Latino Attitudes About Spheres of Political Representation

Attitudes About Spheres of Representation

Although political scientists have long been interested in theoretical and empirical arguments about the congruence (or lack thereof) between the wishes of constituents and policy outputs, members of Congress do a lot more than write and pass legislation. They also help individual constituents in their interactions with government, such as helping a veteran figure out how to secure military benefits (casework); they help direct federal funds to their districts, such as for the construction of highways or hospitals (allocation); they evaluate the performance of government agencies in implementing public policy (oversight); and they pursue objectives that they deem to be in the national interest, even if their constituents are not concerned with (or disagree with) the particular objective in question (pursue national issues). Together, these activities make up the different spheres of representation (Griffin & Flavin, 2011; Mezey, 2008).

Only a handful of studies have examined whether the public deems policy congruence to be the most important sphere of political representation. In general, the delegate sphere does rise to the top, but there are noteworthy variations by race and ethnicity in how people prioritize these spheres. In 1978, the American National Election Study (ANES) asked respondents to prioritize activities of members of Congress, including helping people in the district, making sure the district gets its fair share of government money, and working on national legislation. Cain et al. (1987) found that Blacks “ranked the roles of representatives very differently” from Whites, prioritizing helping people over policy making (p. 42). The 1978 data, however, did not permit analysis of other minority groups and did not contain sufficient questions to allow Cain and colleagues to investigate the reasons for the different priorities of Whites and Blacks.

Almost 20 years later, Tate (2004) analyzed a version of this question from the 1996 National Black Election Study (NBES) and likewise found that national issues were a low priority for Blacks. The NBES allowed for a more in-depth investigation of the factors that shape how Blacks prioritize spheres of representation. Tate showed that having a sense of racial solidarity did not affect one’s prioritization but that Blacks with higher socio-economic status were more likely than other Blacks to prioritize working on national issues and less likely to prioritize casework. Employing an experimental approach, Harden (2011) found that Whites gave better evaluations than Blacks to a representative portrayed as pursuing policies that the constituency wants, while Blacks gave more favorable ratings to representatives portrayed as emphasizing casework and allocation. He was not able, however, to explore the reasons for these differences.

Finally, Griffin and Flavin (2011) extended the analysis of public opinion about spheres of representation to include Latinos. They found that Whites, Blacks, and Latinos all ranked “learning about district constituents’ opinions in order to better represent their views” as most important, but that Latinos and Blacks were more likely than Whites to say that allocation is most important (Griffin & Flavin, 2011, p. 522). Although they did not examine why Blacks and Latinos feel differently than Whites, they offered the provocative speculation that racial minorities might be less likely to prioritize policy spheres

in part because these groups have experienced a long history of unequal representation. There may be a vicious cycle in which disadvantaged groups are underrepresented, which leads these groups not to expect substantive representation, which leads to even worse representation. (Griffin & Flavin, 2011, p. 530)

Implicit in this suggestion is the notion that an awareness of group position in the arena of electoral politics and a heightened concern with the group might explain their finding. Determining whether such awareness and concern are the mechanisms that shape how Latinos prioritize spheres of representation is the goal of this study. ³

Beyond Dummy Variables: Group Identification, Linked Fate, and Discrimination

At the heart of the assumption that perceptions of group location matter is the idea that people rely on connections to groups to help them make sense of politics. Groups serve as valuable information shortcuts that provide cues for processing complex political information (Converse, 1964; Dawson, 1994; Kinder & Herzog, 1993; Walsh, 2004). Among traditionally disadvantaged groups in American society, three aspects of group membership have been shown to be particularly influential in shaping political attitudes and behaviors: group identification, linked fate—which is the view that one’s own fate is inextricably tied to the fate of the group (Dawson, 1994), and perceptions of discrimination toward the group. When scholars find racial differences in political preferences (via dummy variables for race), they assume that these three factors are at work, but they often lack the measures needed for further investigation. When such measures are present, these factors have been shown to shape a range of political attitudes and behaviors among Latinos (and among other ethnic minorities), including support for descriptive representation, interest in politics, political participation, empowerment, and trust in government (Manzano & Sanchez, 2010; Masuoka, 2008; Sanchez, 2006; Schildkraut, 2011, 2013b; Stokes-Brown, 2006; Wallace, 2014). ⁴

But do psychological connections to the group and concerns about its position in the socio-political hierarchy affect how Latinos feel about spheres of representation? Griffin and Flavin suggest that an awareness of group position could make Latinos feel that their preferred policies are unlikely to prevail, which leads them to prefer alternative forms of representation. If this causal mechanism is at work, then Latinos with a greater sense of ethnic identity, who perceive discrimination, and who have a sense of linked fate might be less inclined to prioritize spheres that are oriented toward policy, oversight, and the national interest and more inclined to prioritize spheres that are more narrowly focused, such as casework and allocation. The speculation posed by Griffin and Flavin, along with research on how group membership affects political attitudes and behaviors, thus leads to the following hypothesis:

Data and Measures

Data come from the American Identity and Representation Survey (AIRS), which was commissioned by the author and conducted over the Internet in October of 2012 by Knowledge Networks. The survey has a nationally representative sample of 434 Latinos, along with 448 White, non-Hispanic respondents. A Spanish version of the survey was available and was used by 47.8% of the Latino respondents. ⁵ Fifty-eight percent of the Latino respondents were born outside the United States. Of those, 45% were U.S. citizens. ⁶

Findings

Before examining the determinants of how Latinos prioritize spheres of representation, it is instructive to examine how the spheres of representation were ranked. Figure 1 shows the percentage of respondents that ranked each sphere of representation as most important. Like Griffin and Flavin, I find that pursuing policies in line with constituent wishes (delegate) comes out first, but unlike in their study, it is not selected first by a majority. Only a plurality (31.2%) said that the delegate sphere is most important. Working on national issues was next, with 23.5% of respondents rating it as the most important sphere. In the middle was oversight, with 20% of respondents ranking it most important. Only 16.1% of Latinos said that allocation is the most important sphere of representation, and few respondents prioritized casework (9.2%). Indeed, there was wide agreement that casework is the least important sphere. Figure 2 shows the full range of rankings given to each sphere, and it shows that over 50% of the respondents ranked casework last.

OPEN IN VIEWER

Figure 1.

Most important sphere of representation.

OPEN IN VIEWER

Figure 2.

How Latinos ranked the importance of five spheres of representation.

As noted earlier, the survey also included 448 White non-Hispanic respondents, whose preferences can be compared with those of Latinos. ¹² Figure 3 shows the spheres that each group considered the most important. It shows that a plurality of both groups selected serving as a delegate as the most important part of a representative’s job. Despite this similarity, there are differences among the groups that are in line with findings of existing scholarship. Notably, Latinos were more likely than Whites to rank casework most important, while Whites were more likely than Latinos to rank working on national issues and serving as a delegate most important. Latinos were more likely than Whites to prioritize oversight as well. Whites and Latinos ranked allocation similarly, with 17.8% of Whites and 16.1% of Latinos ranking it most important. ¹³

OPEN IN VIEWER

Figure 3.

Most important sphere, Whites and Latinos.

Differences in how Whites and Latinos rank the importance of allocation emerge, however, when examining the sphere of representation that respondents considered to be the second most important, as Figure 4 shows. The biggest difference between the two groups in this figure concerns allocation: 23.5% of Latinos say that allocation is the second most important aspect of a representative’s job, compared with only 14% of Whites. ¹⁴ In sum, the general trend across Figures 3 and 4 is that serving as a delegate remains a highly valued activity, but Latinos are more likely to value the more narrowly targeted realms of casework and allocation than Whites, while Whites are more likely to value the policy-oriented realms of serving as a delegate and pursuing the national interest.

OPEN IN VIEWER

Figure 4.

Second most important sphere, Whites and Latinos.

Having replicated the general trend established in prior research, the task now is to determine whether group-related concerns account for the ways in which Latinos prioritize the different spheres of representation. Do ethnic group identity, linked fate, and perceptions of discrimination lead some Latinos to prioritize spheres that target specific groups and individuals (casework and allocation), and does the lack of these group-based concerns lead them to prioritize policy-oriented spheres, such as oversight, being a delegate, and working toward the national interest? To examine this question, I ran five ordered probit models, one for each sphere of representation. The dependent variable in each model is the rank that respondents gave to that sphere, recoded to run from 0 to 1, where 1 means the respondent ranked that sphere as most important, and 0 means it was ranked least important. The results appear in Table 1.

OPEN IN VIEWER

Table 1.

Predicting How Latinos Prioritize Spheres of Representation.

	Casework	Allocation	Oversight	Delegate	National issues
Identify primarily as American	0.19	−0.29*	−0.44**	0.15	0.37**
	(0.21)	(0.18)	(0.17)	(0.17)	(0.21)
Linked fate	0.13	−0.22	0.07	−0.11	0.15
	(0.20)	(0.16)	(0.14)	(0.16)	(0.17)
Group discrimination	0.63**	0.11	−0.62**	−0.14	0.05
	(0.29)	(0.30)	(0.31)	(0.31)	(0.31)
Individual discrimination	−0.27	0.03	0.10	0.28	−0.16
	(0.20)	(0.15)	(0.17)	(0.19)	(0.16)
Born in the United States	−0.23	0.29	0.34	0.08	−0.54**
	(0.23)	(0.18)	(0.20)	(0.20)	(0.18)
Took survey in English	0.72**	0.18	−0.37	−0.52**	0.13
	(0.24)	(0.20)	(0.24)	(0.24)	(0.20)
Education	−1.05**	0.16	0.17	−0.10	0.70*
	(0.44)	(0.42)	(0.39)	(0.35)	(0.36)
Political knowledge	−0.10	−0.04	−0.54**	0.74**	−0.11
	(0.25)	(0.25)	(0.39)	(0.24)	(0.26)
Party (Dem → Rep)	0.18	−0.61**	0.23	0.04	0.13
	(0.25)	(0.24)	(0.23)	(0.25)	(0.24)
Male	−0.21	0.01	0.14	−0.24	0.35**
	(0.16)	(0.14)	(0.14)	(0.15)	(0.14)
Same party as representative	−0.27*	0.02	−0.04	0.32**	−0.07
	(0.16)	(0.15)	(0.14)	(0.15)	(0.15)
Latino representative	0.12	0.34**	−0.27*	−0.03	−0.15
	(0.18)	(0.16)	(0.16)	(0.15)	(0.15)
Cut point 1	−0.05	−1.11	−1.18	−1.40	−0.59
Cut point 2	0.38	−0.20	−1.04	−0.61	0.02
Cut point 3	1.03	0.29	−0.42	−0.08	0.58
Cut point 4	1.35	1.06	0.33	0.50	1.28
χ2	20.09	23.09	24.09	27.67	27.74
n	374	377	381	380	377

Source. American Identity and Representation Survey (2012).

Note. Weighted results. Robust standard errors in parentheses. Dependent variables range from 0 to 1, where 1 means respondent ranked that sphere most important and 0 means respondent ranked it fifth.

*

p < .1. **p < .05.

The results validate some of the speculation advanced in earlier research, particularly with regard to group identification and perceptions of discrimination. Regarding group identification, the first row of results shows that people who identify primarily as an American instead of as a member of their racial or ethnic group are more likely to give a high ranking to “working in Congress on bills concerning national issues.” Conversely, people who identify primarily with their racial or ethnic group are more likely to give a high ranking to “making sure the district gets its fair share of government money and projects.” Group identification did not affect the ranking given to casework or serving as a delegate. Unexpectedly, identifying with the racial or ethnic group instead of as American also increased the ranking given to oversight. ¹⁵

With respect to perceptions of discrimination against Latinos and one’s national origin group, people who perceive discrimination are more likely than others to give a high ranking to casework, while people who do not perceive discrimination are more likely than others to give a high ranking to oversight. The final measure of group concern studied here, linked fate, is insignificant in all of the models, as are perceptions of individual-level discrimination.

Overall, the results confirm that there is some basis for the assumption that racial and ethnic differences in how people prioritize spheres of representation are derived from psychological connections to the group and concern about the group’s treatment in society. ¹⁶ However, other factors included in the model also have important roles to play. For example, Table 1 shows that more acculturation leads to more parochial concerns. Foreign-born respondents were more likely than U.S.-born respondents to rank working on national issues highly, and people who took the survey in Spanish were less likely than people who took it in English to rank casework highly. They were also more concerned with the pursuit of policies that align with the constituency. In sum, the less acculturated respondents were more concerned with policy-oriented spheres and less concerned with casework.

This effect for acculturation was not anticipated, though I have identified two possible explanations for it. The first is that these results fit research showing that less acculturated Latinos tend to have a more optimistic image of American government than more acculturated Latinos or native-born Whites and an appreciation for the rule of law, especially when compared with politics in their home countries (Michelson, 2003; Pearson & Citrin, 2006; Schildkraut, 2013a; Weaver, 2003). It is also true that Latinos who are less acculturated are more likely than other Latinos to be concerned about their group’s position (Sanchez & Masuoka, 2010), but once such concerns are controlled for (as in Table 1), acculturation directs attention away from the policy-focused spheres of representation and toward casework. The second possibility is that foreign-born Latinos are more highly attuned to national debates about comprehensive immigration reform and to immigration policy at the state and local levels than are other Latinos. Given that the stakes of immigration policy are so high for this subset of Latinos and given the level of national debate about immigration in 2012, their heightened sensitivity to the issue could crowd out concern for other spheres of representation. Unfortunately, neither of these possibilities can be investigated further with the available data.

The findings with respect to education, political knowledge, partisanship, and gender align with expectations. Higher levels of education are associated with higher rankings for pursuing national issues and lower rankings for casework, while political knowledge increases the importance placed on representing the views of the constituency (though it also decreases the importance placed on oversight). The conventional wisdom regarding differences between the two major parties is on display in Table 1, with Republican Latinos ranking allocation lower than Democratic Latinos. Finally, in line with Harden’s (2011) work, men were more likely than women to rank working on national issues highly.

Table 1 also indicates that the party and race of one’s representative affect how Latinos rank spheres of representation. Sharing the partisanship of one’s representative works in the expected direction: Latinos with a representative of the same party are more likely than other Latinos to rank serving as a delegate highly. Conversely, Latinos who do not share their representative’s party give higher rankings to casework. Regarding descriptive representation, Latinos with a Latino representative are more likely than other Latinos to rank allocation highly, a finding that fits with research showing that descriptive representation leads to more effective allocation of government resources to minority communities (Grose, 2011) and to a greater likelihood of thinking that one’s representative has helped the district (Tate, 2004).

Discussion and Conclusion

These findings contribute to scholarship on racial and ethnic politics in the United States and on political representation. They replicate findings in earlier research by demonstrating differences in the importance Whites and Latinos place on the various activities that members of Congress do as part of their jobs, and it advances that research by examining the factors that lead some Latinos to feel that policy-oriented activities are most important and lead others to feel that allocation or casework are most important. Prior research suggested that connections to the group and attitudes about group treatment lead minorities to feel that their policy concerns are unlikely to get addressed, which leads them to evaluate other forms of representation more highly. The results presented here partially confirm this speculation: Thinking of oneself primarily as a member of an ethnic group instead of as an American increases the importance placed on allocation and decreases the importance placed on pursuing national issues, and perceiving discrimination against one’s ethnic group increases the importance placed on casework and decreases the importance placed on oversight.

I found that other factors associated with the Latino experience in the United States also affect how respondents prioritize spheres of representation. When the effects of ethnic identification and perceptions of discrimination are controlled, acculturation generally contributes to more parochial concerns while Latinos who are less acculturated are more likely to consider policy-oriented representation most important. This pattern is plausibly attributed to either (a) the fact that newcomers to American society have been shown to have greater appreciation for the American political system than others or (b) the prevalence of national and statewide debates over immigration policy. Descriptive representation also matters, with co-ethnic representation leading to greater support for allocation.

Taken as a whole, the results offer a broader understanding of how Latinos feel about representation than earlier research has provided. The speculation that ethnic attachments and perceptions of discrimination matter is correct but they hardly tell the whole story, and linked fate was insignificant. Adding acculturation and descriptive representation to the analysis leads to a more nuanced understanding of how being Latino in the United States affects evaluations of the nation’s representative democracy. Concerns about the group and having co-ethnic representation promote more narrowly focused preferences, but so does acculturation. Policy-oriented representation is preferred not only by those who psychologically distance themselves from ethnic concerns but also by those who are more closely connected to the immigrant experience.

Despite these contributions, limitations of the present analysis exist. First, the sample size precludes analysis by national origin group. Second, it relies on responses to closed-ended survey questions—qualitative approaches that allow respondents to put into their own words the type of representation they value most and why would be a valuable complement to this line of research. Such an approach would help to identify the mechanism leading less acculturated Latinos to be more likely than others to prioritize policy-oriented representation. Third, it accounts for attitudes toward representation at only one point in time. Ideally, future surveys of Latino political attitudes will include questions about spheres of representation as well.

Scholars and pundits have claimed that failing to enact policies that Latinos desire, such as immigration reform, will spell electoral doom. That may very well be the case, especially for the plurality of Latinos who say that serving as a delegate is a representative’s most important task. But I found that 56% of Latinos say that casework and allocation are either the first or second most important spheres of representation, and they do so because they identify primarily with their group, think the group is discriminated against, and have a Latino representative. There are two prescriptive conclusions one could draw from these findings. In the short run, representatives who provide representation in the form of casework and allocation to their Latino constituents could be more effective in meeting constituent needs and desires than those representatives who focus more on policy. In the long run, however, creating a socio-political environment in the United States that is more welcoming to Latinos could help them feel more American and ameliorate their perceptions of discrimination, which, in turn, could make them feel that their policy concerns might actually get addressed. If the delegate sphere of representation is normatively paramount, as some scholars have argued (e.g., Brunell, 2008; Rehfeld, 2005), then the long run is the route to pursue.