The Big Tent Effect

Abstract

In this study, we assess whether Blacks and/or Latinos are more likely to identify with political parties that nominate a U.S. House candidate who shares their race/ethnicity using the 2010 Cooperative Congressional Election Study (CCES). Our results indicate that Blacks are more likely to identify with both the Democratic and Republican Party when they nominate successful Black candidates for the House of Representatives. To assess the temporal order of the relationship, we examine the differences in Black Democratic partisanship before and after Obama’s election to the White House and changes in Republican partisanship among Blacks in districts before and after the nomination of a successful Black Republican candidate. In combination, we find that both political parties can make gains in the Black community through the nomination of co-racial candidates. While descriptive candidates consistently influence Black partisanship, we find that Latino partisanship is not significantly affected by the presence of co-ethnic candidates.

Keywords

Black politics Latino politics partisanship descriptive representation Congress

The 2012 presidential election demonstrated the growing importance of minority voters in American political elections. In spite of receiving less than 40% of the White vote, Obama was able to win re-election through building a coalition of Black, Latino, and Asian American voters. The results of the 2012 election, in some ways, are emblematic of the future of American politics. By 2050, the United States will be a majority–minority country, and to succeed, both major political parties will have to attract the support of the minority electorate.

Faced with this challenge, both parties are exploring different options to enhance their standing in minority communities. One strategy that is becoming increasingly common among both the Democratic and Republican Parties is to highlight and promote their party’s racial/ethnic diversity. Traditionally, this has been done through the strategic appointment of minorities to leadership and cabinet-level positions (Philpot, 2007). Beyond appointing African Americans and Latinos to strategic posts, both parties are also ramping up their efforts to recruit minority candidates. For example, former Black U.S. House Representative J. C. Watts has created a non-profit charged with targeting and promoting potential minority candidates and congressional staffers for the Republican Party.¹ Not to be outdone, the Democratic Congressional Campaign Committee has also created a Latino Council, which focuses on increasing the number of Democratic Latino candidates for Congress.²

While both parties are making a concerted effort to diversify their candidate pool in hopes of attracting Black and Latino support, the efficacy of this strategy has not yet been empirically studied. Previous research has examined the role minority candidates play in increasing turnout, efficacy, and numerous other political attitudes and behaviors, but less is known about how the presence of Black and Latino candidates influences political partisanship (Barreto, 2010; Bobo & Gilliam, 1990; Michelson, 2000; Sanchez & Morin, 2011; Tate, 2003).

In this study, we assess whether Blacks and/or Latinos are more likely to identify with political parties that nominate a U.S. House candidate who shares their race/ethnicity and whether this effect is contingent on the success of the candidate.³ To address this question, we begin by discussing how political partisanship is formed and how the presence of underrepresented candidates may shape party identification for minority voters. We then test our hypothesis using the 2010 Cooperative Congressional Election Study (CCES). Among other things, the 2010 CCES is useful for our purposes because it asks respondents to identify the race/ethnicity of the congressional candidates in their district.

Our findings indicate that Blacks are more likely to identify with the Democratic Party when they nominate a Black candidate, regardless of the candidate’s performance. Our results also indicate that the nomination of a successful Black Republican candidate can improve the Party’s standing with the Black community. In contrast to Blacks, we do not find Latino partisanship to be significantly affected by the presence of co-ethnic House candidates.

To assess the temporal order of the relationship between the Democratic and Republican Party’s nomination of Black candidates and transformations in Black partisanship, we examine the difference in Black Democratic partisanship before and after Obama’s election to the White House. We also explore differences in Republican partisanship among Black voters in districts before and after the nomination of a successful Black Republican candidate. In combination, we find support for the claim that both political parties can make gains in the Black community through the nomination of co-racial candidates.

Race, Ethnicity, and Partisanship in the United States

Traditional explanations of party identification assert that individuals inherit their partisan loyalties from their parents, which stick with them over the course of their lives and provide a lens through which they understand the political world (Campbell, Converse, Miller, & Stokes, 1960). Green, Palmquist, and Schickler (2002) show that party attachment is so stable that even when political and economic evaluations change, party identification does not.

In contrast, others contend that partisan development and change is a much more dynamic process, making it less stable over the course of one’s life (Fiorina, 1981; Franklin & Jackson, 1983; Page & Jones, 1979). Allsop and Weisberg (1988) are even able to show that some individuals change their partisanship over the course of one election cycle. Previous research has demonstrated that partisanship is susceptible to change when the policy positions of individuals’ preferred political party become incongruent with their own (Carsey & Layman, 2006; Fiorina, 1981; Franklin, 1992), and when individuals’ residential context and social networks are unsupportive of their party identification (Lyons, 2011). Most pertinent to our study, other research demonstrates that short-term changes in partisanship can be tied to the personalities and evaluations of a party’s candidates, and alterations in perceptions of the party’s image (Brody & Rothenberg, 1988; Philpot, 2007; Rapoport, 1997).

Along these lines, a growing body of work suggests that minority partisanship in the United States is far from settled. Surveys show that Latinos are reluctant to identify with either political party, reporting higher levels of non-partisanship than Whites and Blacks (see Hajnal & Lee, 2011). While African Americans have clear behavioral preferences for the Democratic Party, they are far from homogeneous in their identification with the Party (Frymer, 1999; Hajnal & Lee, 2011). Rather than stemming from policy disagreements, or a growing conservatism and economic diversity, the Democratic Party’s inattention and lack of effort to pursue Black issues has resulted in a decline in Black identification with the Party (Bositis, 2010; Hajnal & Lee, 2011; Tate, 1991).

Acknowledging the importance of racial identity and group consciousness, a number of scholars have argued that minority partisanship is reflective of Blacks’ and Latinos’ perception that the Democratic Party works harder for their interests than does the Republican Party (Cain, Kiewiet, & Uhlaner, 1991; DeSipio, 1996; Philpot, 2007). Much of the recent literature has reiterated this point by demonstrating that the amount of effort put forth by a political party to appeal to Blacks and Latinos is a strong predictor of its partisan identification (Bositis, 2010; Evans, Franco, Polinard, Wenzel, & Wrinkle, 2012; Frymer, 1999; Hajnal & Lee, 2011). Hajnal and Lee (2011), for example, demonstrate that Blacks are much more likely to identify as Democrats when they feel that the Party works hard to improve the lives of African Americans. Along the same lines, Evans et al. (2012) find that while Latinos are largely non-partisan, the belief that either the Democratic or Republican Party is working for Latino interests increases party identification.

Beyond perceptions of the Democratic Party working for the interest of Blacks and Latinos, some have argued that the Republican Party’s poor showing with minority voters is due to perceptions of Republican hostility toward minority political interests (Cain et al., 1991; King, 2006; Nuño, 2007; Philpot, 2007). Candidates from the Republican Party have at times aired racially divisive advertisements, invoking racial conservatism (Mendelberg, 2001). For instance, in his 1990 senatorial campaign, Republican Jesse Helms ran his infamous “White hands” advertisement, which showed the hands of a White man crumpling up a rejection letter from an employer because the job was given to “a minority.” The advertisement attempted to incite fears of racial favoritism in hopes that it would decrease support for Helms’ Black opponent Democrat Harvey Gantt. More recently, Republican U.S. Senate candidate Sharron Angle ran an ad that depicted recent White graduates celebrating while a group of Latinos, portrayed as illegal immigrants and gang members, looked on grimacing.

In addition, the Republican Party has long been perceived as being inhospitable to minority voters because of its support of policies viewed as hostile toward minorities, such as cutting bilingual education programs or limiting social welfare programs, as well as its perceived lack of attentiveness to the needs of predominantly minority communities, as some deemed to be evident in the Bush Administration’s response to Hurricane Katrina (Coffin, 2003; Mendelberg, 2001; Nuño, 2007; Philpot, 2007; White, Philpot, Wylie, & McGowen, 2007). King (2006) and Philpot (2007), in their studies of Black partisanship, provide evidence that even Blacks who may feel comfortable with some of the Republican Party’s policies do not join the Party because of negative perceptions of the Party’s relationship with the Black community. Bowler, Nicholson, and Segura (2006) find that the Republican Party’s repeated support for racially charged issues benefited the Democratic Party by reversing a trend of increasing Latino support for the GOP.

Descriptive Representation and Partisanship

The battle for minority support, in many ways, lies in the ability of Democrats and Republicans to get Blacks and Latinos to believe that their party takes their concerns seriously and will better represent their interests. By nominating co-racial/ethnic candidates, both parties may be able to signal their commitment to minority inclusion, while attempting to extend the bonds of trust Blacks and Latinos feel with their descriptive representative to the party as a whole.

Research shows that minorities perceive descriptive representatives as being more responsive to their needs (Bobo & Gilliam, 1990; Gurin, Hatchett, & Jackson, 1989; Mansbridge, 1999). In his study of representation in Congress, Grose (2011) finds that Black representatives not only were stronger proponents of Black political interest but also put forth more effort to keep in touch with Black constituents (see also Canon, 1999, and Tate, 2003). Minta (2011) also finds that both Black and Latino legislators are better at providing resources to co-racial/ethnic constituencies. Seeing that party perceptions are based in part on candidate evaluations (Brody & Rothenberg, 1988; Philpot, 2007; Rapoport, 1997), Blacks and Latinos may view a political party as more closely aligned with and more responsive to their interests when a descriptively representative candidate is the party’s standard bearer.

Beyond perceptions of attentiveness, the presence of minority candidates may improve perceptions of each political party, which in turn may influence partisanship. Philpot (2007) and Nuño (2007) both identify the lack of minority representation as one of the reasons Blacks and Latinos have been impervious to appeals from the Republican Party. Through several interviews, Philpot (2007) shows that Blacks perceive the Republican Party and its members as being too different from themselves to be considered a viable alternative to the Democratic Party. In spite of highlighting growing diversity within the Party, Philpot (2007) finds that putting Blacks in prime positions at the National Convention did not influence Black attitudes about the Republican Party. The ineffectiveness of this strategy, Philpot (2007) contends, rests in Blacks’ perceptions that the Party was making merely cosmetic changes to its image rather than genuinely attempting to diversify the Party and work for minority interests.

In contrast to Philpot’s (2007) findings about Republicans, Tate (1991) finds evidence suggesting that descriptively representative candidates can influence Black support for the Democratic Party. She finds that frustration with the Democratic Party over Jesse Jackson’s prolonged struggle with party leadership and his second failed bid for the Party’s nomination may have been the cause of lower Black turnout in the 1988 presidential election. Thus, despite policy congruence between Blacks and Democratic candidates, the lack of diversity in important positions within the Party may diminish Black support.

Along the same lines as Tate (1991), Nuño (2007) and Barreto and Nuño (2011) find that co-ethnic Republican contact increases positive feelings about Republican leadership and policies. Both of these studies argue that co-ethnic contact makes the Republican Party seem more diverse, less hostile to Latino interests, and as more welcoming to Latinos. These results may indicate that some Latinos would be open to identifying with the Republican Party, but a lack of ethnic diversity within the Party has made them hesitant to join (Nuño, 2007).

While previous research demonstrates that “cosmetic” changes have mixed effects on individual party images, nominating descriptive candidates may have a more substantial effect on minority partisanship for several reasons. First, nominating minority political candidates in many cases carries with it more risk, as parties could lose if they nominate the wrong candidate. This action signals to minorities that parties are serious about being more inclusive. Second, considering that Blacks and Latinos view co-racial/ethnic candidates as better representative of their interests, nominating Black and Latino candidates may signal that the party is serious about furthering the goals of minority communities. Moreover, beyond altering issue positions, which parties are hesitant to do, nominating Black and Latino candidates may help political parties more successfully demonstrate that their party platform benefits minorities. This may be particularly true if the party’s nominee succeeds in his or her bid for elected office.

The Conditional Impact of Descriptive Candidates and Minority Partisanship

Although we anticipate the nomination of descriptively representative candidates to influence minority partisanship, we do not expect these effects to be uniform across all candidates, racial/ethnic groups, and political parties. Black and Latino partisanship should be strongest when a co-racial/ethnic candidate is present and the ideological proximity between the racial/ethnic group and the party is close. Given that minority voters tend to have more policy agreements and are ideologically closer with the Democratic Party, the effects of descriptive candidates on partisanship should be greater where a co-racial/ethnic Democratic candidate is nominated than in districts where a co-racial/ethnic Republican is on the ballot (Dawson, 1994; DeSipio, 1996; Hajnal & Lee, 2011).

Current research shows that while Blacks and Latinos hold policy preferences that are more closely aligned with the Democratic Party, they at times feel like a captured electorate whose loyalty to the Party is taken for granted (Frymer, 1999; Hajnal & Lee, 2011). The Democratic Party may be able to address this concern by nominating co-ethnic candidates to political office. Such outreach demonstrates that the Party is actively seeking to represent Blacks and Latinos, which may help alleviate concerns that the Party is ignoring the minority community. Through the diversification of the Party, we may observe an increase in Democratic identification among these underrepresented groups.

In contrast, due to its history of racial animus and hostility, when the Republican Party nominates co-racial/ethnic candidates, particularly non-viable minority candidates, it is viewed as more of a surface-level change without any real substantive effect, and thus met with hesitance. Therefore, even though the low levels of minority Republican identification provide significant room for growth, Republicans must make stronger appeals to overcome negative party perceptions. As Philpot (2007) found, cosmetic changes without shifts in the Republican Party’s platform have only a mild effect on Blacks’ perception of the Republican Party. While we expect that descriptive Republican candidates should improve their Party’s standing with minority voters, we do not expect Black and Latino partisanship to be affected as much or as consistently when a co-racial/ethnic Republican is nominated than when a co-racial/ethnic Democrat is on the ballot. Given this, our first hypothesis states,

Hypothesis 1: All else being equal, Black and Latino partisanship will be more responsive to Democratic co-racial/ethnic candidates than Republican descriptive candidates.

We also expect that the candidate’s level of success will be an important moderating factor between the nomination of a descriptive candidate and the partisanship of the electorate. The nomination of a candidate with a low probability of success does little to show that the party is exerting a significant effort to appeal to minority communities as it carries little risk to nominate a “sacrificial lamb” Black or Latino candidate in a context where the party is sure to lose anyway. Moreover, unsuccessful minority candidates may do little to improve the substantive representation of their co-racial/ethnic constituents in government.

The candidate’s performance may be particularly important for the Republican Party. Considering the findings of Philpot (2007) that cosmetic or surface-level changes have little impact on increasing Blacks’ ideological closeness with the Republican Party, we should expect that the nomination of an unsuccessful minority Republican candidate will do little to strengthen minority partisanship in that district. However, when the Republican Party nominates competitive and successful minority candidates, some may perceive this action as the Party making a genuine effort to appeal to minority voters. Thus, our second hypothesis states,

Hypothesis 2: We expect that the influence of co-racial/ethnic candidacies on partisanship is moderated by the candidate’s level of success.

Data and Method

We use the 2010 Cooperative Congressional Election Study to test whether the presence of minority U.S. House candidates improves the Democratic or Republican Party’s standing with minority voters. This is an ideal data set to assess our hypothesis for several reasons. First, the 2010 CCES has a large sample size (n > 50,000) and includes a substantial number of both Black and Latino respondents. The large sample of minorities in a diverse set of congressional districts provides the opportunity to examine whether the perceived presence of a minority candidate increases minority identification with either of the major political parties in the United States. Second, the data set is unique in that it asks respondents to identify the race of the congressional candidates in their districts. Thus, we can be better assured that respondents are aware (or at least think they are aware) of whether the Republican or Democratic U.S. House candidate in their district is Black or Latino.⁴ As a result, our treatment is not based on the assumption that respondents know the race of the U.S. House candidates; the data set provides us an explicit measure of their knowledge about the candidate’s racial/ethnic identity.⁵ Finally, 2010 was a notable election year because of the high number of Black and Latino Republican U.S. House candidates.

Our dependent variable measures political partisanship on a 7-point scale that ranges from 1 (strong Democrat) to 4 (Independent) to 7 (strong Republican)⁶ and was measured after the 2010 midterm election occurred.⁷ Our independent variables of interest measure whether Black or Latino respondents resided in a congressional district with a successful/unsuccessful perceived co-racial/ethnic candidate. We use a two-step process to create these categories. First, we measured the perceived race/ethnicity of the U.S. House Democratic and Republican candidates utilizing a question in the 2010 CCES that asks, “What race or ethnicity is Democrat/Republican House Candidate [insert name]?” We then used official election results to assess whether the Democratic or Republican candidate succeeded or failed in the respondent’s district. Using these measures, we created several dummy variables that assess whether the respondent resided in a district with a perceived Black or Latino U.S. House candidate who either won or lost in his or her election.

For example, for African American candidates, we used the 2010 CCES question about the race of the candidate and the election results to create four dichotomous variables for (a) respondents in districts with a successful Democratic Black U.S. House candidate, (b) respondents in districts with an unsuccessful Democratic Black U.S. House candidate, (c) respondents in districts with a successful Republican Black U.S. House candidate, and (d) respondents in districts with an unsuccessful Republican Black U.S. House candidate. We then replicated this process to create an additional four categories for Latino candidates. In all, we created eight dichotomous independent variables that assess whether individuals lived in districts with successful or unsuccessful minority candidates (1) or whether respondents did not identify the congressional candidate in their district as Black (0).

We then use these variables to create eight interaction terms that capture Black and Latino respondents who have the opportunity to vote for a perceived co-racial/ethnic U.S. House candidate. For example, we created four interaction terms for Latinos (e.g., Latino × Latino Republican U.S. House candidate [won], Latino × Latino Democrat U.S. House candidate [won], Latino × Latino Republican U.S. House candidate [loss], Latino × Latino Democrat U.S. House candidate [loss]) and an additional four interaction terms for Blacks.

To isolate the influence of the perceived race of the candidate on measures of partisanship, we control for several factors. First, we control for the gender, age, and socio-economic status of the respondents. Previous research has demonstrated that these factors are closely tied with partisanship, particularly for African Americans (Bositis, 2010; DeSipio, 1996). We then control for the ideology of the respondent that is measured on a 7-point scale and ranges from 1 (extremely liberal) to 7 (extremely conservative). Ideology has long been considered a primary determinant of political partisanship (Campbell et al., 1960).

To ensure that the respondent’s partisan context is not influencing our results, we control for whether the respondent lived in the South and the respondent’s district’s Cook’s Partisan Voting Index (PVI) score. The PVI measures whether the district the respondent resides in leans toward Republican or Democratic Party and by how much. The models also include controls for whether the candidate was campaigning for elected office in the South, the percentage of the Black/Latino population in each congressional district, and whether the Republican candidates won (1) or lost (0) in the district.

Finally, for the Latino candidate model, we control for the number of generations individuals and their ancestors have lived in the United States. This ranges from first to third generation. Previous research demonstrates that Latinos who are relative newcomers to the United States generally have lower levels of party identification (Cain et al., 1991; DeSipio, 1996; Hajnal & Lee, 2011). In this model, we also include a control for Latino respondents who resided in Florida. Given that most Cuban candidates and voters are already Republican and reside in the state of Florida, it is important to account for this pre-existing relationship.

Results

Figure 1A shows the partisan makeup for Blacks in five different conditions: Blacks in districts with Black Democratic U.S. House candidates who won their election (N = 1,362), Blacks in districts with Black Democratic U.S. House candidates who lost their election (N = 94), Blacks in districts with successful Black Republican candidates (N = 54), Blacks in districts with unsuccessful Black Republican candidates (N = 251), and Blacks in districts where U.S. House candidates of both parties were perceived to be White (N = 4,861). The results presented in Figure 1A provide initial support for our hypothesis that descriptive representation can influence partisanship for both political parties. However, this is partially contingent on the minority candidate’s level of success.

Figure 1.

(A) Black and (B) Latino partisanship with different descriptive candidates.

For example, Black Democrats comprised 83% of the population in districts where a co-racial Democratic candidate succeeds. In comparison, Black Democrats only comprised about 76% of the Black population in districts with no Black Democratic candidate. Thus, Black Democrats comprise 7% more of the population in districts where a Black Democratic candidate has succeeded than their counterparts. Moreover, a lot fewer of these descriptively represented respondents identify as political independents or as Republicans. Similarly, the composition of Black Republicans in districts where the successful Republican candidate is perceived to be Black is almost twice as large as the proportion of Black Republicans in districts with no African American U.S. House candidates. In combination, when parties nominate successful minority candidates, they appear to receive an increase in support from the Black community. In contrast, the nomination of unsuccessful Black candidates does not appear to influence Black partisanship for either party.

Resembling the pattern found in Figure 1A, Figure 1B indicates that both political parties improve their standings with Latino voters by nominating co-ethnic candidates. Latino Democrats who have a chance to support a successful co-ethnic Democratic candidate (N = 384) make up about 55% of the Latino population in these districts. In comparison, Latino Democrats in districts with only White major-party candidates (N = 4,514) make up only 44% of the population. Moreover, Latino Republicans make up a greater portion of the population in districts with successful Latino Republican candidates (N = 259) than Latino Republicans in districts with no Latino candidates. However, the Republican Party does not appear to make gains with the Latino community when it nominates unsuccessful Latino candidates (N = 174). Overall, the results suggest that with the right candidates, both the Democratic and Republican Parties may have a lot to gain through the nomination of Blacks and Latinos. However, to ensure that these results are not artifacts of other factors, we turn to regression analysis.

Figure 2 presents the partial results of two separate ordinary least squares (OLS) regression analyses predicting partisanship on a 7-point scale. The two models are separated by the race/ethnicity of the candidate. Model A examines the relationship between Blacks and their descriptive candidates, and Model B examines the effect of co-ethnic candidates on Latino partisanship. Negative values in Figures 2A and 2B indicate closeness to the Democratic Party, whereas positive values suggest that the respondent is closer to the Republican Party. This analysis includes robust standard errors that account for the statistical non-independence of individuals in the same district.⁸ The figure displays a point estimate represented by a hollow circle and the 90% confidence intervals.

Figure 2.

Partial regression results predicting partisanship (1 = strong Democrat, 7 = strong Republican) for (A) Blacks and (B) Latinos.

Both the Democratic and Republican Parties appear to make substantial gains with the black electorate by increasing the number of successful Black candidates who represent their party.⁹ Blacks in districts with successful Black Democratic U.S. House candidates are significantly more likely than Whites in districts with perceived Black Democratic U.S. House candidates and other Blacks to identify more closely with the Democratic Party. Figure 3A presents the predicted values for Democratic partisanship for Blacks who perceive the Democratic U.S. House candidate in their district to be Black and for those who do not. Blacks in districts with perceived Black Democratic U.S. House candidates have about a 0.13-point lower score on the 7-point Democrat-to-Republican scale than others. This result indicates that Democrats can retain their long-standing friendly relationship with the Black community through nominating successful Black candidates.

Figure 3.

Predicted values of partisanship for (A) Blacks in districts with Black Democratic candidate (W), (B) Blacks in districts with Black Republican candidates (W), and (C) Blacks in districts with Black Democratic candidates (L).

The Republican Party also makes substantial gains with the Black electorate through the nomination of successful Black candidates. According to Figure 3B, Blacks in districts with successful Black Republican candidates have over a half of a point higher score on the 7-point partisan index than comparable Blacks in districts where the candidate is not perceived to be Black. In combination with the results presented in Figure 3A, we see that Black partisanship is at least partially driven by descriptive representation.

Black Democrats who fail in their bids for elected office also appear to have some impact on partisanship. Figure 3C indicates that Blacks in districts with unsuccessful Black Democratic candidates have about a half of a point lower score on the 7-point partisan scale than Blacks in districts with no African American U.S. House candidates on the ballot. Thus, even the nomination of unsuccessful Black Democratic candidates may be an important outreach tool that the Democratic Party could use to appeal to Black voters. Figure 2A also demonstrates that the presence of losing Black Republicans may actually drive Blacks to the Democratic Party. While only marginally significant (p < .10), Blacks in districts with an unsuccessful Black Republican candidate are substantially more likely to identify as being closer to the Democratic Party. This result, however, may be attributed to the fact that non-Blacks in these contexts are significantly more likely to identify with the Republican Party. Thus, compared with Blacks in other districts, the baseline for comparison is skewed toward the Republican Party.

While Black descriptive candidates in some cases influence Black partisanship, we find no evidence that non-Blacks in districts with successful or unsuccessful Black Democratic candidates and successful Black Republican candidates have different levels of partisanship than non-Blacks in districts where neither candidate is perceived to be Black. This result indicates that symbolic appeals to Black voters by either party does very little to change levels of allegiance from other voters. Thus, the nomination of successful Black candidates may yield a net positive for the Democratic and Republican Party.

Unlike what was found for Blacks, the results presented in Figure 2B suggest that controlling for ideology, age, income, and other factors, successful or unsuccessful Latino candidates are not associated with significant differences in Latino partisanship. While all of the coefficients have the correct direction, none of the interaction terms are significant at the .10 level.¹⁰ Ultimately, there is not strong evidence that Latinos are swayed by symbolic outreach from either the Republican or Democratic Party’s nomination of a Latino candidate when accounting for other variables. This lack of a finding, however, may be attributed to the fact that congruence between the pan-ethnic identity of the candidate and the respondent may not be strong enough to be associated with any changes in partisanship. For example, a Mexican American in Florida may not be swayed by the Republican Party’s nomination of a Cuban American candidate.

Both Figures 3A and 3B show that Blacks and Latinos in districts without co-racial/ethnic candidates identify significantly more closely with the Democratic Party than do Whites. This result, while not unsurprising, indicates that even as minority partisanship changes, the Democratic Party continues to perform better in minority communities than it does among White voters.

Cause or Effect?

The preceding analysis has provided evidence that Blacks move closer to identifying with both the Democratic and Republican Parties when the parties nominate successful candidates who share their race. This fits our expectation that the significant potential for Black representation should improve both political parties’ standing with Black voters. However, the results at this point only demonstrate an associational relationship between the presence of successful Black Democratic and Republican candidates and Black Support for either the Democratic or Republican Party. Based solely on the results in Figure 2A, it is plausible that the presence of successful Black Democratic and Republican candidates does not change Black partisanship; instead it is the already high numbers of Black co-partisans that lead these candidates to succeed.¹¹ Thus, more needs to be done to determine the temporal ordering of successful descriptive candidates’ effect on Black partisanship.

To better assess whether descriptive representation leads to changes in the average Black partisanship over time, we combine our analysis of the 2010 CCES with the Congressional Cooperative Election Studies conducted during the previous two elections, 2006 and 2008.¹² If our hypothesis is correct, we should observe higher levels of Black Democratic partisanship across the United States after Obama’s election to the White House. However, there should be some exceptions to this growing level of Black allegiance to the Democratic Party, namely, Blacks who reside in the two districts where Black Republicans succeeded in 2010 (Tim Scott’s South Carolina 1st district and Alan West’s Florida 22nd district) should move significantly closer to identifying with the Republican Party. In combination, the 2006, 2008, and 2010 CCESs provide us the opportunity to better examine the temporal ordering between the nomination of successful Black candidates and differences in partisanship.

Using the combination of these three election data sets, we examine whether there are differences in Black partisanship measured on a 7-point scale between these three election years. In our first model, we examine whether Blacks grew significantly more in their Democratic partisanship between 2006 and 2008 than individuals of other races/ethnicities. We assess this by using an interaction term that measures whether the respondent was African American and was in the 2008 rather than the 2006 data set (Black × 2008). If our hypothesis is correct, Blacks should identify more closely with the Democratic Party after Obama’s election to the White House (i.e., in the 2008 CCES). In addition to examining whether Obama’s election increased Black identification with the Democratic Party, we also explore whether Black respondents who resided in one of the two districts with a successful Black Republican candidate identified as being closer to the Republican Party after the 2010 elections than they were in either 2008 or 2006, before the success of these candidates. We test this hypothesis in a second model, which includes only respondents from Alan West’s Florida 22nd district and Tim Scott’s South Carolina 1st district, and using an interaction term for the race of the respondent and the year in which Scott and West succeeded in their bids for Congress (2010).

Our dependent variable again measures partisanship on a 7-point scale that ranges from 1 (strong Democrat) to 7 (strong Republican). To isolate these variables’ influence on partisanship, we control for the respondents’ gender, age, education, income, and normalized self-ideology scores¹³ for both models. We also control for the respondents’ Congressional district’s Cook’s PVI score, the percentage of Black voters in the district, and finally whether the respondent resided in the South for the Obama model. Such controls are not necessary for the West–Scott models because they include respondents from only two Congressional districts. Thus, there is not enough variation in the scores to warrant them being placed in the model. Unfortunately, the 2006 CCES does not include measures of the candidate’s subjective race. Thus, these measures were not included in the models.

Consistent with our expectations, Model A in Table 1 demonstrates that Blacks had significantly lower scores on the 7-point partisanship measure after Obama’s election to the White House.¹⁴ The negative and statistically significant interaction term displayed in Table 1 suggests that differences in Black partisanship in 2008 were significantly greater among Blacks than other groups and that Blacks were significantly more likely to identify with the Democratic Party than Blacks in the previous election (2006). Figure 4A, which presents the predicted values derived from the regression results, indicates that Black partisanship on the Democratic-to-Republican scale declined by almost half a point between 2006 and 2008 indicating a substantial movement of Blacks back to the Democratic Party. The results indicate that Obama’s election to the White House may help reverse growing levels of political independence in the Black community and improve the Democratic Party’s standing with African Americans.

Table 1.

OLS Regression Predicting Change in Black Partisanship (1 = Strong Democrat, 7 = Strong Republican) Between 2006 and 2010.

	Model A: Obama model (2006-2008)		Model B: Black Republican model (only two districts)
Black × 2008	−0.37***	Black × 2010	1.31**
Black × 2008	(0.06)	Black × 2010	(0.60)
Black	−1.09***	Black	−1.81***
Black	(0.04)	Black	(0.47)
Latino	−0.37***	Latino	0.08
Latino	(0.03)	Latino	(0.32)
Other	−0.08**	Other	0.39
Other	(0.03)	Other	(0.38)
2008	−0.11***	2010	−0.24
2008	(0.02)	2010	(0.16)
Female	−0.15***	Female	−0.21
Female	(0.01)	Female	(0.15)
Age	−0.01***	Age	−0.01*
Age	(0.00)	Age	(0.01)
Income	0.00***	Income	0.00
Income	(0.00)	Income	(0.00)
Education	0.03***	Education	0.03
Education	(0.00)	Education	(0.04)
Normalized ideology	5.68***	Normalized ideology	5.75***
Normalized ideology	(0.03)	Normalized ideology	(0.28)
PVI	−0.01***
PVI	(0.00)
South	−0.05***
South	(0.02)
Percent Black Dist	0.00
Percent Black Dist	(0.00)
Constant	0.76***		0.69
Constant	(0.06)		(0.61)
Observations	46,830		429
R ²	.53		.54

Source. 2010 and 2006 CCES.

Note. Robust standard errors are given in parentheses. Model A contains all respondents from the 2006 and 2008 CCESs. Model B contains all respondents from the two districts in which Black Republicans succeeded (i.e., Florida’s 22nd district and South Carolina’s 1st district) from the 2006, 2008, and 2010 CCESs. CCES = Cooperative Congressional Election Study; OLS = ordinary least squares; Partisan Voting Index = PVI.

p < .1. **p < .05. ***p < .01.

Figure 4.

Predicted values of partisanship for (A) Blacks and non-Blacks before and after Obama’s election to the White House (2006-2008) and (B) Blacks and non-Blacks in districts with only successful Black Republicans in the 2010 elections.

While most Blacks moved toward the Democratic Party in the Obama years, according to the results presented in Model B in Table 1, Blacks in districts with successful Black Republican candidates moved in the opposite direction. According to Figure 4B, Blacks in districts with successful Black Republican candidates had more than a point higher score on the 7-point partisan scale than comparable Blacks who resided in these districts before a Black Republican was elected to the U.S. House of Representatives. Thus, Blacks moved closer to the Republican Party after the Party nominated successful Black candidates in these districts in 2010. In combination, the results indicate that the nomination of successful Black candidates does have a dynamic effect on shaping Black partisanship.

Conclusion

The growing importance of the minority vote increases pressure on both major political parties to attract and retain Black and Latino support. One such strategy that is being increasingly used is the diversification of each party’s candidate pool. While the Democratic Party has long had a number of minority candidates and elected officials, the Republican Party is now also making a concerted effort to increase the diversity within its ranks. The results of this study indicate that candidate diversity in some circumstances can improve a party’s position with minority voters. The Democratic and Republican Parties appear to be able to make particularly substantial gains with Black voters by increasing the number of successful African American candidates who carry their party’s label. Blacks who had the opportunity to vote for a successful African American for Congress were much more likely to identify with either the Democratic or Republican Party.

Instead of policy disagreements leading to declining levels of Democratic partisanship, lack of outreach from the Party specifically to the Black community has been deemed a culprit for rising levels of Black political independence (Bositis, 2010; Hajnal & Lee, 2011; Tate, 1991). Given the strong preference of Blacks for descriptive representation (see Schildkraut, 2013), the nomination of African American candidates may be a strategy the Democratic Party could use to demonstrate that it works for Black political interests. This strategy may also carry little risk of backlash, given that we did not observe significant differences in partisanship patterns among non-Blacks in districts with Black Democratic candidates.

The Republican Party may also be able to attract some Blacks by nominating credible Black politicians. While the Republican Party has put forth a number of minority candidates, a vast majority of these candidates are not viable and lack elected experience. In 2010, only 2 of 14 Black Republicans had any elected experience, and all but one (Tim Scott’s South Carolina 1st district) was challenging a well-established incumbent. Given these candidates’ dismal prospects, it is not surprising to find that a vast majority of these candidates fail to gain any traction in the Black community. However, if the Republican Party could increase the number of Black Republican elected officials through the nomination of credible candidates, it, according to this analysis, could make significant inroads with Blacks. While Blacks as a group still largely disagree substantively with the Republican Party (see Hajnal & Lee, 2011), a portion of Blacks who are attracted to the Party’s platform may move closer to identifying as Republican because they perceive it as being less hostile to individuals who share their race.

However, the results pertaining to Republican candidates are far from conclusive. Our findings related to this group are based on only two candidates (Tim Scott’s South Carolina 1st district and Alan West’s Florida 22nd district). While we account for the possibility of outliers influencing these findings in robust regressions reported in Table SA3 in the supplemental appendix, the findings in this study may not be fully generalizable to all Black Republican candidates given the small number of successful Black Republicans in the data that we analyze. Nonetheless, we hope that future studies replicate our analysis as/should Black Republican elected officials become more common. Nonetheless, our results show that descriptive representation plays a strong role in influencing Black partisanship. As previous studies have found a link between descriptive representation and changes in Black political behaviors and attitudes (Bobo & Gilliam, 1990; Griffin & Keane, 2006; Tate, 2003), our results indicate that it also plays a significant role in shaping Black partisanship.

While Blacks appeared to be significantly influenced by the nomination of co-racial candidates, Latino partisanship did not appear to be tied to the ethnicity of the candidate. While a number of studies demonstrate that Latinos respond positively to descriptive representation, partisanship and perceptions of political parties do not appear to be influenced by the presence of co-ethnic candidates for this rapidly growing ethnic group. Thus, if both parties want to improve their standing with the Latino community, they may have to take a different approach.

That being said, we hope that future studies revisit this issue with the following considerations. First, we hope that future research examines our analysis with panel survey data to better isolate the influence of co-racial/ethnic representation on changes in partisanship. While we utilize a number of different methods and data sources to address this issue, we believe that panel survey data would provide more conclusive evidence in support of our hypothesis. Second, the lack of a national origin measure in the data we use for this project may lead us to erroneously disregard the importance of descriptive candidates in shaping partisanship for Latinos. While pan-ethnic commonalities between candidates from different political parties and Latino voters may not lead to changes in partisanship, a shared national origin between candidates and voters may yield different results.

Third, future studies should explore internal differences between Blacks and Latinos when assessing the influence of descriptive candidates on partisanship. It is possible that some minorities may be more persuaded by the nomination of co-racial/ethnic candidates than others. For example, Blacks and Latinos with higher levels of group consciousness may be more receptive to parties that nominate candidates who share their race. Also, those who are more politically sophisticated may be less swayed by symbolic politics. Finally, our study examined the dyadic relationship between voters and co-ethnic candidates. While a descriptive candidate in isolation may not lead to changes in partisanship for Latinos, a collective growth in the party’s diversity may be influential. Thus, we hope that future research continues to examine the link between descriptive candidates and party identification.

Footnotes

Appendix

Table A1.

Regression Results Predicting Partisanship (1 = Strong Democrat . . . 7 = Strong Republican) for A: Blacks and B: Latinos.

	A: Black candidates		B: Latino candidates
Black × Black Dem Cand (W)	−0.17** (0.07)	Latino × Latino Dem Cand (W)	−0.16 (0.11)
Black × Black Rep Cand (W)	0.50** (0.25)	Latino × Latino Rep Cand (W)	0.15 (0.15)
Black × Black Dem Cand (L)	−0.48*** (0.18)	Latino×Latino Dem Cand (L)	0.02 (0.15)
Black × Black Rep Cand (L)	−0.20* (0.12)	Latino × Latino Rep Cand (L)	0.12 (0.15)
Black Dem Cand (W)	0.05 (0.05)	Latino Dem Cand (W)	0.01 (0.07)
Black Rep Cand (W)	0.06 (0.10)	Latino Rep Cand (W)	0.08 (0.08)
Black Dem Cand (L)	−0.09 (0.06)	Latino Dem Cand (L)	0.05 (0.08)
Black Rep Cand (L)	0.20*** (0.07)	Latino Rep Cand (L)	0.05 (0.08)
Female	−1.39*** (0.03)	Female	−1.44*** (0.03)
Black	−0.24*** (0.03)	Black	−0.25*** (0.03)
Latino	−0.10*** (0.03)	Latino	−0.10*** (0.03)
Other	−0.10*** (0.02)	Other	−0.11*** (0.02)
Age	−0.00*** (0.00)	Age	−0.00*** (0.00)
Income	0.00*** (0.00)	Income	0.00*** (0.00)
Education	0.02*** (0.00)	Education	0.02*** (0.00)
Ideology	0.87*** (0.00)	Ideology	0.87*** (0.00)
PVI	−0.00*** (0.00)	PVI	−0.00*** (0.00)
South	0.04** (0.02)	South	0.02 (0.02)
Percent Black district	−0.00 (0.00)	Percent Black district	0.00 (0.00)
GOP Win	0.08*** (0.02)	GOP Win	0.06*** (0.02)
		Generation	−0.01 (0.02)
		FL	0.04 (0.03)
Constant	0.08 (0.07)		0.12 (0.08)
Observations	39,147		39,056
R ²	.59		.59

Source. 2010 Cooperative Congressional Election Study.

Note. Robust standard errors in parentheses. PVI = Partisan Voting Index score. GOP=Republican Party.

p < .1. **p < .05. ***p < .01.

Acknowledgements

The authors thank Regina Branton, Andrea Benjamin, Sophia Wallace, Phil Habel, Tobin Grant, Scott McClurg, the editor of American Politics Research Brian J. Gaines, and the four anonymous reviewers for their helpful suggestions and comments on earlier drafts of this article.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Christopher T. Stout PhD is an assistant professor at Southern Illinois University, Carbondale. His research agenda crosses the fields of race and ethnicity, political behavior, representation, and public opinion. He is also the author of the forthcoming book Bringing Race Back In: Black Politicians, Deracialization, and Voting Behavior in the Age of Obama (University of Virginia Press).

Jennifer R. Garcia is a PhD candidate at the University of California, Irvine and an affiliate with the Center for the Study of Democracy. Her research focuses on the political representation of racial and ethnic minorities, minority legislative strategies in the U.S. Congress, public opinion and voting.

References

Allsop

Weisberg

H. F.

(1988). Measuring change in party identification in an election campaign. American Journal of Political Science, 32, 996-1017.

Barreto

M. A.

(2010). Ethnic cues: The role of shared ethnicity in Latino political participation. Ann Arbor: University of Michigan Press.

Barreto

M. A.

Nuño

S. A.

(2011). The effectiveness of coethnic contact on Latino political recruitment. Political Research Quarterly, 64, 448-459.

Bobo

Gilliam

F. D.

Jr. (1990). Race, sociopolitical participation, and black empowerment. The American Political Science Review, 84, 377-393.

Bositis

(2010). The political orientations of young African Americans. In Marable

Clarke

(Eds.), Barack Obama and African American empowerment: The rise of Black America’s new leadership (pp. 81-90). New York, NY: Palgrave-MacMillan.

Bowler

Nicholson

S. P.

Segura

G. M.

(2006). Earthquakes and aftershocks: Race, direct democracy, and partisan change. American Journal of Political Science, 50, 146-159.

Brody

R. A.

Rothenberg

L. S.

(1988). The instability of partisanship: An analysis of the 1980 presidential election. British Journal of Political Science, 18, 445-465.

Cain

B. E.

Kiewiet

D. R.

Uhlaner

C. J.

(1991). The acquisition of partisanship by Latinos and Asian Americans. American Journal of Political Science, 35, 390-422.

Campbell

Converse

P. E.

Miller

W. E.

Stokes

D. E.

(1960). The American voter. New York, NY: Wiley Press.

10.

Canon

D. T.

(1999). Race, redistricting and representation: The unintended consequences of black majority districts. Chicago, IL: University of Chicago Press.

11.

Carsey

T. M.

Layman

G. C.

(2006). Changing sides or changing minds? Party identification and policy preferences in the American electorate. American Journal of Political Science, 50, 464-477.

12.

Coffin

(2003). The Latino vote: Shaping America’s electoral future. The Political Quarterly, 74, 214-222.

13.

Dawson

M. C.

(1994). Behind the mule: Race, class and African American politics. Princeton, NJ: Princeton University Press.

14.

DeSipio

(1996). Counting on the Latino vote: Latinos as the new electorate. Charlottesville: University of Virginia Press.

15.

Evans

Franco

Polinard

J. L.

Wenzel

J. P.

Wrinkle

R. D.

(2012). Ethnic concerns and Latino party identification. The Social Science Journal, 49, 150-154.

16.

Fiorina

M. P.

(1981). Retrospective voting in American national elections. New Haven, CT: Yale University Press.

17.

Franklin

C. H.

(1992). Measurement and the dynamics of party identification. Political Behavior, 14, 297-309.

18.

Franklin

C. H.

Jackson

J. E.

(1983). The dynamics of party identification. American Political Science Review, 77, 957-973.

19.

Frymer

(1999). Uneasy alliances: Race and party competition in America. Princeton, NJ: Princeton University Press.

20.

Green

D. P.

Palmquist

Schickler

(2002). Partisan hearts and minds: Political parties and the social identities of voters. New Haven, CT: Yale University Press.

21.

Green

D. P.

Schickler

(1993). Multiple-measure assessment of party identification. Public Opinion Quarterly, 57, 505-535.

22.

Griffin

J. D.

Keane

(2006). Descriptive representation and the composition of African American turnout. American Journal of Political Science, 50, 998-1012.

23.

Grofman

Handley

(1989). Minority population proportion and Black and Hispanic congressional success in the 1970s and 1980s. American Politics Research, 17, 436-445.

24.

Grose

C. R.

(2011). Congress in Black and White. New York, NY: Cambridge University Press.

25.

Gurin

Hatchett

Jackson

(1989). Hope and independence: Blacks’ response to electoral and party politics. New York: Russell Sage.

26.

Hajnal

Z. L.

Lee

(2011). Why Americans don’t join the party: Race, immigration, and the failure (of political parties) to engage in the electorate. Princeton, NJ: Princeton University Press.

27.

King

(2006). A black/non-black theory of African-American partisanship: Hostility, racial consciousness and the Republican Party (Doctoral dissertation). Retrieved from University of North Texas Library. (70699470)

28.

Lublin

(1999). The paradox of representation: Racial gerrymandering and minority interests in Congress. Princeton, NJ: Princeton University Press.

29.

Lyons

(2011). Where you live and who you know: Political environments, social pressures, and partisan stability. American Politics Research, 39, 963-992.

30.

Mansbridge

(1999). Should Blacks represent Blacks and women represent women? A contingent “yes.” The Journal of Politics, 61, 628-657.

31.

Mendelberg

(2001). The race card: Campaign strategy, implicit messages, and the norm of equality. Princeton, NJ: Princeton University Press.

32.

Michelson

(2000). Political efficacy and electoral participation of Chicago Latinos. Social Science Quarterly, 81, 136-150.

33.

Minta

M. D.

(2011). Oversight: Representing the interests of Blacks and Latinos in Congress. Princeton, NJ: Princeton University Press.

34.

Nuño

S. A.

(2007). Latino mobilization and vote choice in the 2000 presidential election. American Politics Research, 35, 273-293.

35.

Page

B. I.

Jones

C. C.

(1979). Reciprocal effects of policy preferences, party loyalties and the vote. The American Political Science Review, 73, 1071-1089.

36.

Philpot

(2007). Race, Republicans, and the return of the party of Lincoln. Ann Arbor: University of Michigan Press.

37.

Rapoport

R. B.

(1997). Partisanship change in a candidate-centered Era. The Journal of Politics, 59, 185-199.

38.

Sanchez

G. R.

Morin

J. L.

(2011). The effect of descriptive representation on Latinos’ views of government and of themselves. Social Science Quarterly, 92, 483-508.

39.

Schildkraut

D. J.

(2013). Which birds of a feather flock together? Assessing attitudes about descriptive representation among Latinos and Asian Americans. American Politics Research, 41, 699-729.

40.

Tate

(1991). Black political participation in the 1984 and 1988 presidential elections. The American Political Science Review, 85, 1159-1176.

41.

Tate

(2003). Black faces in the mirror. Princeton, NJ: Princeton University Press.

42.

White

I. K.

Philpot

T. S.

Wylie

McGowen

(2007). Feeling the pain of my people: Hurricane Katrina, racial inequality, and the psyche of Black America. Journal of Black Studies, 37, 523-538.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB