the amount of errors they bring into the process goes beyond exponential levels, every incremental accuracy they pretend to be flooding, "he said in an interview.
"If the reason is:" There are some cases we can not bring in because the survey data are not accurate enough "to believe that this question will be added in this environment without testing it – and this is unthinkable – it basically goes by the titanic adage, "We're pretty sure this will not cause a problem, but just in case, yes, we'll throw a massive iceberg in the way."
If we It is not of fundamental size.
Justin Levitt, former Deputy Assistant Attorney General for Civil Rights
Although the Census Bureau may spend years examining the wording of a question, it has I have not conducted any tests at all about the implications of adding a citizenship question, it will essentially go blind in the 2020 test and the risk that a citizenship question does not suppress response rates. Civil rights groups say that immigrants and other minority groups are less responsive to a citizenship issue because they fear that the Trump government might use their response to find them and deport them or deport someone they know
Census Bureau Ross warned against adding the question and said it was too risky. But Ross and the White House downplayed the risk and said that their presence on the ACS is a test and there is "limited empirical evidence" that fewer people will respond.
The Census Bureau rejected the citizenship question in 1960. In early 1970, questions were asked only about the longer questionnaires as population sampling became more popular in the 20th century, said Margo Anderson, historian of the Wisconsin-Milwaukee University Census , But asking a question about these long polls is not a substitute for testing the impact of the question on the actual ten-year form.
"1950's been a long time," Anderson said. "The normal method would be to say," OK, if you want to do that, let's test it. "
Kathleen Joyce Weldon, Director of Data Operations and Communications at the Roper Center for Opinion Research, pollster said are cautious about including citizenship questions in surveys for fear that people will not respond, and the seriousness of the census, a national survey conducted by the government, has only kept people from reacting.
Apart from the differences in content between the long-term ACS and the 10-year census, it is risky to accept this question that works well in one survey will work on another.Social scientists are wary of the way survey questions are asked and even with the way questions are asked because the wording of a single fra It may influence how a person responds to all other questions, or whether they continue to answer.
In a January letter to Ross, the American Statistical Association warned that adding a citizenship question would be a problem the careful testing and calibrations the Census Bureau had made so far to prepare the questions for the ten-year survey could invalidate.
"Adding a question at this late stage of the census process leaves no time to adequately test new questions, especially when the tests reveal significant problems, and a new question undermines the validity of the extensive testing of the current issues We believe that the potential for a citizenship issue to create problems strongly contradicts its inclusion at this advanced stage, with the likeliest outcomes being a lower response rate, more non-participation and, ultimately, a sharp increase in costs for non-response tracking. "