The Voting Rights Act and Latino Voter Registration

Abstract

This study explores how the language minority provisions in the Voting Rights Act (VRA) affect Latino voter registration. We are particularly interested in how these provisions affect Latino citizens with varying levels of English language proficiency. Using data from the 2006 National Latino Survey, we find that Latino citizens with limited English skills register to vote at about the same rate whether or not they live in a county mandated by the VRA to provide registration and voting materials in Spanish. However, for Latinos who speak English “very well,” we find that access to these materials is associated with increased registration rates, all else equal. We interpret these findings to suggest that the positive effects of VRA coverage on Latino registration are due to a symbolic “welcoming” effect, rather than substantial reductions in administrative barriers to registration.

Keywords

Latino political participation voter registration VRA

Broad based electoral participation is the hallmark of a vibrant democracy. Yet, in the United States, participation rates have continuously fallen short of expectations, with many Americans foregoing their right to participate in the selection of their representatives. Participation rates are particularly low among minority citizens, especially those in language minority communities (e.g., Leighley, 2001; Leighley & Vedlitz, 1999). Studies show that limited English skills within these communities can be a significant barrier to broader electoral participation, marginalizing millions of non-English-speaking Americans from the political system (e.g., Hopkins, 2011; Parkin & Zlotnick, 2011; Ramakrishnan, 2005; Uhlaner, Cain, & Kiewiet, 1989). This is a particularly critical issue in a “nation of immigrants,” where minority populations continue to grow rapidly (Gaouette, 2006).

Recognizing the importance of this barrier, Congress added provisions to protect and facilitate the voting rights of language minorities in the 1975 reauthorization of the Voting Rights Act (VRA). Sections 4(f)(4) and 203 of the VRA mandate that counties with a significant language minority population (i.e., more than 10,000 or 5% of voting age citizens) must provide election materials (e.g., registration and voting notices, forms, instructions) in the language of the applicable minority group (for details, see Tucker & Espino, 2007). According to the Department of Justice (2012),

The requirements of the law are straightforward: all election information that is available in English must also be available in the minority language so that all citizens will have an effective opportunity to register, learn the details of the elections, and cast a free and effective ballot.

The VRA’s objectives are quite clear; however, there is some uncertainty and debate about how well these objectives are being met. The Supreme Court’s recent invalidation of the VRA’s preclearance formula in Shelby County v. Holder highlights the need for a clearer understanding of how and when legislation is successful in expanding political participation. Although the provisions requiring non-English election materials remain in place, states and counties with a history of low minority participation and discriminatory election practices, including heavily Latino Texas, Arizona, and parts of California, are no longer required to seek approval from the Department of Justice before making changes to voting procedures or district boundaries that could reduce the electoral power of minorities. Writing for the majority, Chief Justice Roberts found that substantial improvements in minority participation since the passage of the law showed that cultural changes in the South have eliminated the discriminatory tendencies that had earlier warranted the preclearance requirement. In a dissenting opinion, Justice Ginsburg refuted the majority’s reasoning, arguing that the gains in participation are evidence of the law’s effectiveness, rather than an indication that it is obsolete. This disagreement illustrates the need for a more nuanced understanding of how the various provisions of the VRA affect political participation. Without a firm and detailed understanding of how and why the law has been effective, it is difficult for researchers to predict the impact of changes to the law, policymakers to improve those parts that have not been effective, and advocates to defend the law from its opponents.

Additional research is further justified by the fact that most academic studies evaluating the VRA have tended to focus on its relationship to voter turnout with little or no attention given to its potential impact on voter registration (e.g., de la Garza & DeSipio, 1997; Hopkins, 2011; Jones-Correa, 2005; cf. Parkin & Zlotnick, 2011; Timpone, 1998). This is an important oversight for two reasons: (1) It neglects the primary and necessary first step in the participation process—voting is only possible after registration, and (2) it conflates more participation with broader participation. As such, we have an incomplete picture of how the VRA’s language provisions affect participation in language minority communities.

To address this gap, we focus our study on the relationship between the VRA’s language minority provisions and voter registration. We are particularly interested in the VRA’s effect on registration in the Latino community, which is by far the largest language minority community in the United States and now the fastest growing segment of the American population (Gaouette, 2006). We assess whether the VRA’s language provisions are meeting one of their most critical objectives: Are they increasing the registration rate among citizens with limited English skills and if not, why not? We start in the next section with a review of the relevant literature and a discussion of our hypotheses. We then describe our data and methodology before discussing our results and their implications.

The VRA and Latino Electoral Participation

Existing research on the language minority provisions in the VRA is largely limited to studies of turnout in covered jurisdictions. The results have been mixed but suggest a conditional relationship between English language skills, VRA coverage, and voter turnout. In the earliest study, de la Garza and DeSipio (1997) find little effect on Latino voter turnout, which they attribute to the fact that most American Latinos speak English well, have little need for Spanish language election materials, and report using them at low rates. Hopkins (2011) and Parkin and Zlotnick (2011), using different datasets, find a conditional positive effect that supports de la Garza and DeSipio’s intuition: VRA coverage increases turnout among Spanish-speaking Latinos but has no effect on fluent English speakers. Because many Latino citizens speak English, the effect for Spanish speakers is obscured when Latinos with differing language skills are pooled together. Parkin and Zlotnick additionally test the impact that VRA coverage has on registration and find no statistically significant effect in multiple regression models; however, the reliability of their finding is limited by imperfectly estimated geographic data on respondent location. Fraga (2010) considers the short-term effect of VRA coverage by examining changes in voter turnout between 2000 and 2004 in those precincts that gained VRA coverage between the elections and finds an increase in Latino and non-Latino voter turnout. Finally, pointing out that Congress’ intention in passing the language provisions of the VRA was not to facilitate voting by immigrants but rather to increase participation by native-born language minorities, Ramakrishnan (2002) and Jones-Correa (2005) have evaluated generational differences in the effect of VRA coverage. Their results are mixed, with Ramakrishnan finding a greater benefit for the native-born and Jones-Correa finding a stronger effect for first-generation immigrants.

Although the better part of the existing literature on the VRA’s language provisions considers their effect on Latinos, there have been a small number of studies that consider other American language minorities. In contrast to the findings on Latinos, Lien (2001) and Ancheta (1998) find that Asian Americans report using foreign language election materials at high rates, suggesting that the effects of VRA coverage may vary between the groups covered.

The existing literature’s near exclusive focus on turnout has overlooked the potential for the legislation to have an effect beyond simply facilitating voting by those already likely to participate by additionally bringing politically marginalized minority groups into the political process. Increasing turnout alone indicates more participation, but it does not necessarily indicate broader participation, which requires expansion of political interest and activity among those segments of target populations that have traditionally eschewed politics entirely. If the VRA improves turnout among the registered but not registration itself, we may suspect that the legislation facilitates voting among those who are already politically involved enough to register to vote. If, however, VRA coverage also increases registration rates, we can infer that the election materials and assistance provided by the legislation also expand participation by those who would otherwise remain on the sidelines of the political process.

We begin our analysis by testing the effect that VRA coverage has on Latino registration. We analyze the expectation that, all else equal, registration rates will be higher among Latino citizens living in VRA covered counties than they are for those living in counties that are not covered by the legislation. Our past results notwithstanding (Parkin & Zlotnick, 2011), there are two primary reasons to expect support for this hypothesis. First, Spanish-language registration materials should ease the administrative burden of registering. Whereas Spanish-language materials have a conditional effect on Latino turnout (e.g., Hopkins, 2011; Parkin & Zlotnick, 2011), they could have an even more noticeable effect on Latino registration given that registration is typically more cumbersome than voting (Tam Cho, 1999; Timpone, 1998). In fact, VRA coverage should have a particularly pronounced impact on registration rates because it simplifies what might otherwise be perceived as a daunting task. Second, providing registration materials in Spanish ought to welcome Latino citizens into the political process. Even if VRA coverage does little to reduce the administrative burden, particularly for those who already speak English fluently, the provision of Spanish-language registration materials sends a clear message that Latino participation is welcomed and encouraged. This ought to provide an important psychological motivation for registering.

We also expect that VRA coverage will have a stronger effect on Latino citizens with relatively weak English skills than it does on those with greater English fluency. There are currently an estimated 6.5 million Latino citizens in the United States who speak little to no English (Shin & Kominski, 2010). This represents roughly 30% of Latino Americans, who we believe will be the primary beneficiaries of the VRA’s language minority provisions. Whereas the administrative assistance offered by Spanish-language materials is of limited practical value to English-speaking Latinos, it should be quite helpful to those with little or no English proficiency. In fact, VRA mandates might be critical in allowing these citizens to register and become part of the political process.

Data and Method

We test these predictions with data from the 2006 National Latino Survey (NLS). The NLS was conducted with self-identified Latino residents of the United States between November 2005 and August 2006. Given our focus on registration, we exclude all non-U.S. citizens. This leaves us with a nationally representative sample of 4,764 Latinos with U.S. citizenship.

Our dependent variable is individual-level registration, based on the question “Are you currently registered to vote in the U.S.?” (yes, registered; no, not registered). We find that 74.9% of our sample reports that they are registered to vote, which is generally in line with past measures (Bass & Casper, 2001; Calvo & Rosenstone, 1989; de la Garza & DeSipio, 1997; Jamieson, Shin, & Day, 2002; Schildkraut, 2005; Uhlaner et al., 1989).

Our primary independent variable is coverage under the language provisions of the VRA—sections 4(f)4 and 203. Unlike past studies that have estimated coverage using the respondent’s Metropolitan Statistical Area (Parkin & Zlotnick, 2011), we use the restricted county identifier variable in the 2006 NLS to determine which respondents live in covered areas and which do not. Specifically, we match the respondent’s county with data from the Department of Justice website on which counties were covered in 2005-2006 when the NLS was conducted (Department of Justice, 2012). For example, respondents in Broward County, Florida, are coded as VRA covered (1) while those in Benton County, Arkansas, are coded as not covered (0). In our sample, 70.6% of respondents live in covered jurisdictions and 29.4% do not, reflecting that most Latino citizens reside in areas subject to the VRA language requirements.

Our models also include independent variables measuring English language skills, the size of the Latino community, and county-level voter registration requirements. We use the following question to measure English language proficiency: “How good is your spoken English: Would you say you could carry on a conversation in English (both understanding and speaking) very well (4), pretty well (3), just a little (2) or not at all (1)?” While others have measured English proficiency with the language used to conduct the survey (e.g., Hopkins, 2011), we elected to use the self-assessment measure because such measures have proven to be valid (Bachman & Palmer, 1989; Lutz, 2006; Oscarson, 1989), and this particular measure offers a more fine-grained assessment of actual English skills. A respondent who speaks English well may, for whatever reason, elect to take the survey in Spanish, which would lead us to mistakenly categorize him or her as having limited English skills. We believe that this type of mismeasurement is less likely with the self-assessment question. As it turns out, the decision to use a self-assessment measure is further justified by the fact that it correlates highly, but not perfectly, with the language used to take the survey (.713, p = .000).

We measure the size of the Latino community for each respondent using county-level data from 2000 U.S. Census. For example, the Latino population in Texas’s Bexar County was recorded as 54.3% in 2000, while the Latino population in Colorado’s Jefferson County was 10%. The average respondent in our sample comes from a county where 30.1% of the population is Latino. This measure allows us to control for the potential impact that geographic concentration and personal interaction within the Latino community might have on registration activity (Barreto, Segura, & Woods, 2004; Leighley, 2001.) We include this variable because Latino citizens, particularly those with limited English skills, might be more likely to register if they live in areas with abundant social encouragement and assistance.

We control for variation in registration requirements by including a variable based on 2004 state voter registration deadlines. For example, voters in Arizona had to register 29 days before Election Day, while those in West Virginia had to be registered 20 days before votes were cast. The average respondent in our sample has a registration period that is 25.5-day long. We include this variable in our models because we anticipate that stricter registration laws may dampen the registration rate (e.g., Ansolabehere & Konisky, 2006).

Following past research in this area, we also include controls for age, education, annual household income, nativity, years in the United States (identical to age for American-born respondents), church attendance, family’s national heritage, and political interest (e.g., Bass & Casper, 2001; DeSipio, 1996; Hopkins, 2011; Jackson, 2003; Leighley & Vedlitz, 1999; Parkin & Zlotnick, 2011). The average age in our sample is 36.8 years, average education level is “high school graduate,” average annual household income is somewhere between US$35,000 and US$45,000, and church attendance is a little more than “once a month.” Within the sample, 34.9% are naturalized citizens; among this group, the average time since naturalization (i.e., length of citizenship) is 8.2 years. As for national heritage, 64.6% of the sample identifies as Mexican, 13.8% as Puerto Rican, 5.2% as Cuban, and the remaining 16.4% identify as other Central or South American. Finally, 48.7% of respondents claimed to be “somewhat interested” in politics, while 25% were “very interested” and 26.3% were “not very interested.”

We use these data in a number of tests below. We start in the next section with an analysis of the direct relationship between VRA coverage and voter registration among Latino citizens. We then explore the VRA’s effect on Latino registration, and its conditional effect in relation to English language proficiency, in a series of logistic regressions, controlling for the variables mentioned above.

Results and Discussion

Our first test compares the average Latino registration rate in counties covered by the minority language provisions in the VRA and those that were not covered in 2005-2006 when the NLS was administered. As anticipated, Latino registration rates are higher in those counties covered by minority language provisions. Specifically, 78% of respondents living in covered jurisdictions are registered to vote, compared with 67% of those living in noncovered counties (t = −7.746, p = .000). In an uncontrolled comparison of Latino citizens, those residing in counties covered by VRA provisions are more likely to be registered than those living in counties that do not require the provision of Spanish-language registration materials.

Of course, the difference in registration rates could be caused by other factors that vary simultaneously with VRA coverage. Thus, we next test the effect of these provisions in a larger model with various controls. The logistic regression model in Table 1 predicts the probability of being registered based on the respondent’s location inside or outside of a VRA covered county, controlling for other pertinent factors including English language proficiency, the size of the Latino population, registration regulations, and a number of demographic variables.

Table 1.

Latino Registration.

	B	SE	Exp(B)
VRA coverage	.172	.126	1.188
English proficiency	.356***	.069	1.428
Latino population %	.008**	.003	1.008
Registration requirements	.001	.007	1.001
Age	.048***	.004	1.049
Education	.253***	.032	1.288
Income	.107***	.025	1.113
Naturalized Citizen	−1.132***	.206	0.322
Naturalized years	.037***	.009	1.038
Church attendance	.031	.036	1.031
Mexican heritage	−.335**	.131	0.715
Puerto Rican heritage	−.103	.185	0.902
Cuban heritage	−.137	.287	0.872
Political interest	.424***	.061	1.528
Constant	−4.400***	.433	0.012
n	3,605
Log likelihood	3,098.230

Note. VRA = Voting Rights Act.

†

p ≤ .1 (two-tailed tests). *p ≤ .05. **p ≤ .01. ***p ≤ .001.

The results in Table 1 show that the relationship between VRA coverage and Latino registration fades to statistical insignificance when controls are added to the model (p = .172 in a two-tailed test). In fact, there are a number of other factors that seem to play a more important role in determining Latino registration rates. These include English proficiency, the percentage of Latinos living in the respondent’s community, age, education, income, being a naturalized U.S. citizen, the length of time since naturalization, Mexican heritage, and political interest. Whereas being a naturalized citizen and having Mexican heritage are associated with lower levels of registration, all of the other factors are positively related to higher registration rates. Calculating the percentage changes in the odds ratio for each variable ((Exp(B) − 1) × 100) shows, for example, that speaking English “very well” increases the odds of being registered by 43% over those who speak English “pretty well”; each additional year of age is associated with a 5% increase in the odds of being registered; and being “very interested” in politics, rather than “somewhat interested,” increases the likelihood of registration by 53%. Church attendance, Puerto Rican and Cuban heritage, and registration requirements have no significant impact on Latino registration in this model.

In the first column in Table 2, we replicate the above model but limit the sample to only those with less than perfect English skills, which includes all respondents who rate their ability to carry on a conversation in English as less than “very well.” Dividing the sample allows us to investigate the impact that minority language provisions have on those who would seem to need them the most—that is, those with limited English-speaking abilities. The results show, however, that these provisions have no consistent impact on this seemingly receptive group of Latino citizens. In fact, the coefficient is far from being statistically significant (p = .476 in a two-tailed test). This leaves us to conclude that, despite expectations, the VRA’s provision of minority language registration materials does little to encourage or facilitate registration among Latino citizens with limited English skills. In other words, the probability of registering among low-English-proficiency Latinos is basically the same whether they live in a community covered by the VRA or not. VRA coverage seems to fall short of expectations for this group, all things being equal.

Table 2.

Latino Registration by English Fluency.

	Less than “very well”			“Very well”
	B	SE	Exp(B)	B	SE	Exp(B)
VRA coverage	−.181	.254	0.834	.242^†	.148	1.274
Latino population %	.006	.006	1.006	.010**	.004	1.010
Registration requirements	−.004	.014	0.996	.003	.008	1.003
Age	.027***	.007	1.027	.052***	.005	1.054
Education	.194***	.053	1.214	.295***	.040	1.343
Income	.150**	.055	1.162	.103***	.028	1.108
Naturalized citizen	−1.312***	.333	0.269	−1.115***	.292	0.328
Naturalized years	.057***	.012	1.058	.035*	.014	1.035
Church attendance	.057	.075	1.059	.012	.041	1.012
Mexican heritage	−.689**	.229	0.502	−.234	.160	0.791
Puerto Rican heritage	.562	.401	1.755	−.223	.213	0.800
Cuban heritage	−.137	.453	0.872	−.058	.373	0.944
Political interest	.257*	.120	1.292	.503***	.072	1.654
Constant	−1.822*	.725	0.162	−3.614***	.411	0.027
n	810			2,795
Log likelihood	817.860			2,264.638

Note. VRA = Voting Rights Act.

†

p ≤ .1 (two-tailed tests). *p ≤ .05. **p ≤ .01. ***p ≤ .001.

This presents something of a puzzle. The language provisions have an impact on Latino registration in a direct model, and they come close to being a significant predictor in a larger regression model with the full sample, and yet, fall well below statistical significance when we focus on Latino citizens with limited English skills, suggesting, counterintuitively, that the effect of the legislation might be found among the strongest English speakers. In the second column of Table 2, we present the results of an identical logistic regression with only those respondents who reported speaking English “very well.” This allows us to see whether the minority language provisions have an effect on those with less of an immediate need for them—that is, those with strong English skills.

The results in column 2 of Table 2 show that Spanish language registration materials have a positive and marginally significant impact on registration rates among Latino citizens with fluency in English (p = .102 in a two-tailed test). Specifically, the estimated percentage change in the odds ratio shows that those living in VRA covered jurisdictions are 27% more likely to be registered than those who live outside a covered area. This provides at least some evidence that the VRA’s minority language provisions have a positive influence on the registration rates of Latino citizens who are comfortable speaking English, even after a host of additional factors are considered.

To confirm this unexpected result, we ran a fourth logistic regression that includes an interaction between VRA coverage and English proficiency. As the results in Table 3 show, the interaction is positive and statistically significant, confirming that the effect that VRA coverage has on registration is greater for English-speaking Latinos than for those who have more limited English skills.

Table 3.

Latino Registration by VRA Coverage and English Fluency.

	B	SE	Exp(B)
VRA coverage	−.767*	.458	0.464
English proficiency	.174	.110	1.190
Latino population %	.008**	.003	1.008
Registration requirements	.001	.007	1.001
Age	.049***	.004	1.050
Education	.247***	.032	1.281
Income	.107***	.025	1.113
Naturalized citizen	−1.152***	.206	0.316
Naturalized years	.038***	.009	1.039
Church attendance	.032	.036	1.032
Mexican heritage	−.350**	.131	0.705
Puerto Rican heritage	−.119	.185	0.888
Cuban heritage	−.132	.287	0.876
Political Interest	.432***	.061	1.540
VRA coverage × English proficiency	.260*	.122	1.297
Constant	−3.719***	.536	0.024
n	3,605
Log Likelihood	3,093.670

Note. VRA = Voting Rights Act.

†

p ≤ .1 (two-tailed tests). *p ≤ .05. **p ≤ .01. ***p ≤ .001.

These results are surprising. We find that VRA coverage increases registration among those who speak English fluently or very well but not among the target group of those with weaker English skills. We now turn to explaining this puzzle: Why should the availability of Spanish-language registration materials have an effect on those who speak English well enough to make these materials seem unnecessary?

The most obvious interpretation is that the provision of registration materials in Spanish is an effective means for overcoming some types of obstacles but not others. Specifically, VRA coverage appears to fulfill a symbolic “welcoming” function but falls short of the level of practical assistance needed for weak English speakers to overcome substantial barriers to registration.

We test this mechanism by determining whether strong English speakers report higher levels of political efficacy (i.e., sense of political inclusion) in counties covered by the VRA than in those not covered by the legislation. Respondents were asked whether they agree or disagree with the following statements: “People are better off avoiding contact with government” and “Sometimes politics and government seem so complicated that a person like me can’t really understand what’s going on.” Responses were coded as either “agree” (0) or “disagree” (1). We run two logistic regressions with the measures of political efficacy described above as dependent variables and limit the sample to those who report speaking English “very well.” Again, VRA coverage is the independent variable of interest, controlling for the same factors as above (except the registration requirement variable, which is not included in this model because it should have little bearing on feelings of political efficacy).

As can be seen in Table 4, we find a negative and marginally significant (p = .095 in a two-tailed test) effect of VRA coverage for attitudes toward government interaction: Strong English speakers in VRA covered areas are less wary of interaction with the government than equally strong English speakers in noncovered counties, all else equal. In other words, VRA coverage is associated with the idea that people should feel comfortable contacting and interacting with their government. The estimates show that Latinos in VRA covered jurisdictions are about 19% less likely than those outside VRA coverage to claim that government contact should be avoided. We also see a negative and significant (p = .059 in a two-tailed test) effect with regard to personal political competency: Strong English speakers in VRA covered counties are more likely to disagree with the statement that politics is too complicated to understand than their counterparts living in areas not covered by the VRA. Again, the percentage change in odds of agreeing with this statement is about 19% between covered and noncovered Latinos who speak English fluently. Taken together, these results suggest that VRA coverage is associated with a stronger sense of being able to comprehend and engage with politics and feeling able and included in the political process, all else equal.

Table 4.

Political Efficacy Among Latinos Who Speak English “Very Well.”

	Avoid government contact			Politics is too complicated
	B	SE	Exp(B)	B	SE	Exp(B)
VRA coverage	−.208^†	.124	0.812	−.214*	.113	0.808
Latino population %	.003	.003	1.003	.003	.003	1.003
Age	−.003	.003	0.997	.006**	.003	1.006
Education	−.222***	.033	0.801	−.180***	.032	0.835
Income	−.039^†	.024	0.961	−.054**	.022	0.948
Naturalized citizen	.483**	.228	1.621	−.355^†	.213	0.701
Naturalized years	−.011	.008	0.989	.014^†	.008	1.014
Church attendance	−.026	.035	0.975	.091**	.031	1.095
Mexican heritage	.043	.134	1.044	.163	.120	1.177
Puerto Rican heritage	−.088	.178	0.915	.028	.157	1.029
Cuban heritage	−.222	.280	0.801	−.191	.223	0.826
Political interest	−.488***	.061	0.614	−.496***	.059	0.609
Constant	1.842***	.289	6.311	2.479***	.275	11.932
n	2,664			2,704
Log likelihood	2,997.281			3,470.812

Note. VRA = Voting Rights Act.

†

p ≤ .1 (two-tailed tests). *p ≤ .05. **p ≤ .01. ***p ≤ .001.

This evidence provides preliminary support for the symbolic effects theory and argues for a more systematic test of the symbolic mechanism. At this point, we find that for those facing psychological barriers to participating in the electoral process, the VRA’s language provisions provide an effective means for encouraging participation. The VRA seems to send a welcoming message such that, all else equal, English-speaking Latinos living in covered jurisdictions are more likely to register than their counterparts in noncovered areas. We have also provided some evidence that this is driven by an increased sense of political efficacy and inclusion.

For those facing an additional language barrier, the provision of Spanish-language registration forms and instructions appears to be insufficient to effectively overcome this obstacle. Latinos with limited English skills tend to register at roughly the same rates whether they live in a county covered by the VRA or not. Despite the VRA’s provisions, it may still be difficult for those with limited English skills to overcome the perceived administrative hurdles to registration. They may find it difficult, for example, to locate registration materials in their primary language, or to acquire enough political information to make elections salient. Studies have also shown that there is significant variation in the quality of precinct-level implementation of the language provisions of the VRA. Jones-Correa and Waismel-Manor (2007) and Tucker and Espino (2007) find significant variability in compliance with VRA requirements across and within covered states. Jones-Correa and Waismel-Manor specifically note that “the overall picture for the provision of translated registration materials is less than sanguine: just under half of the non-individually covered counties visited by [their] research team offered translated registration materials” (p. 175). It is possible that the effectiveness of the VRA at broadening participation is conditional on consistent and complete compliance with the law’s mandates. Our own attempts to locate Spanish-language materials from official online sources revealed that a working knowledge of written English was often required even to locate the appropriate pages. These deficiencies undermine the spirit and letter of the law, making the accommodations that are provided less accessible to those who need them the most.

We do not believe that these findings invalidate the usefulness of the VRA’s language provisions. Past studies have shown that the VRA can increase voter turnout in the Latino community (Hopkins, 2011; Parkin & Zlotnick, 2011). Here, we also find that the language minority provisions can increase registration rates among those Latino citizens who speak English well. While unexpected, this effect is nonetheless valuable: The VRA, as implemented, may not be reaching expectations in terms of significantly increasing registration among its primary target group, but it does seem to be sending a welcoming message to English-speaking Latino citizens who might otherwise stay on the sidelines of American politics. Clearly, the VRA works in ways that have not been fully anticipated.

Other factors may be at play here, and more research is necessary on the role of language in encouraging and facilitating participation among minority groups. Recent campaigns have conspicuously increased efforts to reach out to Spanish-speaking voters (e.g., Alvarez, 2012); yet, there is little research on how this outreach compares with traditional voter mobilization techniques. Studies evaluating whether campaigns employ similar strategies and effort levels for voter mobilization among English- and Spanish-speaking Latinos effectively inform Spanish speakers about resources and rights provided by the VRA and advocate on behalf of voters when their jurisdictions fail to provide the required accommodations would help identify the critical differences in Spanish speakers’ interaction with the political system that contributes to low registration rates.

The obstacles, either practical or psychological, to participation among weak English speakers have not yet been effectively addressed; we should aim to identify the factors that are impeding the success of the VRA’s provisions at increasing participation among this target group. Clearly, work remains on the critical goal of getting these citizens into the political process. At the same time, attention ought to be given to the various, and sometimes unanticipated, consequences that legislative solutions can have on the participation problem. The results presented here demonstrate that the VRA works in ways that few would have expected.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Michael Parkin (PhD, University of Minnesota) is an associate professor of politics at Oberlin College, where he teaches courses on American political behavior. His research focuses on American campaigns and elections and has appeared in journals such as The Journal of Politics, American Political Science Review, Political Research Quarterly, Politics and Policy, and Political Communication.

Frances Zlotnick is a PhD candidate in political science at Stanford University. Her research interests include the political ramifications of demographic change, political communication, and representation.

References

Alvarez

(2012, November 27). For Latino groups, grass roots efforts paid off in higher number of voters. New York Times, p. 22.

Ancheta

A. N.

(1998). Race, rights and the Asian American experience. New Brunswick, NJ: Rutgers University Press.

Ansolabehere

Konisky

D. M.

(2006). The introduction of voter registration and its effect on voting. Political Analysis, 14, 83-100.

Bachman

L. F.

Palmer

A. S.

(1989). The construct validation of self-ratings of communicative language abilities. Language Testing, 6, 14-29.

Barreto

M. A.

Segura

G. M.

Woods

N. D.

(2004). The mobilizing effect of majority-minority districts on Latino turnout. American Political Science Review, 98, 65-75.

Bass

L. E.

Casper

L. M.

(2001). Impacting the political landscape: Who registers and votes among naturalized Americans? Political Behavior, 23, 103-130.

Calvo

M. A.

Rosenstone

S. J.

(1989). Hispanic political participation. San Antonio, TX: Southern Voter Research Institute.

de la Garza

R. O.

DeSipio

(1997). Save the baby, change the bathwater, and scrub the tub: Latino electoral participation after twenty years of Voting Rights Act coverage. In Garcia

F. C.

(Ed.), Pursuing power: Latinos and the political system (pp. 72-126). Southbend, IN: University of Notre Dame Press.

Department of Justice. (2012, March 28). About language minority voting rights. Retrieved from http://www.justice.gov/crt/about/vot/sec_203/activ_203.php

10.

DeSipio

(1996). Making citizens or good citizens? Naturalization as a predictor of organizational and electoral behavior among Latino immigrants. Hispanic Journal of Behavioral Sciences, 18, 194-213.

11.

Fraga

(2010, August). The (surprising) short-term impact of the language provisions of the Voting Rights Act. Paper presented at the annual meeting of American Political Science Association, Washington, DC.

12.

Gaouette

(2006, May 10). Latinos boost U.S. population. Los Angeles Times, p. A4

13.

Hopkins

(2011). Translating into votes: The electoral impacts of Spanish-language ballots. American Journal of Political Science, 55, 814-830.

14.

Jackson

R. A.

(2003). Differential influences on Latino electoral participation. Political Behavior, 25, 339-366.

15.

Jamieson

Shin

H. B.

Day

(March 1, 2002). Voting and registration in the election of November 2000. Retrieved from http://www.census.gov/prod/2002pubs/p20-542.pdf

16.

Jones-Correa

(2005). Language provisions under the Voting Rights Act: How effective are they? Social Science Quarterly, 86, 549-564.

17.

Jones-Correa

Waismel-Manor

(2007). Verifying implementation of language provisions in the Voting Rights Act. In Henderson

(Ed.), Voting Rights Act reauthorization of 2006: Perspectives on democracy, participation, and power (pp. 161-182). Berkeley, CA: Berkeley Public Policy Press.

18.

Leighley

J. E.

(2001). Strength in numbers? The political mobilization of racial and ethnic minorities. Princeton, NJ: Princeton University Press.

19.

Leighley

J. E.

Vedlitz

(1999). Race, ethnicity, and political participation: Competing models and contrasting explanations. Journal of Politics, 61, 1092-1114.

20.

Lien

(2001). The making of Asian America through political participation. Philadelphia, PA: Temple University Press.

21.

Lutz

(2006). Spanish maintenance among English-speaking Latino youth: The role of individual and social characteristics. Social Forces, 84, 1417-1433.

22.

Oscarson

(1989). Self-assessment of language proficiency: Rationale and applications. Language Testing, 6, 1-13.

23.

Parkin

Zlotnick

(2011). English proficiency and Latino participation in U.S. elections. Politics & Policy, 39, 515-537.

24.

Ramakrishnan

S. K.

(2002). Voters from different shores: Electoral participation in immigrant America (Doctoral dissertation). Princeton University, NJ.

25.

Ramakrishnan

S. K.

(2005). Democracy in immigrant America: Changing demographics and political participation. Palo Alto, CA: Stanford University Press.

26.

Schildkraut

(2005). The rise and fall of political engagement among Latinos: The role of identity and perceptions of discrimination. Political Behavior, 27, 285-312.

27.

Shin

H. B.

Kominski

R. A.

(2010, March 1). Language use in the United States: 2007. Retrieved from http://www.census.gov/prod/2010pubs/acs-12.pdf

28.

Tam Cho

W. K

. (1999). Naturalization, socialization, participation: Immigrants and (non-) voting. Journal of Politics, 61, 1140-1155.

29.

Timpone

R. J.

(1998). Structure, behavior, and voter turnout in the United States. American Political Science Review, 92, 145-158.

30.

Tucker

J. T.

Espino

(2007). Government effectiveness and efficiency? The minority language assistance provisions of the VRA. Texas Journal on Civil Liberties & Civil Rights, 12, 163-232.

31.

Uhlaner

C. J.

Cain

B. E.

Kiewiet

D. R.

(1989). Political participation of ethnic minorities in the 1980s. Political Behavior, 11, 195-231.