The following is a chapter from Trumpism and its Discontents. Click to download a PDF of the book here.

By G. Cristina Mora

For Latinx individuals working within civil rights organizations, the 2020 census promised to be the most complete enumeration of their community to date. The count would mark fifty years since community stakeholders, from Latinx politicians to academics to Spanish-language media, had worked together to ensure that Latinx people were fully counted. Groups like UnidosUS (formerly the National Council of La Raza), the National Association of Latino Elected Officials, Univision Media Corporation, the National Hispanic Chamber of Commerce, and many others had worked with the US Census Bureau for decades to help develop the agency’s messaging and enumeration plans within Latinx communities. This effort was necessary because Latinx individuals are hard to count. Indeed, farmworkers, immigrants, and the poor have a higher probability of being missed by census efforts, and over time this factor has negatively affected the Latinx count. To improve this situation, community groups have worked closely with the Census Bureau, even committing their own resources to independently reach out and communicate to Latinx individuals the importance of being counted. Gradually, these collaborative efforts have paid off as the Latinx undercount estimate has shrunk from about 5.5 percent in 1980 to 1.5 percent in 2010.1

Advocates understand that a complete count translates into more financial resources for their community and provides the political tools to shift the narrative about Latinx in the United States. Since helping the Census Bureau to develop a distinct “Hispanic/Spanish Origin/Latino” category in the 1970s, Latinx community organizations have used census figures to show that the growing Latinx population is sizable and thus a force to be reckoned with on a national scale.2 Indeed, post-2000 demographic forecasts about the increasing “browning of America” hinge in large part on a complete count of the Latinx population.3 Such a thorough enumeration, stakeholders anticipate, can show that Latinx people are the demographic future of America and can no longer be ignored.4

Elected officials in progressive areas with large numbers of Latinxs also recognized the critical importance of the 2020 count. In California, which is home to about 30 percent of the nation’s Latinx population, the state legislature allocated about $90 million for census outreach efforts, up from just $2 million allocated in 2010.5 Officials in New York City also allocated close to $5 million for census efforts, marking the first time that the city had ever created an official budget for enumeration.6 Moreover, governors in places with high or increasing Latinx populations have formed coalition boards that are helping with Latinx census outreach.7

The actions of the Trump administration, however, threw much of the Latinx census momentum into a tailspin. This chapter examines the issue of Latinx census politics in the Trump era to show how the policies and practices of the administration represent a severe escalation of a more historic type of census racial politics that aims to curb minority rights. Such politics target minority data collection efforts and destabilize Census Bureau and community stakeholder relationships. Specifically, this chapter focuses on two issues: (1) the decision to dismiss the combined Race and Ethnicity question and (2) the proposal to include a citizenship question. Each issue has the potential to depress Latinx enumeration, drive a wedge between Latinx community advocates and census officials, and influence the overall census count. 

To be sure, the administration’s abrupt actions reflect a longer-term political backlash directed toward Latinx communities and immigrants more generally. The backlash sees census data manipulation as an opportunity to curb minority representation and curb the Latinx community’s political momentum. In the following sections, I first provide an overview of the relationship between Latinx community stakeholders and the Census Bureau. I then describe the decision to eliminate the combined Race/Ethnicity question and proposal to implement a citizenship question, showing how these two actions effected decades of research and outreach toward Latinxs. I conclude by discussing how recent census politics fit into a broader effort to delimit Latinx political power and influence the discourse on race and diversity in America.

Unlikely Bedfellows: Census Bureau Officials and Latinx Advocates 

The relationship between Latinx stakeholders and Census Bureau officials has never been without tension, in part because of the complicated political and even damaging ways that minority census data has historically been used. The first recorded interaction between stakeholders and Latinxs came in 1930 when census officials inserted a “Mexican” racial category on the decennial census form. Until then, Mexican Americans, like all other Latinxs in the country, had been categorized simply as “White,” placing them in a category with the descendants of European migrations. The decision to create the category was motivated in large part by congressional interest in curbing Mexican migration, which had spiked after the Mexican Revolution.8 Indeed, in a congressional hearing in late 1928, Senator William James Harris argued that Mexican immigration was hurting American workers and that the undocumented were entering the country in unprecedented numbers. He inserted two items in the Congressional Record: a letter asking for census counts of Mexican immigrants, describing them as illiterate and undocumented, and a news clipping that noted that Mexicans, who were mostly of “Indian Blood”, were depleting local and state resources in California.9   At the time, Harris’s inclusions echoed much of the anti-immigrant and anti-Mexican sentiment of the 1920s, and calls for Mexican deportation were common in Depression-era newspapers. In their seminal work Decade of Betrayal: Mexican Repatriation in the 1930s, Francisco Balderama and Raymond Rodriguez10 contend that the calls coincided with federal and local-level actions that ultimately deported more than one million Mexicans, some of whom were American citizens.

The Mexican consulate quickly vocalized its disapproval of the new category, marking the first time that a foreign nation would meddle in Census Bureau affairs.11 Mexican American political leaders also criticized the Bureau’s actions. They complained that the category was meant to single out Mexicans at a time of Americans’ increased fear and unjust state violence. Stakeholders likely surmised that the Census Bureau’s new category would also exacerbate fear and mistrust of the government among Latinxs. 

Additionally, organizations like the League of United Latin American Citizens and the American GI forum accused census officials of instituting the Mexican racial category to question the citizenship and belonging of this community. They saw the distinct “Mexican” racial category as an affront to Mexicans’ official “White” status that had been negotiated during the Treaty of Guadalupe and that had provided Mexicans with US citizenship and property rights. Stakeholders reasoned that classifying Mexicans as nonwhite would call into question this historic agreement and would also make the Mexican community an easier target for racial discrimination. At the time, Latinxs were already subject to rampant discrimination and were even targets of racialized terror campaigns and lynching in areas such as Texas and New Mexico.12 Latinx advocates feared that an official status as nonwhite for Latinx individuals would exacerbate this anti-Latinx sentiment. 

The pressure from groups and foreign officials seems to have worked. By the 1940 Census, the Census Bureau had returned to categorizing all Latinxs as “White.” Over the coming years, however, the Bureau would develop uneven indicators of Latinx nationality. In 1950, for example, it developed a “Spanish Surname” category but only for respondents who lived in the Southwest and parts of the Northeast. And in 1960, the Bureau asked a sample number of respondents to indicate if they were “Spanish Speaking.”13

With the advent of the Civil Rights Movement and the rise of minority rights politics, new organizations emerged to focus on the conditions of Mexicans, Puerto Ricans, and even Cubans in the United States. These stakeholders, some with links to the Johnson administration, would eventually come to envision Latinx panethnically and would call for the development of a national “Hispanic” lobby.14 Spanish-language media reinforced these efforts when they covered the movement and simultaneously attempted to connect their own audiences across the country through new United States–produced cultural programming that communicated a unified Hispanic identity.15

Data quickly became a key part of the “Hispanic” panethnic movement of the 1970s. By then, Mexicans, Puerto Ricans, and others realized that without census data they could not make credible claims about underrepresentation and the social ills that plagued Mexicans, Puerto Ricans, and others. Census Bureau reports on poverty, for example, mixed Latinx data with that of Irish and Italians and thus reported only on Black and White income trends. Similarly, the Bureau of Labor only reported on Black and White unemployment trends, thus disguising Latinx realities. Community leaders quickly called on the Census Bureau to find a way to distinguish 
Latinx people from “Anglos” and those of European descent. Groups like the National Council of La Raza and the Mexican American Legal Defense and Education Fund, for example, began threatening legal and protest actions against census officials for classifying Latinxs separately.16

The Census Bureau was initially reluctant to reclassify Latinxs, even though it acknowledged that its “Spanish Surname” and “Spanish Speaking” categories were imperfect. Nevertheless, it soon began working with activists once it was pressured by the Johnson administration to deal with the issue. The Bureau created an advisory committee for the Spanish speaking in 1972, marking the first time that minority group stakeholders would be allowed to advise it.  Working closely with this committee, the Bureau implemented its first question on “Hispanic/Spanish Origin” on the decennial short form in 1980.17

Over time, Latinx advocacy groups would come to see their work with the Census Bureau in two ways. First, they helped the Bureau develop a set of best practices by consulting on issues like question wording and Spanish language–form translation and by advising officials on how to develop a more culturally relevant Latinx public relations strategy. Second, advocacy groups have become ambassadors for the census count. They use their own resources to develop unique census messaging campaigns, and in so doing, they legitimize the count itself. In effect, Latinx groups hope that their organizational support will help Latinxs overcome decades of mistrust and suspicion vis-à-vis the government. Thus, advocates reiterate the issue of census privacy laws and use their credibility to debunk arguments that census data will be shared with the Department of Homeland Security.18

However, despite these efforts, the relationship of Latinx advocates and the Census Bureau still contains some tension. Latinx advocates prefer that the Bureau dedicate more resources to eliminating the undercount, but such extensive canvassing is usually tied to overall census budgets, which are often held hostage politically. Following historical precedent, the Republican-led Congress initially refused to significantly increase the Bureau’s budget in 2017, thus inhibiting the agency’s ability to do more to alleviate the undercounting of minorities. In addition, in 2018, Latinx census advocacy efforts were further tested in a more radical manner. The Trump administration imperiled the 2020 Latinx count by making abrupt decisions about the combined race/ethnicity question and the census citizenship question issue. 

Latinx and the Question of Race and Ethnicity 

Since its implementation, the Hispanic, Latino, or Spanish origin category has always been tallied as a count of ethnicity distinct from race. Census respondents first answer whether they are Hispanic, Latino, or Spanish origin and then mark a category to indicate whether their race is White, Black or African American, Native American or Alaska Native, one of seven Asian (such as Chinese or Korean) or Pacific Islander categories, or “Some other race.” Such a configuration of two separate questions has historically suggested two things. First, it implies that Hispanic/Latino ethnicity is not a race and that this population can be slotted into the various racial categories. Such a view has been promoted by those who argue that Latinxs can be of any race—and thus point to those who describe themselves as Afro-Latino on the one hand and those who describe themselves as White Hispanics on the other. This view has received push-back from advocates who argue that the Latinx identity is itself a mixed, mestizo racial identity. They contend that Latinx are not simply White or Black but rather are inherently mixed and that Latinxs have been historically and systematically racialized. Although skin tones can vary, the identity of being Latinx is connected to minority, outsider status. The second issue that the two questions suggest is that there is a real difference between ethnicity and race. Those advocates who claim that Latinxs are ethnoracial, meaning that their ethnicity is often racialized and that the population is composed of different racial identities, resist the two-question format. 

Although these views could have remained components of theoretical debates, the issue of the two census questions has had real consequences that pose problems for the Census Bureau. The first issue is that about 50 percent of Latinxs consistently mark the “Some other race” (SOR) box on census forms. 19 Moreover, “the overwhelming majority” of the SORs write a Latino/Hispanic panethnic term or nationality in the space provided.20 Following its policy, the Census Bureau responds to this missing data by reclassifying all SOR Latinxs as “White,” a practice that coincided historically with the idea that Latinxs should mark the “White” category on census forms. 

The issue was first addressed in the late 1990s as the Bureau held a series of hearings with Latinx community stakeholders. At these hearings the Bureau noted the difficulties presented by the SOR category and asked community leaders for assistance. Community advocates responded positively as by now nearly every major Latinx advocacy group viewed census enumeration as a central Latinx political issue.21 The Bureau had considered developing a new combined question that would include “Hispanic/Latino” as an option among other racial identities. Advocates’ responses to this proposal, however, were mixed because stakeholders were unsure about how such a change would affect the count. On the one hand, several stakeholders acknowledged the confusion and difficulties that the current two-question format had caused for some respondents,  especially those who considered Latinidad to be a mixed, mestizo racial identity.22 On the other hand, these community advocates had invested various resources in helping the Bureau to legitimate and popularize the two-question format; advising individuals to “check Hispanic first and race second” had become common practice. Stakeholders understood that any change in the questions would require a dramatic shift in messaging and practice. Importantly, some Afro-Latinx organizations emerged as advocates of the two-question format during this time. They feared that the combined question could weaken the Black/Hispanic count and consequently could undermine Afro-Latinx representation. Nevertheless, all Latinx community stakeholders called on the Bureau to continue studying the issue as it prepared for future counts. 

The Census Bureau returned to the issue after the 2010 count when it held a series of public meetings to discuss the new question, tested the combined query extensively on the American Community Survey (ACS), and expanded the Hispanic/ Latino advisory board, with a focus on solving the SOR dilemma. Latinx advocates participated in hearings and on the advisory board and awaited the Bureau’s conclusion after its testing, still unsure how any change in format would affect the overall Latinx count. The Bureau released a report on the SOR issue in 2017, stating that “the combined question formats had significantly lower percentages of respondents reporting Some Other Race or invalid responses, as well as significantly lower percentages of missing responses than the Separate Questions format” had garnered.23 Moreover, the report noted that Hispanic respondents identified themselves as “Hispanic” alone at significantly higher rates when responding to the combined question formats as compared with their responses in the Separate Questions format. Additionally, the Bureau found that all minorities, not simply Latinxs, were more likely to respond to one combined question than to two questions. Last, the single question format also greatly reduced the number of nonresponses to the Hispanic question itself. Thus, by late 2016, the Bureau had moved ahead with a strong recommendation to use a new combined question format. 

The Bureau had finished most of its testing on the issue when President Trump came into office in 2017. Still finishing a term he had started under President Obama, Census Director John Thompson intended to use the single (combined) question on the November 2018 pre-census trials. However, his plans were cut short when the Office of Management and Budget, under Mick Mulvaney, suddenly stopped the momentum on the issue.24 This decision went against the research and the majority of community recommendations to date. In June 2017, Director Thompson announced his retirement, and President Trump did not appoint a replacement in his absence. Instead, responsibility for the census officially fell under the jurisdiction of Wilbur Ross, the Secretary of Commerce. In January 2018, the Census Bureau announced that the 2020 Census would resume using the two-question format. 

Latinx advocacy groups quickly criticized the decision. Arturo Vargas of the National Association of Latino Elected Officials contended that the abrupt decision had overturned years of inquiry and testing and impeded scientific progress.25 Kenith Prewitt, the former head of the Census Bureau, echoed this concern and noted that the decision went against academia’s best recommendations.26

The elimination of the combined question, while abrupt and surprising, reflects more historical top-level administrative tactics to curb efforts to improve minority undercounts. Indeed, during the 1980s, arguments about undercounts of minorities had led Census Bureau statistician Barbara Bailar, then head of the American Statistical Association, to study the issue of adjustment. Bailar developed an analytical strategy that could help impute the number of missing persons, mainly the poor and racial minorities. Republicans quickly pushed back against this effort and lobbied to use unadjusted numbers for reapportionment in 1990. Soon after, the Commerce Department under President George H.W. Bush announced that it would not use “hotdecking” or any other adjustment techniques to account for missing minority data. Bailar then quit, denouncing the Bush administration for ignoring years of government and academic research and recommendations.27

At the same time, the SOR question controversy has incited conservative criticism about racial classification, led perhaps most vocally and visibly by Ward Connerly and Mike Gonzalez of the Heritage Foundation. They contend that the Bureau should eliminate racial categories altogether because such classifications tend to “straightjacket” Americans, especially those of mixed descent, into Black/Hispanic/ White racial categories. They also argue that racial classification encourages a divisive understanding of American identities.28 In effect, the new issue of the combined question was used to drum up an old argument used frequently by the advocates of color-blind politics. Latinx and many other minority advocates see this development as a dangerous proposition used to undermine data collection efforts and hide the discrimination and underrepresentation trends that census data can reveal.29

There was little further engagement on the subject after January 2018. Latinx leaders believed that the Bureau’s own report on the question spoke for itself and that some hidden maneuvering at the level of the Trump administration—likely motivated by racial animus and employed in an attempt to curb Latinx representation—was at work. But with fewer than two years remaining until the count and finding little recourse to overturn the administration’s decision, there was not much they could do. Moreover, Latinx advocates were preparing themselves to face a much bigger controversy that would officially unfold only weeks later.   

The Census Citizenship Question 

On March 26, 2018, Secretary of Commerce Wilbur Ross announced that the 2020 Census would now include a new question on the decennial short form: “Is this person a citizen of the United States?” Such a question marked a radical departure from census protocol. The Census Bureau had not asked all Americans about their citizenship since 1940, when the nation was embroiled in World War II and had established the national registry. The question was subsequently removed in 1950 after the war had ended, and since then the question has either been excluded from census enumeration efforts or included only on sample long forms.30 However, in 2000, the citizenship question was added to the American Community Survey. 

The introduction of the new citizenship question on the 2020 Census came as a surprise to many, including Latinx advocacy organizations, for at least three reasons. First, the request for the question came at the behest of Wilbur Ross, the head of the parent body, the Department of Commerce, and not from anyone within the Census Bureau. There had been no public support from within the Bureau to implement the question. Moreover, there had been no official discussions with the Latinx Advisory Commission on the issue. Like the sudden decision to eliminate the combined race/ethnicity question, the addition of the citizenship question seemed to come from the top, bypassing the work of the Census Bureau and advocates. 

Second, the insertion of the question contradicted institutionalized census practices of vigorously testing and analyzing the effects of new questions. Although the Bureau had a practice of experimenting with different types of questions for the ACS, the decennial count was approached differently. The figures estimated in the decennial count were tied to resources, apportionment, and a variety of other representation issues, so it was unprecedented for the Bureau to insert a new question without testing it first. 

Third, the insertion of the citizenship question was surprising because census directors had long contended that such a question on the decennial form would compromise an accurate enumeration. In 1980, the Federation for American Immigration Reform brought a lawsuit, which was ultimately unsuccessful, against the Census Bureau, calling the enumeration effort unconstitutional because it did not have a citizenship question and did not produce an alternative way to identify “illegal” immigrants. The Bureau, however, contended that “Any effort to ascertain citizenship [through the census] will inevitably jeopardize the overall accuracy of the population count.” Moreover, census officials declared that “questions as to citizenship are particularly sensitive in minority communities and would inevitably trigger hostility, resentment[,] and refusal to cooperate.”31

Within days of the 2018 announcement, several Latinx civil rights advocacy groups as well as cities and states with large Latinx populations filed suits against Secretary Ross and the US Census Bureau. These suits were combined, with the New York and California lawsuits maintaining the highest profiles. Plaintiffs claimed that the new citizenship question represented animus against Latinxs and immigrants in general and was designed to produce a chilling effect on these individuals’ participation in the census. The plaintiffs contended that in the current era of anti-immigrant sentiment, such a question could lead noncitizens and their relatives to not report on Bureau documents. This consequence would have deleterious downstream effects on federal funds and other forms of representation within the Latinx community.32

In his March 2018 memo and in a written statement in June 2018, Ross maintained that the decision to institute a citizenship question had come at the behest of the Department of Justice (DOJ), which he claimed had asked for the question in order to better enforce the Voting Rights Act. However, the discovery process in the New York case (State of New York v. Department of Commerce, 2019) revealed that Ross had spoken with presidential advisor Steve Bannon and Kris Kobach, then secretary of state of Kansas, about adding the citizenship question long before he had received such a request from the DOJ.  Additionally, about a month before the Ross memo, the Trump Re-election Campaign had sent supporters an email asking them if they agreed with the Trump administration’s desire to have a citizenship question on the 2020 Census form.33 And soon after the Ross citizenship question memo had been published, the re-election team sent a second email citing the addition of the citizenship question to the upcoming census as an accomplishment of the Trump administration. This political maneuvering, plaintiffs argued, coupled with several anti-Latinx and anti-immigrant statements from the administration, constituted proof of racial animus.  

Kris Kobach, as Kansas secretary of state, along with other state officials in the South and Midwest, filed briefs supporting the addition of the citizenship question. They argued that states needed block-level citizenship data to protect individual voting rights and to ensure proper representation in Congress. They also rejected the notion that the question would produce a chilling effect on respondents, stating that no comprehensive study of the issue existed and that the implementation of the question on the Census Bureau’s American Community Survey (ACS) suggested that the Bureau itself regarded the citizenship question as valid. On May 21, the state of Alabama filed its own suit against the Trump administration to challenge the Bureau’s policy of including undocumented immigrants in the decennial count.  

Support for the plaintiffs in the combined cases was strong. Latinx advocates and immigrant rights groups across the country filed briefs contending that the Bureau’s decision to include the citizenship question was racially motivated and was designed to suppress the Latinx count. Five former directors of the US Census also filed briefs, stating that ACS and administrative data have long adequately supported Voting Rights Act (VRA) enforcement needs and that the citizenship question “seriously jeopardizes the accuracy of the [census] count.”34 Moreover, several academic organizations, including the American Statistical Association and the American Sociological Association, followed with similar statements of support for the plaintiffs.35 On January 15, 2019, Judge Jesse Furman of New York ruled that the Census Bureau could not move forward with the 2020 Census citizenship question because doing so would have a disparate impact on minority and immigrant communities. The California case was later decided in favor of the plaintiffs on March 6. Judge Richard Seeborg decided in that case that the citizenship question would have deleterious impact on Latinx communities and that Commerce Secretary Wilbur Ross had breached the Administrative Procedures Act by capriciously and arbitrarily adding the question.36 Latinx advocates hailed both decisions as an initial victory, but they also cautiously awaited the Supreme Court’s assessment. 

As the cases progressed, Latinx advocacy groups found themselves in a difficult situation. After fifty years of campaigning to make sure that Latinxs were fully counted, the addition of the new citizenship question threatened to undermine their efforts and credibility. Given the increased visibility of deportations in the Trump Era, Latinx advocacy groups would only be able to ensure a complete count if they could also assure Latinxs that their information would indeed remain confidential. Some groups continued their efforts as usual, noting the disastrous economic and political effects that a depressed count would have on their communities. In public these groups pointed to Title 13, which prohibits the Bureau from disclosing information for 72 years. Others took comfort in a 2010 memo from the DOJ stating that the Patriot Act cannot override the confidentiality of the census. 

Yet much uncertainty still remained. In May of 2018 Jimmy Gomez (D-CA) submitted a written question to the DOJ asking if it still agreed with the 2010 Patriot Act memo. Court filings now show that DOJ officials discussed how to answer Gomez’s question in a way that would leave the issue open for future debate.37 The email findings became part of the court documents filed in California. Moreover, many acknowledge that the citizenship question controversy had already inflicted much damage. The issue had already created a media stir within the Latinx community, and issues about the confidentiality of census responses, citizenship, and government trust had already been primed. Whether or not the courts ruled in favor of Latinx advocates, much had already been done to stoke suspicion about the census within the community. 

In June of 2019, the Supreme Court issued its opinion on the matter. In a ruling of 5-4 Justice Roberts ruled that the administration had insufficiently made an administrative procedure case for the question. They invited the administration to revise its arguments and bring the case back to the courts.38 Yet soon thereafter, the Trump administration held a press conference and stated that the administration would no longer pursue the citizenship question issue. 

The Trump administration placed decades of Latinx census efforts in peril. Advocates’ work to assist the Census Bureau in developing new question formats that would improve minority response rates has been disregarded. Moreover, the years of efforts to legitimize enumeration and convince communities to trust the Census Bureau have been jeopardized by the proposal of a new citizenship question, which stokes decades-old fears of government betrayal. 

Offhand, the efforts seem new. They represent actions taken at the administrative level and encourage a sharp break in Census Bureau protocol. However, upon deeper reflection, it is obvious that the Trump administration’s decisions reflect more historical racial politics that target data in an effort to depress minority rights. For decades, conservative administrations have meddled with census budgets and contested enumeration strategies in an effort to curb the minority count. The Trump administration’s actions concerning the combined race/ethnicity question showed that administrative appointees were also now willing to intervene in question wording and configuration. Moreover, as described previously, almost forty years ago,  right-wing anti-immigrant groups attempted to raise the issue of including a citizenship question in the courts, and it was dismissed. Such talk has been revived countless times through the efforts of Republican congressional leaders. Since the 1980s, conservative congressional leaders have repeatedly asked census directors to testify on the issue of including a citizenship question. And in each previous instance, census directors have rejected the idea, citing internal research and expert opinion.39 The Trump administration’s actions have followed these historic race politics. However, this time the pressure for a citizenship question came from within the administration itself. 

The efforts to attack data on minorities have a two-fold effect. First, they weaken the empirical evidence that minority groups have about their conditions. This situation, in turn, denies stakeholders the data that they need to mobilize on behalf of minority rights. By eliminating the combined question, which would have improved minority response rates, and by calling for a citizenship question, the Trump administration has weakened Latinx political power. Second, such census manipulations reinforce a particular image of the racial landscape of America. This is one in which the voices of racial minorities are not fully represented and in which their undercounting provides a less Black and Brown picture of the American landscape. As one Latinx census advocate, Representative Jimmy Gomez (D-CA), recently stated, “[The administration’s actions] threaten to erase [Latinx] from our country’s records.”40 Such tactics ultimately play into the national image of who is genuinely American and reinforce the second-class citizenship of Latinx and other minorities more generally.

  • 1Edward Fernandez, Using Analytic Techniques to Evaluate the 1990 Census Coverage of Young Hispanics (Washington, DC: US Bureau of the Census, 1995); US Census Bureau, Census Bureau Releases Estimate of Under- and Overcount in the 2010 Census (Suitland, MD: US Census Bureau, May 22, 2012).
  • 2G. Cristina Mora, Making Hispanics: How Activists, Bureaucrats, and Media Constructed a New American (Chicago: University of Chicago Press, 2014).
  • 3Michael Rodríguez-Muñiz, “Cultivating Consent: Nonstate Leaders and the Orchestration of State Legibility,” American Journal of Sociology 123 (2017): 385–425.
  • 4See National Association of Latino Elected Officials (NALEO), A Roadmap for Census 2020: Securing a Fair and Accurate Count of Latinos in California (Los Angeles: National Association of Latino Elected Officials, 2018).
  • 5“Governor Brown Makes a Deal on the State Budget. It’s Pretty Good” (Editorial), Sacramento Bee, June 12, 2018,; Nick Brown, “Worried About Undercount, States and Cities Spend to Promote 2020 Census, Reuters, Oct. 1, 2018.
  • 6Brown, “Worried About Undercount.”
  • 7See Ibid.
  • 8Julie Dowling, Mexican Americans and the Question of Race (Austin: University of Texas Press, 2014).
  • 9Senator William James Harris, cited in Federal Register, p. 1176, Dec. 15, 1928.
  • 10Francisco Balderrama and Raymond Rodriguez, Decade of Betrayal: Mexican Repatriation in the 1930s (Albuquerque: University of New Mexico Press, 2006).
  • 11Dowling, Mexican Americans and the Question.
  • 12Rodolfo Acuna, Occupied America: A History of Chicanos (New York: Pearson, 2014); Cynthia Orozco, No Dogs, Women, or Mexicans Allowed: The Rise of the Mexican American Civil Rights Movement (Austin: University of Texas Press, 2009).
  • 13Mora, Making Hispanics.
  • 14Ibid.
  • 15Arlene Davila, Latinos, Inc: The Making and Marketing of a People (Berkeley: University of California Press, 2001).
  • 16Mora, Making Hispanics, 83-118.
  • 17Ibid.
  • 18Rodríguez-Muñiz, “Cultivating Consent,” 385-425.
  • 19Kelly Mathews et al.,, 2015 National Content Test: Race and Ethnicity Analysis Report (Washington, DC: US Bureau of the Census, 2017) p. 9.
  • 20Ibid., p. ix.
  • 21Rodríguez-Muñiz, “Cultivating Consent.”
  • 22Clara Rodriguez, Changing Race: Latinos, the Census, and the History of Ethnicity (New York: New York University Press, 2000).
  • 23Mathews et al., 2015 National Content Test, p. xii.
  • 24See Grace Guarnieri, “2020 Census Will Ask White People About Origins but Leave Out Questions About Hispanic and Middle Eastern Identities,” Newsweek, Feb. 1, 2018,….
  • 25Hansi Lo Wang, “2020 Census to Keep Racial, Ethnic Categories Used in 2010,” National Public Radio, Jan. 26, 2018,….
  • 26Ibid.
  • 27See Margo Anderson and Stephen Feinberg, Who Counts?: The Politics of Census Taking in Contemporary America (New York: Russell Sage, 2006).
  • 2828. Ward Connerly and Mike Gonzalez, “It’s Time for the Census Bureau to Stop Dividing America,” Spokesman-Review (WA), Feb. 28, 2019; Mike Gonzalez, The Division for the Consequences of the Census Bureau’s Committee on Race (Washington, DC: Heritage Foundation, 2018).
  • 29Rodriguez, Changing Race.
  • 30Frederick Bohme, Twenty Censuses: Population and Housing Questions, 1790–1980 (Washington, DC: US Bureau of the Census, 1979).
  • 31See Federation of American Immigration Reform v. Klutznick, 486 F. Supp. 568 (DDC 1980).
  • 32See New York State v. Department of Commerce, No. 18-CV-2921 (SDNY 2019).
  • 33Tara Bahrampour, “Trump’s Re-election Campaign Calls for Adding Citizenship Question to 2020 Census amid Criticism He is Politicizing the Count,” Washington Post, Mar. 20, 2018.
  • 34See Brief of Amici Curaie, Former Census Bureau Directors in Support of Plaintiff’s Trial Position, State of New York v. US Department of Commerce, No. 1:18CV-2921 (JMF) (SDNY 2019), p. 1.
  • 35Mathew Snipp, “ASA Files Amicus Brief Objecting to Citizenship Question in the 2020 Census,” Footnotes 46, no. 5 (2018)
  • 36See State of California v. Wilbur Ross et al., No. 18-CV-01865-RS (N.D. CA 2018).
  • 37Hansi Lo Wang, “DOJ, Asked About Census Confidentiality, Crafted Intentionally Vague Answer,” National Public Radio, Nov. 19, 2018,….
  • 38Department of Commerce et. al v. New York et. Al 588 U.S. (2019)
  • 39See State of New York No. 1:18-CV-2921 (JMF).
  • 40Congressman Jimmy Gomez, quoted in “Groundswell of Opposition Tells Commerce Department to Scrap the Citizenship Question on 2020 Census,” Common Cause, 2018 groundswell-opposition-tells-commerce-department-scrap-citizenship-question-2020