Measuring the Impact of Political Ideology on the Adoption of English-Only Laws in the United States

Dublin Core

Title

Measuring the Impact of Political Ideology on the Adoption of English-Only Laws in the United States

Description

This study tackles whether political ideology influences the adoption of English-Only laws in the United States'-and, if so, to what degree? To answer this question, we collected data from across various U.S. states that we then analyzed, using logistic regression, complemented with a geographical information system (GIS) survey of the states. This Article breaks down into four parts. We first review the still-sparse literature on the topic. Next, we define and operationalize (i.e., establish indicators of measurement) key terms in the study. The third section specifies the hypotheses and assumptions underlying those hypotheses. The fourth presents our results and interpretations.

Creator

Philip C. Aka, Lucinda M. Deason, Augustine Hammond

Publisher

The Scholar: St. Mary's Law Review on Race and Social Justice, St. Mary's University School of Law San Antonio Texas, St. Mary's University School of Law, Sarita Kenedy Law Library

Date

2010

Rights

Copyright to The Scholar: St. Mary's Law Review on Race and Social Justice

Relation

The Scholar: St. Mary's Law Review on Race and Social Justice

Format

RFC3778

Language

English, en-US

Type

Text

Identifier

STMU_TheScholarStMarysLRev_v13i1p0001_Aka_Deason_Hammond

PDF Search

Text

ARTICLE

MEASURING THE IMPACT OF POLITICAL IDEOLOGY ON THE ADOPTION OF ENGLISH-ONLY LAWS IN THE UNITED STATES DR. PHILIP C. AKA* DR. LUCINDA M. DEASON** DR. AUGUSTINE HAMMOND***

I. Introduction and Purpose of Study ..................... II. Review of Literature on This Topic ..................... III. Definitions and Operationalization of Key Terms..........

2 5 9

* Professor of Political Science, Chicago State University; Adjunct Professor of Law, Indiana University School of Law at Indianapolis. B.A. (magna cum laude), Edinboro University of Pennsylvania; M.A., University of North Texas; J.D., Temple University; LL.M. (summa cum laude), Indiana University at Indianapolis; Ph.D., Howard University. Dr. Aka has published extensively on issues related to human rights. His recent publications include Politics and Economic Development in Africa: Incorporating the Influence of the Movement for Popular Participationon Assessments of NEPAD, Hum. RTS. & GLOBALIZATION L. REV. (forthcoming Summer 2010). This Article benefitted from the data collection assistance of a number of Dr. Deason's graduate students. We particularly acknowledge the diligent energies of Elijah Agyapong and William OpokuAgyeman. We also gratefully acknowledge the outstanding editorial assistance of The Scholar: St. Mary's Law Review on Minority Issues editorial board. ** Associate Professor of Political Science, University of Akron. B.S., Wayne State University; MPA, University of Michigan-Dearborn; Ph.D. Michigan State University. Dr. Deason has published on issues related to cultural competence, English-Only laws, underserved populations, and nursing home regulations. Her recent publications include Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53 (2009) (with Philip C. Aka). *** Assistant Professor of Political Science, Augusta State University. B.Ed., University of Cape Coast; MPA, University of Ghana and University of Akron; Ph.D., University of Akron. Dr. Hammond has published on issues related to drug abuse prevention, and nursing home regulations. His recent publications include Do Adolescents Perceive Police Officers as Credible Instructors of Substance Abuse Prevention Programs?, 23 HEALTH EDUc. REs. 682 (2008) (coauthored with six others).

1

2

THE SCHOLAR

[Vol. 13:1

IV. Hypotheses and Underlying Assumptions ................ A. Political Party Affiliation Variables................. B. Additional Assumption Concerning the Political Party Affiliation Variables.... ......................... ................................. C. Foreign-Born D. Other Languages............................... E. Region. ...................................... V. Results and Interpretations... ...................... VI. Conclusion ......................................
I.
INTRODUCTION AND PURPOSE OF STUDY

11 12 13 14 14 14 15 22

The central question this study tackles is whether political ideology influences the adoption of English-Only laws in the United States'-and, if so, to what degree? To answer this question, we collected data from across various U.S. states that we then analyzed, using logistic regression, complemented with a geographical information system (GIS) survey of the states. 2 Altogether, we analyzed a sample size of 857 cases covering the period from 1990 to 2007.' This piece is an offshoot of an extensive study on the effect of EnglishOnly laws on culturally competent public services by two of the authors.4 Specifically, that broader study highlighted the constraints English-Only measures pose for delivery of public services to limited English proficiency (LEP) populations in the United States' and "devise[d] a solution to those barriers centered around the application of cultural compe-

1. Definition and measurement of this and other key terms in the study are saved for Part III of this Article. 2. For more on the tools of measurement used in this study, see Parts III (dealing with definitions and operationalization of key terms) and V (results and interpretations) of this Article. 3. See Part V of this Article (results and interpretations) for more details. 4. See generally Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53 (2009). 5. See generally id. LEP persons are individuals with a "low level of skill in comprehending, speaking, reading or writing the English language because of being from an environment in which another language is spoken." Ruiz v. Hull, 957 P.2d 984, 990 n.5 (Ariz. 1998) (en banc). The Department of Justice observed that, "[diespite efforts to learn and master English," the English language proficiency of LEP persons "may be limited for some time." National Origin Discrimination Against Persons with Limited English Proficiency (LEP Guidance), 65 Fed. Reg. 50123, 50124 (Aug. 16, 2000). LEP communities include not only foreign-born persons (discussed in Parts IV and V of this Article), but also encompass some persons born within the United States, such as Native Americans. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L. 53, 71 (2009) (finding the LEP population of the United States to be substantial when compared to the overall population of the country).

2010]

ENGLISH-ONLY LAWS

3

tency."6 In developing our argument in the larger study, one of the issues we examined was the possible role of politics in the evolution of EnglishOnly laws.' We surmised that "[p]olitical ideology tends to play a major role in recent debates on and adoption of English-Only laws"' and, among other things, analyzed the attitudes of recent U.S. presidents toward English-Only laws, 9 and the constitutional amendment incorporating an English-Only measure approved in November of 2008 in Missouri.' 0 The possible relationship between political ideology and adoption of English-Only laws was also among various topics that, in commenting on possible avenues for future research into culturally competent public services and English-Only laws, we marked out as "[u]nfinished businesses."' We recommended "a more systematic examination regarding the influence of political ideology on the adoption of English-Only

6. Phillip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 128 (2009). Cultural competency "focuses on the distinctiveness of an individual or his or her group-and public services tailored to such individuality." Id. at 120. Culturally competent public services "take due account of the different cultural compositions of [recipients of such services], with positive outcome for all concerned." Id. at 112 (footnote omitted). Impetuses for culturally sensitive public services include growing racial and ethnic diversity, as well as demographic changes related to the growth of the U.S. population. See id. at 112-17. Moreover, as we argued in our larger piece, culturally competent public services do not run afoul of the tenets of the Weberian concept of bureaucracy, grounded in impersonal features, including "equal treatment for all who deal with the organization." Id. at 120 (quoting MICHAEL E. MILAKOVICH & GEORGE J. GORDON, PUBLIC ADMINISTRATION IN AMERICA 158 (10th ed. 2009)). 7. Id. at 90-94 (questioning whether a connection exists between politics and EnglishOnly laws and how they have changed over time). 8. Id. at 90 (2009) (footnote omitted) (according to James Crawford, politicians are one of numerous groups that facilitated the transitioning of the English-Only movement from a mere periphery organization to something more formidable). 9. Id. at 91-93 (detailing the viewpoints of the United States' most recent presidents on English-Only laws). From Ronald Reagan to Barack Obama, the attitude of presidents has vacillated in tune with the change of political parties occupying the White House. Id. 10. Mo. CONsT. art. I, § 34; Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 93-94 (2009) (demostrating overwhelming support for the movement when the amendment passed with an 86.3% majority). The amendment established English as the official language of governmental communications, requiring "all official proceedings in the state be conducted in English alone." Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and EnglishOnly Laws, 53 How. L.J. 53, 93 (2000). The role of political ideology in the adoption of the constitutional amendment in Missouri became apparent when considered in light of a post-election poll finding that 57% of Missouri Democrats favored the amendment while an overwhelming 96% of Missouri Republicans supported the measure. Id. 11. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 131 (2009) (recommending six different topics that are ripe for examination to better ascertain the true impact of English-Only laws).

4

THE SCHOLAR

[Vol. 13:1

laws,"1 reasoning that, "[a]lthough a narrative (or qualitative) analysis of the type undertaken [in our previous study] may be useful, the recommended mode is a statistical study composed of a larger sample [size] than the survey we used" in the article. 1 3 "Doctor, heal thyself," is a common adage in our culture. By this Article, we take our own counsel by seeking to implement the systematic, statistical analysis we recommended in our larger study. Findings from this study demonstrate that, based on the data we examined, political ideology has an impact on the adoption of English-Only laws at a level that is generally statistically significant.14 All the nine hypotheses we tested in this study showed a relationship between political ideology and adoption of English-Only laws." Of the nine, five were also statistically significant, although one of these moved in the opposite direction. 16 Our geographical information system (GIS) analysis also importantly converged with these statistical findings." Among other things, the GIS analysis showed that, regionally, the South has the most states with English-Only laws.1 " A revealing finding from the study is the interaction effect arising from the mixture of being foreign-born and speaking languages other than English.19 Specifically, the result statistically confirmed that states with a higher than average percentage of foreign-born residents who speak languages other than English are more likely to adopt English-Only laws, compared to states without these features.20 Another revealing discovery is that full Republican control of the government, embodied in control of the governorship and the two chambers of a state legislature, is necessary for passage of English-Only laws, whereas

12. Id. (suggesting a study of the connection between political ideology and the rise of English-Only laws as one way to better understand English-Only laws and their many facets). 13. Id. (determining a different methodology would be preferable in the examination of any influence that political ideology might have on English-Only laws and their adoption). 14. See Table 3 in the appendix containing our logistic regression. See also Table 2 in the appendix providing a description of variables used in our study, including statistical information, such as the mean, standard deviation, range, and sample sizes tied to these variables. 15. See Part V of this Article (containing our results and interpretations as well as the recapitulation of our hypotheses and results). 16. See Part V of this Article. The hypothesis that moved in the opposite direction is Hypothesis Number 7. 17. See Figure 1 in the appendix. 18. See Figure 1. 19. See discussion in Part V of this Article (results and interpretations) relating to Hypothesis Number 8. 20. See discussion in Part V of this Article (results and interpretations) relating to Hypothesis Number 8.

2010]1

ENGLISH-ONLY LAWS

5

departure from full control, beginning with even the least emblem of reduced control, like a legislature whose control is split between the Republican and Democratic parties, 21 decreases the likelihood of an adoption of such laws. This Article breaks down into four main parts. Following this introduction, we first review the still-sparse literature on the topic. Next, we define and operationalize (i.e., establish indicators of measurement) key terms in the study. The third major section specifies the hypotheses of this study and the assumptions underlying those hypotheses. The fourth presents our results and interpretations.
II.
REVIEW OF LITERATURE ON THIs

Topic

The practice in social science research, especially empirical studies, such as we undertake here, is to perform a literature review that will form the basis for generation and exploration of hypotheses.22 However, to the best of our knowledge, there have been little, if any, works published specifically on the role of political beliefs in the adoption of English-Only laws-and, therefore, little prior research to anchor our own scholarship. Instead, this study is the first of its kind to systematically assess the impact of political ideology on English-Only laws. In the absence of previous qualitative-narrative or statistical research on the topic, we are left with little choice than to fall back on the literature that we generated in our larger study on this topic, referred to before in this piece. Fortunately, our literature review also encompasses materials germane to this topic outside our larger study, such as the sources relating to executive leadership and southern politics that we cited in discussing the bases for the assumptions supporting our hypotheses in Part IV. Back again to our larger study, there we indicated, "[o]f the two major political parties of the U.S., the one that tends to be associated with advocacy for and adoption of English-Only initiatives has been the Republican Party."2 3 We explained the progression of English-Only sentiments and movement towards implementing English-Only laws in modern times occurred largely in the 1980s while Republicans held the reins of national powers and leaders of the movement tended to be affiliated with the Re21. Explanation of "split" control is provided in Part III of this Article. 22. See CHRIS HART, DOING A LITERATURE REVIEw: RELEASING THE SOCIAL SCIENCE RESEARCH IMAGINATION 1 (1st ed. 1998). The purpose of a literature review is to gain an understanding about the topic and see what research has already been conducted on the subject matter. Id. 23. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 90 (2009) (explaining that individuals who identify themselves as Republican tend to support the adoption of English-Only laws in larger numbers than people who identify with the Democratic party).

6

THE SCHOLAR

[Vol. 13:1

publican Party, including the late Senator Samuel I. Hayakawa.2 4 Still on this movement, all of the six English-Only measures we analyzed in our larger piece, including Senator Hayakawa's "English Language Amendment" (ELA), were sponsored by lawmakers affiliated with the Republican Party.25 Additional to the ELA, for the Senate, there were Bill 356, the "Language of Government Act," sponsored by Senator Richard C. Shelby (Republican from Alabama) and twenty-two others, introduced into the Senate on February 3, 1995; and Bill 1335, the "S.I. Hayakawa Official English Language Act," sponsored by Senator James Inhofe (Republican from Oklahoma) and nine others, introduced into the Senate on May 8, 2007.26 In the House of Representatives, the measures in question were the "Bill Emerson English Language Empowerment Act," sponsored by Representative Randall Cunningham (Republican from California) and 167 others, introduced into the House in 1997; H.R. 769, the "National Language Act," sponsored by Representative Peter King (Republican from New York) and sixty-seven others, introduced into the House on January 31, 2007; and H.R. 997, the "English Unity Act," sponsored by Representative Steve King (Republican from Iowa) and twentyfive others, introduced into the House on February 12, 2007.27 Similarly, at the sub-national levels, sponsors of English-Only policies, as well as individuals who lead interest groups, that advocate such policies, such as ProEnglish, tend to be affiliated to the Republican Party.

24. Id. at 90-91 (illustrating the clear divide between Republicans and Democrats in their assessment of and advocacy, or lack thereof, for English-Only laws). 25. Id. at 61-64. Of the six English-only measures, three were proposed in the Senate, while the other three were sponsored in the House of Representatives. Id. As we indicated in our larger piece, the six initiatives illustrate, rather than exhaust the numerous English-Only measures from that period, none of which we traced to the Democratic Party. Id. at 60-61. 26. Id. at 62-63 (identifying other Republican Senators responsible for sponsoring other English-Only legislation). 27. Id. at 63-64 (listing the sponsoring Republicans associated with introducing English-Only legislation in the House of Representatives). The "Bill Emerson English Language Empowerment Act" was the House version of Senate Bill 356, the "Language of Government Act," while the "National Language Act" was the House version of Senate Bill 1335, named after Senator Hayakawa. Id. at 64. 28. For a measure at the local level, see Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 87 (2009), commenting on a measure in Davidson County and the Nashville metropolitan government, all in Tennessee. Regarding interest groups, see id. (pointing out that "various individuals" on the board of advisors of ProEnglish "identify themselves as Republicans," and none as Democrats). ProEnglish is based in Arlington, Virginia. ProEnglish, Who We Are: All About ProEnglish, http://www.proenglish.org/main/gen-info.htm (last visited June 9, 2010). The organization proclaims that it "work[s] through the courts and in the court of public opinion to defend English's historic role as America's common, unifying language, and to persuade lawmakers to adopt English as the official language at all levels of government."

2010]1

ENGLISH-ONLY LAWS

7

We next examined the attitudes of recent U.S. presidents toward English-Only laws, and analyzed the adoption of an English-Only measure passed in Missouri on November 4, 2008.29 Regarding the attitudes of U.S. presidents to English-Only measures, our finding was that the English-Only movement gained vitality under the watch of a succession of Republican administrations 3o and that, at best, Republican Party presidents, such as George W. Bush, displayed an orientation of "antipathy toward English-Only measures."" In contrast, "[u]nder President William Jefferson Clinton, a Democrat, there was an opposition or at least an attempt to minimize the worst effects of English-Only laws. Clinton signed Executive Order No. 13166, which sought to improve access to [public] services for persons with limited English proficiency." 3 2 What is more, Clinton signed the order in the aftermath of his administration's "strong opposition to the Language of Government Act of 1995, proposed in the Senate of the 104th Congress, which the Republican Party controlled."3 3 Following our review of the attitude of recent U.S. presidents toward English-Only laws, we concluded: "It must be this tendency of Republican administrations to advocate for or support English-Only programs, and that of Democratic governments to oppose them, which led ProEnglish to conclude, as it did, that the U.S. government, under President Barack Obama, will 'certain[ly]' oppose official English."3 4 A conceivable objection to the foregoing literature review is that it speaks to English-Only initiatives at the national level, whereas, as we indicate below, the level of analysis that, for reasons of data constraint, we chose for this study is the state.3 5 However, such an objection is rebutted and overcome by the fact that English-Only initiatives at the national level, beginning

ProEnglish, Who We Are: All About ProEnglish, http://www.proenglish.org/main/gen-info. htm (last visited June 9, 2010). 29. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 91-94 (2009) (reviewing the previous areas of inquiry in an effort to better understand English-Only laws). 30. Id. at 91-92 (arguing that under Ronald Reagan, English-Only laws were given preference in their cause to establish English as the de jure language of the United States). 31. Id. at 92. The conclusion that President George W. Bush's position concerning English-Only laws was more moderate derived from the former President's indication that he was not in favor of the laws, even though he never took advantage of his right as Commander in Chief to repeal Executive Order No. 13166. Id. at 92-93. 32. Id. at 92. 33. Id. (footnote omitted) (emphasizing the fact that President Clinton chose to employ a non-legislative route for his remedy which conspicuously does not require Congressional approval or participation of any kind). 34. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 93 (2009). 35. See Part III of this Article.

8

THE SCHOLAR

[Vol. 13:1

from the 1980s, impacted adoption of English-Only measures at the subnational levels, particularly the states. As we pointed out in our larger study, "[m]ost of the [state] adoptions of English-Only laws occurred after 1981." 36 Prior to 1981, merely four states-Hawaii, Illinois, Louisiana, and Nebraska-had English-Only measures, or just three, excluding Hawaii, which is officially bilingual. Regarding the adoption on November 4, 2008, of an English-Only measure in Missouri that we analyzed in our larger study, we made the following observations. The measure, a constitutional amendment identifying English as the official idiom of the state government, was overwhelmingly ratified when 86.3% of the voters in Missouri voted to adopt the constitutional amendment.3 1 It "mandated that all official proceedings in the state be conducted in English alone and that English must be used at 'all governmental meetings, at which any public business is 3 discussed, decided, or public policy is formulated.' "" The results of an online survey on voters' attitudes conducted following the election4 0 appear to validate our hunch on the tendency of Republicans to favor English-Only laws and the tendency of Democrats to oppose or counter those measures. The survey was conducted on November 7, 2008, and involved 808 voters. 41 It consisted of two questions: (1) "[W]hether the voters voted for or against English as the official language in Missouri;" and (2) whether the voters "would support or oppose adoption of English as the official language in the U.S."4 2 "[Ninety-six percent] of Republicans, compared to 57% of Democrats, said they voted for the amendment
... [and] 96% of Republicans, compared to 59% of Democrats, indicated

they would vote to support making English the official language of the U.S."4 3 In summation, on both questions, an overwhelming number of individuals who identified themselves as Republicans, compared to those who identified themselves as members of the Democratic Party, favored English-Only laws.

36. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 66 (2009). 37. Id. (identifying the number of English-Only laws in the several states before 1981). 38. Id. at 93. 39. Id. 40. Id. The survey was conducted by Zogby International, a U.S. polling agency. Id. For more on this organization, see Zogby International, About Us, http://www.zogby.com/ about (last visited June 10, 2010). 41. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 93 (2009). The survey had a margin of error of 3.5%. Id. 42. Id. 43. Id. at 93-94 (footnote omitted).

2010]

ENGLISH-ONLY LAWS

9

III. DEFINITIONS AND OPERATIONALIZATION OF KEY TERMS
Our dependent variable (or matter under examination) in this study is English-Only laws, and the independent variables (hypothesized to affect the dependent variable) are political ideology, foreign-born persons, speaking other languages, and region. One of our major independent variables is political ideology. English-Only laws are defined as "measures, such as ordinances, statutes, constitutional amendments, policies, rules, and regulations, that mandate people to 'speak English only' by banning or restricting the use of other languages by government agencies and, ... private businesses, as well."4 4 Here, as in our larger piece, our definition of English-Only laws embraces measures others might assess as merely symbolic.4 5 The term is coded in this study as a dichotomous variable operationalized in terms of whether or not a state has adopted such law. Our larger study analyzed the manifestations of English-Only measures at all three levels of the U.S. political system, namely, the national, state, and local. 46 However, for reasons of manageability and access to data, in this study, our focus is limited to states. Staying with this level of analysis, 47 thirty states have adopted English-Only laws. Table 1 in this Article identifies these states, along with details relating to their adoptions. The data is complemented with a GIS mapping of the states (Figure 1) with pertinent details, such as English-Only status, political party affiliation, and census region. The major independent variable in this study, political ideology, is defined as "a consistent set of values and beliefs about the proper purpose and scope of government." 48 Ideology is a property that elites and the masses alike possess, 4 9 and our larger study covered these two entities. Although much of our analysis dwelt on "politicians" (political elites),

44. Id. at 57 (defining the term English-Only laws for purposes of this examination and the previous one). 45. Id. at 68-69 (asserting that some entities including the Arizona Supreme Court and Professor Baron believe English-Only laws to be symbolic). 46. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 60-68 (2009) (examining English-Only laws in the
previous study at the three different levels of government).

47. For an elaboration of this concept in political science, see BRUCE RUSSETr ET AL., WORLD POLITIcs: THE MENU FOR CHOICE 10-16 (6th ed. 2000). 48. KENNETH JANDA ET AL., THE CHALLENGE OF DEMOCRACY: AMERICAN GOVERNMENT IN A GLOBAL WORLD 20 (10th ed. 2009). 49. See generally MICHAEL J. SODARO ET AL., COMPARATIVE POLITICS: A GLOBAL INTRODUCTION 98-123 (3d ed. 2008) (commenting on "power," specifically "on the various ways power can be used within countries," rather than political ideology as such) (emphasis added). Elites, particularlypoliticalelites, are "people who have prominent positions either

in government or in nongovernmental organizations and professions that have a real effect on government actions." Id. at 99. In contrast, masses, often denoting a variegated popu-

10

THE SCHOLAR

[Vol. 13:1

including recent presidents, we also covered ideology among mass publics within the context of the online survey relating to the adoption of an English-Only measure in Missouri in 2008.5o Here, as in our larger study, ideology is an attribute that we ascribe both to elites and the masses, a quality that, in our assessment, both leaders and their followers share, even if elites possess the property in more elevated levels than non-elites. Some critics will challenge our assessment that the quality applies to the masses in this case, given that the dependent variable, that forms the focus in this study, English-Only laws, is the legislative handiwork of elites. However, because in our governmental system political leaders represent their constituencies and, hopefully, in the best traditions of democratic government, the will of their people, we make no sharp distinction concerning possession of this quality between elites and mass publics. Political ideology is what, for lack of a better term, we denominate a politicalparty affiliation variable (PPAV). It is a dummy variable created from the political party affiliation of state officials. PPAV is operationalized in terms of whether or not a governor belonged to the Democratic or Republican Party and whether one of these two major parties has a majority in the two chambers of the legislature of an affected state ("unified" in favor of the majority party) or split (something which occurs when neither major party has majority).5 ' There are six possible combinations or categories: (1) Democratic governor and a unified Democratic legislature, (2) Democratic governor and a unified Republican legislature, (3) Democratic governor and a split (non-unified) legislature, 52 (4)

lation divided on ethnicity, religion, class or other lines, "refers broadly to the rest of the population" beyond or besides elites. Id. at 100. 50. See Mo. CONST. art. I, § 34. 51. The tie of PPAV to the Democratic and Republican parties raises the question whether these two private organizations, between themselves, exhaust the rank of political parties in the U.S. political system. The answer is that the United States has a strong twoparty system where, more than is the case in other two-party systems, these two major political parties dominate the political and electoral landscape. See KENNETH JANDA ET
AL., THE CHALLENGE OF DEMOCRACY: AMERICAN GOVERNMENT IN A GLOBAL WORLD

227, 229 (10th ed. 2009); see also THOMAs E. PATTERSON, THE AMERICAN DEMOCRACY

199 (9th ed. 2009) (stating that the Democratic and Republican parties today are the only two parties "with a realistic chance of acquiring political control" in the United States). Although many Americans identify themselves as "independents" (or disinclined from either of these two parties) and numerous parties have evolved over the years that tap into this "independent" mood, these third parties, through their platforms or postures, maintain some degree of affiliation to one of the two major parties. See Minnesota DemocraticFarmer-Labor Party, Introduction to the DFL Party, http://dfl.org/introduction (last visited June 10, 2010). 52. This occurs within the context of a bicameral legislature where either the Democratic or Republican Party control one chamber of the legislature, while the other party controls the other.

2010]

ENGLISH-ONLY LAWS

11

Republican governor and a unified Republican legislature, (5) Republican governor and a unified Democratic legislature, and (6) Republican governor and a split (non-unified) legislature. Consistent with the practice in statistical research, only five of these categories were directly investigated, while the sixth combination, category Number 4, was used as a comparison with the other five categories. Other variables we determined to influence the adoption of EnglishOnly laws are (being) foreign-born, (speaking) other language(s), and (census) region.5 4 Foreign-bornis measured by the percentage of foreignborn persons in the state, including naturalized U.S. citizens and non-U.S. citizens. Other language is measured by the percentage of persons five years and above who speak languages other than English. Region is measured by dummy variables created from the four population regions of the United States Census Bureau: Northeast, Midwest, South, and West." The area that, for this variable, we seclude as comparison category is the Southern region. The data for the PPAV came from the World Almanac and Book of Facts," supported with materials from the National Governors Association," while the ones for foreign-born, other language, and region, were taken from the U.S. Census Bureau.5 8
IV.
HYPOTHESES AND UNDERLYING AssumerloNs

Based on the measurement variables specified above, we developed the following nine hypotheses for this study.

53. See DAMODAR N. GUJARATI, ESSENTIALS OF ECONOMETRICS 257 (1992) (advising that where a dummy variable has "in" categories, the analyst should introduce only (m-1) dummy variables). 54. Our general sense is that, together, these terms influence the adoption of EnglishOnly laws. Particularly, the first two are directly connected to LEP communities that have formed a point of focus in our general research on English-Only laws, including this study. 55. See WordlQ, United States Census Bureau-Definition, http://www.wordiq.com/ definition/UnitedStates CensusBureau (last visited June 10, 2010); U.S. Census Bureau, About Us, http://www.census.gov/aboutus/ (last visited June 10, 2010). 56. WORLD ALMANAC BOOKS, THE WORLD ALMANAC AND BOOK OF FACTS (C. ALAN JOYCE & SARAH JANSSEN EDS., WORLD ALMANAC, 2009). We obtained the political party affiliation of the governor and the majority party in each chamber of the state legislature from the corresponding editions of the World Almanac and Book of Facts. 57. National Governors Association, http://www.nga.org (last visited June 10, 2010). We used this website to corroborate the political party affiliation data for the governor of each state with the information in the World Almanac and Book of Facts. 58. See, e.g., U.S. Census Bureau, American Community Survey, http://www.census. gov/acs/ (last visited June 10, 2010); U.S. CENSUS BUREAU, PROFILE OF THE FOREIGNBORN POPULATION IN THE UNITED STATES: 2000 (2001), available at http://www.census. gov/prod/2002pubs/p23-206.pdf.

12

THE SCHOLAR

[Vol. 13:1

A.

Political Party Affiliation Variables

1. States with a Democratic governor and a legislature controlled by or unified in favor of the Democratic Party are less likely to adopt EnglishOnly laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. Several assumptions undergird this hypothesis. The first is that all the EnglishOnly bills recounted in this Article that have been introduced in the U.S. Congress (the Senate and the House of Representatives), including the "English Language Amendment" proposed by Senator Hayakawa, were sponsored or spearheaded by lawmakers affiliated to the Republican Party.59 None of these measures, to the best of our knowledge, was spearheaded by a lawmaker affiliated with the Democratic Party. Second, related to the first point, of the United States' two major political parties, "the one that tends to be associated with advocacy for and adoption of English-Only initiatives has been the Republican Party." 6 0 Thirdly, tied to the first two points, there has been a tendency at the national level for Republican administrations "to advocate for or support English-Only programs, and that of Democratic governments to oppose" or counter those measures. 6 ' 2. States with a Democratic governor and a legislature controlled by or unified in favor of the Republican Party (governments whose control is divided between the Democratic and Republican Parties) are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. 3. States with a Democratic governor and a legislature whose control is split, in the sense that the majority party in one chamber is the Democratic Party while the majority party in the other chamber is the Republican Party, are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. 4. States with a Republican governor and a legislature controlled by or unified in favor of the Democratic Party (i.e., governments whose control is divided between the two major political parties) are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. 5. States with a Republican governor and a legislature whose control is split, in the sense that the majority party in one chamber is the Republi59. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 60-64, 87, 90-91 (2009). 60. Id. at 90. 61. Id. at 93.

2010]

ENGLISH-ONLY LAWS

13

can Party while the majority party in the other chamber is the Democratic Party, are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. B. Additional Assumption Concerning the Political Party Affiliation Variables With respect to the political party affiliation variables delineated above, we make a general assumption, not in our literature review, that is tied to our supposition above about the tendency of Republican governments to adopt English-Only laws and the disinclinations of Democratic governments toward such laws. 6 2 The assumption is that executive leadership, of the type a state governor embodies, matters.6 3 As Professors Milakovich and Gordon elaborate, within the limited context of bureaucratic leadership, but arguably applicable here, too, "[t]he quality and style of leadership practiced by" chief executives, "are key factors in how public agencies perform their duties and achieve their goals" partly because chief executives, governors included, "have increasingly been regarded as logical choices for the task of maintaining some measure of operational control and accountability within their administrative establishments." 6 4 In a nutshell, we assume that the political party affiliation of a governor makes a difference in whether or not a state will adopt English-Only laws, with Republican governors tending to support such measures while governors belonging to the Democratic Party tend to oppose them. The basis for this assumption is the reduced synergy, nay dissonance, arising from a divided government, half-Republican, half-Democrat, compared to the "ideal" scenario of a fully Republican government of our comparison category. The departure from full Republican control, necessary to facilitate the adoption of English-Only laws, increases in situations where, additional to Democratic control of the governorship, legislative control in a state is split between the Democratic and Republican Parties.

62. Id. at 93. 63. See, e.g., MICHAEL E. MILAKOVICH & GEORGE J. GORDON, PUBLIC ADMINISTRATION IN AMERICA 253-56 (10th ed. 2009).
64. Id. at 253-54. Milakovich and Gordon importantly observed that "[t]o a great extent, electoral outcomes-[including] whether elected executives remain in office-are determined by public perceptions of these leadership duties and how they are handled."

Id. at 254.

14

THE SCHOLAR

[Vol. 13:1

C.

Foreign-Born

6. States with higher than average percentage of foreign-born persons are more likely to adopt English-Only laws, compared to states with lower than average percentage of foreign-born individuals. 65 The assumption underlying this hypothesis came from our findings in reviewing the history and evolution of English-Only laws in our larger study.66 "[A]nxiety and expression of xenophobia followed the passage of the 1965 immigration [reform] law" that increased the quota of immigrants from non-European countries. More specifically, many nativists in the past perceived increased migration to the United States, together with the new languages the migrants bring with them, as an assault on their way of life-and the preeminent status of English-and have often responded with the passage of English-Only measures.68 D. Other Languages

7. States with a higher than average percentage of residents who speak languages other than English are more likely to adopt English-Only laws, compared to states with a lower than average percentage of residents who speak languages other than English.69 The assumption underlying this hypothesis is the same as the one in Hypothesis Number 6 above regarding foreign-born residents. 8. States with a higher than average percentage of persons, who speak languages other than English and are foreign-born, are more likely to adopt English-Only laws, compared to states with a lower than average percentage of persons who speak languages other than English and are foreign-born. E. Region

9. States in the Southern region are more likely to adopt English-Only laws, compared to states in other regions, such as the Midwest, Northeast, and West. The assumption undergirding this hypothesis is two-fold. The
65. The percentage of U.S. residents who are foreign-born is indicated in Table 2 at the appendix of this Article. 66. See Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 74-80 (2009) (tracing English-Only laws from their origins, including comments indicating that John Adams, John Jay, and Benjamin Franklin had affirmative English-Only inclinations). 67. Id. at 79. 68. See id. at 77-80. Moreover, anti-immigrant sentiments have sometimes been mixed with racism in the United States. Id. at 85-86. 69. The percentage of U.S. residents five years and older who speak a language other than English at home is indicated in Table 2 at the appendix of this Article.

2010]

ENGLISH-ONLY LAWS

15

first relates to the xenophobia and racism that swirls around passage of English-Only laws.70 The second is the general context of Southern regional politics once marked by segregation.7 1 It was also a political arrangement, intriguingly, once dominated by the Democratic Party, an occurrence duly memorialized in the appellation of "solid South,"7 2 but, since the 1980s, has become controlled by the Republican Party. V. RESULTS AND INTERPRETATIONS To test the foregoing hypotheses, we collected data on English-Only laws involving all U.S. states, excepting Nebraska, which has a singlechamber (unicameral) legislature. Our study covered the period 1990 to 2007,74 and our sample size ("n") comprised 857 cases, excluding missing 76 data in the three categories of political party affiliation, foreign-born,
70. See Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 85-86 (2009). 71. See, e.g., William S. McFeely, Foreword to C. VANN WOODWARD, THE STRANGE CAREER OF JIM CROW (3d prtg. 1969); see also JAMES M. GLASER, THE HAND OF THE PAST IN CONTEMPORARY SOUTHERN POLITICS (2005); EARL BLACK & MERLE BLACK, POLTICS AND SOCIETY IN THE SouTH (1987); V.O. KEY, JR., SOUTHERN POLITICS IN STATE AND NATION (2d prtg. 1950). 72. See William S. McFeely, Foreword to C. VANN WOODWARD, THE STRANGE CAREER OF JIM CROW (3d prtg. 1969); see also JAMES M. GLASER, THE HAND OF THE PAST IN CONTEMPORARY SOUTHERN POLITICS (2005); EARL BLACK & MERLE BLACK, POLITICS
AND SOCIETY IN THE SoUTH (1987); V.0. KEY, JR., SOUTHERN POLITICS IN STATE AND NATION (2d prtg. 1950). 73. EARL BLACK & MERLE BLACK, THE RISE OF SOUTHERN REPUBLICANS 1-5 (2002). Elaborating on the Republican Party domination, these two southern scholars recounted that "[i]n 1950[,] there were no Republican senators from the South and only 2 Republican representatives out of 105 in the southern House delegation." Id. at 2. However, "[a] half-century later[,] Republicans constituted majorities of the South's congressional delegations-13 of 22 southern senators and 71 of 125 representatives." Id. at 3. 74. The endpoint, 2007, is the most recent data available. As for 1990, that is the census date closest to the recrudescence of English-Only laws in the 1980s that followed the proposal of the "English Language Amendment" by Senator Samuel I. Hayakawa.

Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and EnglishOnly Laws, 53 How. L.J. 53, 61-62 (2009).
75. We found 20 cases where the party of the governor was identified as "independent." These instances included four cases in Alaska from 1990 to 1993, 4 cases in Connecticut from 1991 to 1994, 8 cases in Maine from 1995 to 2002, and 4 cases in Minnesota from 1999 to 2002. In each of these instances, we coded the party affiliation as missing. There were 29 cases where a question arose regarding which political party was dominant, given that control of either of the chambers of the legislature was tied 50-50 between the Democratic Party and the Republican Party. Nine of those instances involved a lower house, while the remaining 20 involved an upper house. In these instances, we coded whichever party that controls the other house as dominant party. Finally, we classified the Minnesota Democratic-Farmer-Labor Party as the Democratic Party. The basis for this decision is that the party is affiliated with the Democratic Party. See Minnesota Demo-

16

THE SCHOLAR

[Vol. 13:1

and other languages." To measure this data statistically, we used logistic regression." Table 3 in the appendix presents our results. The table incorporates two sets of findings, denominated Models 1 and 2. Model 1 summarizes the results of politicalparty affiliation variables,foreign-born, other language, and regional/spatialcontrols on the passage of EnglishOnly laws, while Model 2 integrates the interaction term of foreign-born and other language." We took a number of additional steps designed to solidify our results and increase confidence in our findings. The first is our GIS analysis."o Second, to get around potential problems arising from heteroskedasticity8 ' and autocorrelation 8 2 in the errors associated with time series, cross-

cratic-Farmer-Labor Party, Introduction to the DFL Party, http://dfl.org/introduction (stating that: "The Minnesota DFL supports and works to enact the ideals and principles of the Democratic Party"). 76. Data for 1991 to 1999 and for 2001 to 2002 were not available. As a result, 1990 data was also used for 1991 to 1999 and 2000 data was used for 2001 and 2002. In effect, there was no variance within states for this variable from 1990 and 1999 as well as for 2000 and 2002. 77. Data for 1991 to 2002 were not available. As a result, 1990 data were also used for 1991 to 1999 and 2003 data were used for 2000 to 2002. In effect, there was no variance within states for this variable from 1990 to 1999 as well as from 2002 to 2003. 78. Logistic regression is a statistical device used in the social sciences when the dependent variable (or matter under investigation) is dichotomous and the independent variables are of any type. See, e.g., JOSEPH M. HILBE, LOGISTIC REGRESSION MODELS (2009); SOLEMAN H. ABu-BADER, ADVANCED & MULTIVARIATE STATISTICAL METHODS FOR SoCIAL SCIENCE RESEARCH WITH A COMPLETE SPSS GUIDE 127-55 (2010). 79. The rationale for including the interaction term is to account for the possible differential effect of foreign-born populations who speak only English and those who speak languages other than English. For more on interaction effects, see, for example, JAMES J. JACCARD AND ROBERT TURRISI, INTERACTION EFFECTS IN MULTIPLE REGRESSION (2d ed. 2003). 80. See Figure 1 in the appendix, along with the material in this section immediately following recapitulation of our hypotheses and results. 81. Heteroskedasticity means "differing variance" and comes from the Greek hetero ("different") and skedasis ("dispersion"). See W. PAUL VOGT, DICTIONARY OF STATISTICS
& METHODOLOGY: A NONTECHNICAL GUIDE FOR THE SOCIAL SCIENCES 104 (1993). In

statistics, a sequence of random variables is heteroscedastic or heteroskedastic, if the random variables have different variances. Id. The flipside is homoscedasticity, a sequence of random variables with constant variance. Heteroskedasticity occurs when there is a large difference among the sizes of the observations. See, e.g., SOLEMAN H. ABu-BADER, AD& MULTIVARIATE STATISTICAL METHODS FOR SOCIAL SCIENCE RESEARCH WITH A COMPLETE SPSS GUIDE 78-79 (2010); Halbert White, A Heteroskedasticity-Consistent
VANCED

CovarianceMatrix Estimatorand a Direct Test for Heteroskedasticity, 48 ECONOMETRICA 4, 817-38 (1980). 82. Autocorrelation means "[c]orrelation between members of a series of observations, such as weekly oil prices or interest rates" and "occurs when residual error terms from observations of the same variable at different times are correlated. Such correlations

2010]1

ENGLISH-ONLY LAWS

17

sectional panel data,83 we used robust statistics to yield more conservative inferences.' Finally, Table 3 is supplemented by another chart, Table 2, also in the appendix, which provides important information on the means, standard deviations, ranges, and sample sizes of various variables used in this study, including the percentage of U.S. residents that are foreign-born, and the percentage of residents (five years and over) who speak a language other than English at home. As the discussion below makes clear, the statistical details in Table 4 are particularly helpful in Hypotheses Numbers 7 and 8 for the light they throw with respect to the measurement of the percentage of foreign-born residents and the percentage of these individuals who speak languages other than English. We now present the results of the hypotheses outlined in Part IV; we take the hypotheses in their numerical order. Hypothesis Number 1 predicts that states with a Democratic governor and a legislature controlled by or unified in favor of the Democratic Party are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. There is support in our data in both Models 1 and 2. Additionally, the results are statistically significant. Hypothesis Number 2 predicts that states with a Democratic governor and a legislature controlled by or unified in favor of the Republican Party are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. There is support in our data in both Models 1 and 2; however, the results are not statistically significant. Hypothesis Number 3 predicts that states with a Democratic governor and a legislature split, in the sense that the majority party in one chamber is the Republican Party while the majority party in the other chamber is the Democratic Party, are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. There is statistically sig-

can raise several kinds of interpretive problems." W. PAUL VOGT, DICTIONARY OF STATISTICS & METHODOLOGY: A NONTECHNICAL GUIDE FOR THE SOCIAL SCIENCES 14 (1993). 83. In statistics, the term panel data refers to two-dimensional data, specifically a data set containing observations on multiple phenomena observed over multiple time periods. See generally, CHENG HsIAO, ANALYSIS OF PANEL DATA (2d ed. 2003).
84. Robust statistics provide an alternative approach to classical statistical methods in an attempt to produce estimators that are not unduly affected by deviations from model assumptions, such as outliers. See generally, PETER J. ROUSSEEUW & ANNICK M. LEROY, ROBUST REGRESSION AND OUTLIER DETECTION (1987); W. PAUL VOGT, DICTIONARY OF STATISTICS & METHODOLOGY: A NONTECHNICAL GUIDE FOR THE SOCIAL SCIENCES 198 (1993) (stating that robustness serves to keep a statistics "useful even when one (or more) of its assumptions is violated."); see also Table 3 in the appendix (figures for robust standard errors in the chart are secluded in parentheses).

18

THE SCHOLAR

[Vol. 13:1

nificant support (at the 0.01 level) in the findings for this hypothesis in both our main and interaction effect models. We surmise that the result agrees with our assumption regarding the importance of executive leadership, particularly in the situation of a split legislature where, besides controlling the executive branch, one political party additionally shares control of the legislature. For the Democratic Party, the situation is second only to the "ideal" of full control (Democratic governor and a legislature controlled by or unified in favor of the Democratic Party) which, based on the hypotheses we formulated for this Article, sets the stage for non-passage of English-Only laws. This result bears comparison with the situation in Hypothesis Number 2 where the governorship is in the hand of the Democratic Party while the legislature is Republicancontrolled. Although, as we saw in that hypothesis, while the finding indicated support for the hypothesis in both Models 1 and 2, the results were not statistically significant. Hypothesis Number 4 predicts that states with a Republican governor and a legislature controlled by or unified in favor of the Democratic Party (i.e., a government whose control is divided between the two major political parties) are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. This hypothesis has support in our findings in both Models 1 and 2. However, the findings were not statistically significant. Hypothesis Number 5 predicts that states with a Republican governor and a legislature split, in the sense that the majority party in one chamber is Republican and the majority party in the other chamber is Democratic, are less likely to adopt English-Only laws, compared to states with a Republican governor and a legislature controlled by or unified in favor of the Republican Party. This hypothesis has support in our findings in both the main effect (Model 1) and interaction effect (Model 2). Additionally, the findings were statistically significant in both models at the .01 level. One explanation of this intriguing result is that full control of the administration in a Republican government, defined to encompass Republican governorship plus a unified Republican legislature, is necessary to guarantee passage of English-Only laws, and that sharing legislative control between the Democratic and Republican Party eats away some of the full control necessary for the adoption of English-Only measures. Assuming the reader accepts this explanation, the results seem to contradict our assumption regarding the influence of executive leadership, Democratic or Republican, in the passage of English-Only laws. If, as we indicated, analyzing Hypothesis Number 3, the scenario of split legislative

85. See Part IV.B. of this Article.

2010]

ENGLISH-ONLY LAWS

19

control (second only to the "ideal" of Democratic governorship and a unified Democratic legislature) leaned toward the logical outcome of non-passage of English-Only laws, the scenario here should tend toward passage, not non-passage, of English-Only laws. For, as in Hypothesis Number 3, split legislative control for the Republican Party is second only to the "ideal" situation involving control of both the executive and legislative arms of a government. A possible rebuttal could be that the adoption of English-Only measures requires affirmative energy, 8 6 whereas non-passage is more passive or unobtrusive. However, such rebuttal lacks persuasiveness. Hypothesis Number 6 postulates that states with higher than average percentage of foreign-born persons are more likely to adopt English-Only laws, compared to states with a lower than average percentage of foreign-born residents." The hypothesis is supported in both Models 1 (main effect) and 2 (interaction effect) in our logistic regression analysis. However, the findings in the interaction term, Model 2, are not statistically significant, in that the significance disappears when the interaction term is included. A conceivable explanation is nativist anxiety and xenophobia arising from increased immigration, a factor leading to adoption of English-Only laws, appears to not be felt by the mere presence of foreign-born persons, but rather by the presence of foreign-born persons who, probably because they are not "Americanizing" quickly or fully enough, are viewed by nativists as deserving of the supposed "unity" therapy of English-Only laws."

86. Such energy includes tension within the electorate generated from division between supporters and opponents of such measures. See, e.g., Larry Rohter, Repeal Is Likely for 'English Only' Policy in Miami, N.Y. TIMES, May 14, 1993, available at 1993 WLNR 3428791 (recounting the dispute and surrounding tension between advocates and opponents of an English-Only measure that opponents wanted repealed). As one newspaper reporter somewhat wryly puts it, reporting on the adoption of an English-Only measure in a Tennessee locality, such measures "[q]uiet [a]ll but English." Robbie Brown, In Nashville, A Ballot Measure That May Quiet All but English, N.Y. TIMES, Jan. 11, 2009, at A16, available at 2009 WLNR 549411. 87. The percentage of U.S. residents who are foreign-born is indicated in Table 2 at the appendix of this Article. 88. English-Only advocates claim using foreign languages threatens unity within the United States, while opponents deny this contention, pointing to the predominance of the English language in this country. Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L.J. 53, 81, 84 (2009). Opponents of English-Only laws explain that what unites Americans "is not linguistic or ethnic homogeneity[,] but rather a shared commitment to democracy, liberty, and equality." Id. at 84 (quoting American Civil Liberties Union of Florida, English Only, http://www.aclufl.org/ takeaction/downloadresources/info papers/6.cfm). For example, the American Civil Liberties Union protested that "[a]n English Language Amendment to the [U.S.] Constitution," of the type that Senator Hayakawa proposed, "would transform the U.S. Constitu-

20

THE SCHOLAR

[Vol. 13:1

Hypothesis Number 7 posits that states with a higher than average percentage of residents who speak languages other than English are more likely to adopt English-Only laws, compared to states with a lower than average percentage of residents who speak languages other than English.89 The findings supported this hypothesis in both Models 1 and 2, and the results were statistically significant. However, the result moved in the opposite direction of what we postulated. This is an intriguing result suggesting, as in the explanation we set forth in Hypothesis Number 8, the variable here requires an interactive force to activate it in the postulated direction. Hypothesis Number 8 asserts that states with a higher than average percentage of persons who speak languages other than English and have a higher than average percentage of individuals who are foreign-born, are more likely to adopt English-Only laws, compared to states with a less than average percentage of residents who speak languages other than English, along with a lower than average percentage of foreign-born residents.9 0 This was only tested in Model 2. The findings supported our hypothesis and were statistically significant. We interpret this result to mean that, although states with a substantial number of persons (higher than average percentage) who speak languages other than English were less likely to adopt English-Only laws, when these states have a higher than average percentage of foreign-born residents who speak languages other than English, they become more likely to adopt English-Only laws. Thus, it is not speaking a language other than English, per se, that triggers the anxiety and expression of xenophobia observed with increased immigration, leading to passage of English-Only laws (after all, some nativeborn Americans speak languages other than English), but rather being foreign-born and speaking languages other than English. Our ninth and final hypothesis predicts that states in the Southern region are more likely to adopt English-Only laws, compared to states in other regions, such as the Midwest, Northeast, and West. This hypothesis has support and is statistically significant at the 0.01 level. Below is a recapitulation of the results from the hypotheses we tested. Recapitulation of Hypotheses Tested and their Respective Findings

tion from being a charter of liberties and individual freedom into a charter of restrictions that limits, rather than protects, individual rights." Id. (quoting American Civil Liberties Union of Florida, English Only, http://www.aclufl.org/takeaction/downloadresources/ info-papers/6.cfm). 89. The percentages of U.S. residents who are foreign-born and speak languages other than English at home are indicated in Table 2 at the appendix of this Article. 90. Information related to these averages is provided in Table 2 in the appendix relating to these two variables.

2010]1

ENGLISH-ONLY LAWS
CHART

21

A
Findings Supported in both Models 1 and 2 and is statistically significant

Hypothesis Number 1

Statement Democratic governor and a unified Democratic legislature, compared to a Republican governor and unified
Republican legislature

Number 2

Democratic governor and a unified Republican legislature, compared to a Republican governor and a unified

Supported in both Models 1 and 2, but both are not statistically significant Supported in both Models 1 and 2, but both are not statistically significant Supported in both Models 1 and 2, but both are not statistically significant Supported in both Models 1 and 2 and is statistically significant.

Republican legislature
Number 3 Democratic governor and a split legislature, compared to a Republican governor and unified Republican legislature Republican Governor and a unified Democratic legislature, compared to a Republican governor and unified Republican legislature Republican governor and a split legislature, compared to a Republican governor and unified Republican legislature Relationship between percentage of foreign-born persons and adoption of English-Only laws Relationship between speaking other languages and the adoption of English-Only laws Relationship between speaking other languages, foreign-born, and adoption of English-Only laws Relationship between a state's location in the Southern region and the likelihood of adopting English-Only laws

Number 4

Number 5

Number 6

Supported in both Models 1 and 2, but the findings are not statistically significant in Model 2. Supported in both Models 1 and 2 and is statistically significant but moved in the opposite direction we did not expect Tested only in Model 2. The findings support the hypothesis and are statistically significant Supported in both Models 1 and 2 and the findings are statistically significant in both models

Number 7

Number 8

Number 9

The geographical analysis (Figure 1 in the appendix) converged with the logistic regression analysis findings. The figure reveals that, of the thirty states that have adopted English-Only laws, ten states, or 30% of the overall number, had a Republican governor and a legislature controlled by or unified in favor of the Republican Party, compared to seven states, or 23% of the overall number, with a Democratic governor and a legislature controlled by or unified in favor of the Democratic Party. Conversely, only two states, or 7%, of the states without English-Only laws had a Republican governor and a legislature controlled by or unified in favor of the Republican Party, compared to seven, or 23%, with a Democratic governor and a legislature controlled by or unified in favor of the Democratic Party (Figure 1). Regionally, the South had the most

22

THE SCHOLAR

[Vol. 13:1

states with English-Only laws, while the Northeast had the least. In sum, of the thirty states with English-Only laws, eleven, or 37% of the whole, are in the South, nine, or 30%, are in the West, eight, or 27%, are in the Midwest, while only two, or 7%, are located in the Northeast.
VI. CONCLUSION

In our larger study analyzing the impediments English-Only laws pose for public service delivery to LEP populations, we postulated a relationship between political ideology and English-Only laws that we have systematically measured and analyzed in this study, using data from fortynine states for the seventeen-year period from 1990 to 2007. The fortynine states, over the period 1990 to 2007, yielded a sample size of 857, excluding missing cases. Statistical tests of our hypotheses demonstrated political ideology-measured in terms of political party affiliation-foreign-born populations, persons speaking languages other than English, and regional factors all influenced state adoption of English-Only laws in this study. Findings from this study bring us full circle to our larger work: Given the steady growth in the U.S. immigrant community (made possible by liberalization of U.S. immigration law and the forces of globalization), 9 matched against the widespread prevalence of English-Only laws, foreign-born persons and others in the LEP community will experience unequal access to government services. For instance, LEP persons might encounter life-threatening situations requiring emergency medical attention at public hospitals that they will not get due access to because of barriers imposed by English-Only measures. This, and other possible outcomes, are matters for careful contemplation by policymakers who should work to expand the limited tools, centered around Executive Order 13166 and its progeny, for promoting equal access to public service delivery for LEP populations.9 2 Future studies on this topic should identify other variables that can be added to the model here to account for more variation in the dependent variable, or use a sample size of elongated duration beyond the one here that could, for example, include data from the 2010 census. Depending on data availability, such future studies could also assess the factors that

91. On liberalization of U.S. immigration law, see, for example, Philip C. Aka & Lucinda M. Deason, Culturally Competent Public Services and English-Only Laws, 53 How. L. 53, 73-74, 79-80 (2009); on the impetus of globalization, see id. at 114. 92. See id. at 106-10. For our suggestion on strengthening these tools to promote culturally competent public services, see id. at 121-28 (recommending the adoption of the national standards for culturally and linguistically appropriate services in health care as the best practice in public administration).

2010]

ENGLISH-ONLY LAWS

23

influence the adoption of English-Only laws in the national and local levels of the U.S. political system. Although this study improved upon our presentation in our larger study, and, although it sheds useful light on some of the systematic factors influencing the adoption of English-Only laws, filling the knowledge-gap that remains on this important topic requires that we know more about occurrences at the national and local levels of the U.S. system.

24
TABLE 1. State

THE SCHOLAR

[Vol. 13:1

STATES THAT HAVE ADOPTED ENGLISH-ONLY LAWS Form of Law Adoption Year(s)

Alabama Alaska Arizona Arkansas California Colorado Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Massachusetts Mississippi

Constitutional amendment passed by ballot Statutory voter initiative Statute Constitutional amendment Constitutional amendment Constitutional amendment Statute Constitutional amendment by the legislature95 Statute Statute Statute Statute Statute Statute Adopted English as precondition for admission into the Union as a state Original constitution interpreted as having established an official language
93

1990 1998 2006 1987 1986 1988 1988 1986, 1996 1978 2007 1969 1984 2002 2007 1984 1807 1975 1987

Constitutional amendment 94

Statute

93. In March of 1999, an Alaskan superior court declared the state's English-Only law unconstitutional. Kritz v. Alaska, Nos. 3DI-99-12 CI, 3AN-99-4488 CI., 2002 WL 34220502 (Super. Ct. Mar. 2002), affd in part sub nom. Alaskans for a Common Language, Inc. v. Kritz, 170 P.3d 183 (Alaska 2007) (granting a preliminary injunction preventing enforcement of the successful ballot measure requiring the use of English only as the official language of the Alaskan government); see James Crawford, Issues in U.S. Language Policy: Language Legislation in the United States of America, http://www.languagepolicy.net/ arch ives/langleg.htm (last visited June 10, 2010) (documenting various English-Only laws in the United States and their history, including the history of Alaska's English-Only proposal). The English-Only initiative in Alaska was partially upheld and partially reversed by the Supreme Court of Alaska in 2007. Alaskans for a Common Language, Inc. v. Kritz, 170 P.3d 183, 202, 211 (Alaska 2007), affg in part Kritz v. Alaska, Nos. 3DI-99-12 CI, 3AN99-4488 CI., 2002 WL 34220502 (Super. Ct. 2002) (invalidating and severing part of the Alaska initiative making English the only official language of the government). 94. In 1988, the Arizona Supreme Court overturned the state's English-Only amendment, see Ruiz v. Hull, 957 P.2d 984, 987 (Ariz. 1998) (en banc). But voters reinstated the measure in Nov. 2006. See ProEnglish, Legal Status of Official English in Arizona, available at http://www.proenglish.org/states/arizona.htm. 95. This amendment recognized English and Native Hawaiian as official languages, an occurrence which makes this state bilingual. HAW. CONST. art. XV, § 4.

2010] Missouri Montana Nebraska New Hampshire North Carolina North Dakota South Carolina South Dakota Tennessee Utah Virginia Wyoming

ENGLISH-ONLY LAWS Constitutional amendment by popular vote Statute Constitutional amendment by popular vote Statute Statute Statute Statute Statute Statute Statute passed by petition initiative Statute Statute 2008 1995 1920 1995 1987 1987 1987 1987 1984 2000 1981, 1996 1996

25

Source: Adapted from ProEnglish, English as Our Official Language: English in the 50 States, available at http://www.proenglish.org/issues/offeng/states.html (last visited June 18, 2010).

26
TABLE Variable

THE SCHOLAR

[Vol. 13:1

2.

DESCRIPTION OF VARIABLE USED IN THE STUDY Mean (SD) Range N

Definition

Dependent Variable
English-Only-Laws

Whether or not a state had
adopted an English-Only

0-1

862

Law Independent Variables
PoliticalParty Affiliation

(PPA) Demo. governor and unified Demo. legislature Demo. governor and unified Rep. legislature Demo. governor and split legislature Rep. governor and unified Demo. legislature Rep. governor and unified Demo. legislature Rep. governor and split legislature
Non-PPA Independent Variables

96

0-1

862

0-1 0-1 0-1 0-1 0-1

862 862 862 862 862

Foreign-born Other Language

%-age of U.S. residents not born U.S. citizens 5 years and older who speak a language other than English at home
97

6.12 (5.37) 11.36 (8.62)

0.8-27.42 2.09-42.61

881 877

Regional/SpatialControls Northeast Region Midwest Region West Region

0-1 0-1 States in West region 0-1

882 882 882

96. Each political party affiliation variables was a dummy variable, compared to states with a Republican governor and legislative houses controlled by or unified in favor of the Republican Party. 97. Each of the regional/spatial control variables was a dummy variable compared with the Census Bureau's Southern region.

2010]
TABLE

ENGLISH-ONLY LAWS

27

3.

LOGisTic REGRESSION OF ENGLISH-ONLY LAWS IN THE UNITED STATES: 1990 THROUGH 2007 Model 1 Model 2 z-value -1.65* -1.04 -3.56*** -0.21 -3.01*** Coefficient -. 464 (.244) -. 190 (.256) -. 929 (274) -. 054 (.255) -. 827 (.280) Z-value -1.90* -0.74 -. 39*** -0.21 -2.99***

Varible Political Party Affiliation Variables
(PPAV)

Coefficient -3.90 (.237) -. 270 (.259) -1.003 (282) -. 0500 (.238) -. 812 (.270)

Democratic Governor and Unified Democratic Legislature Democratic Governor and Unified Republican Legislature Democratic Governor and Split Legislature Republican Governor and Split Legislature Non-PPA Variables Foreign-born Other Language Foreign-born * Other Language Regional/Spatial Controls Northeast Region Midwest Region West Region Constant Pseudo R
2

.162 (.030) -. 098 (.020)

5.48*** -5.02***

.026 (.050) -. 141 (.021) .006 (.002)

0.51 -6.58*** 4.06***

-2.366 (.260) -1.174 (.223) -1.084 (.219) 1.231 (.219)

-9.11*** -5.26*** -4.95*** 5.61*** 0.126 132.65*** 857

-2.181 (.263) -1.206 (.227) -1.047 (.219) 1.830 (.279)

-8.31*** -5.31*** -4.78*** 6.57*** 0.139 151.16*** 857

2 Model (Wald) X (10/11)

N

Notes: 1. The state of Nebraska was not included in these models because it has a unicameral legislature. 2. Robust standard errors are in parentheses; statistical significance at the 0.01 level, 0.05 and
0.10 level is indicated by ***, ** and *, respectively.

3. The pseudo R-square, unlike R-square in ordinary least-square (OLS) regression, does not measure percentage of variance in the dependent variable. This is because, technically, there is little valiance in a dichotomous dependent variable (which ranges from 0 to 1). The pseudo Rsquare measures changes in likelihood and related quantities, and is useful for comparing different models. Sources: World Almanac (2009); U.S. Census Bureau (2009a); U.S. Census Bureau (2009b); National Governors Association (2009).

28 FIGURE

THE SCHOLAR

[Vol. 13:1

1. STATES CLASSIFIED BY ENGLISH-ONLY STATUS, POLITICAL PARTY AFFILIATION, AND CENSUS REGION AS OF 2007

DDD RSDD DD

No-EgisDOS statesD

~Eglth-n~ystaesDR ~No-Engsh-Oly sates
____

OD - Denmocratic Govemorand Democrtatic controlled legislature -DemnocraticGovemnor Republican-controled end OS- Deemocratic Govemnor one chamberis controlledby the and and Republicans the otheris controlledby theDemocratn RR - Raeblican Gnvernor and Repubilcan-controffed legislature legislature RD -Republican Govemoor Demoocratiocontrolled end RSRepubianGor theand chamer is by theDmorl controlled RRepbnan thoher anis chbn ontolled by theDeort

legislature

Files

Citation

Philip C. Aka, Lucinda M. Deason, Augustine Hammond, “Measuring the Impact of Political Ideology on the Adoption of English-Only Laws in the United States,” St. Mary's Law Digital Repository, accessed November 20, 2017, http://lawspace.stmarytx.edu/item/STMU_TheScholarStMarysLRev_v13i1p0001_Aka_Deason_Hammond.

Document Viewer