Political Discussion Networks and Information

Chapter 5 examines the relationship among network characteristics, political knowledge, and policy attitudes by ethnorace, nativity, and gender. While we are unable to distinguish between selection and social influence, we uncover some interesting patterns. Network size is positively associated with political knowledge for both men and women, but we observe variation by ethnorace and nativity. Network size is not associated with political knowledge among Latinos, and discussion frequency is not associated with political knowledge among Blacks, Latinos, or Whites. Discussion frequency is positively associated with political knowledge among both the U.S. and foreign born, but network size is only associated with political knowledge among the U.S. born. For policy attitudes, we explored two issues that are important for marginalized ethnoracial group members: environmental policy and social justice policy. We find substantial variation in the relationship between network characteristics and policy preferences among the ethnoracial groups in the sample.

Precursors to Political Engagement

In this chapter, we investigate the extent to which discussion network characteristics are associated with feelings of political efficacy and trust, which are essential precursors to political participation. We explore these factors by of ethnorace, nativity, and gender. We compare levels of trust and efficacy across these groups descriptively, then use ordinary least squares reg ression models to examine how network characteristics are associated with efficacy and trust. We supplement these empirical findings with evidence from our qualitative interviews. We find that network characteristics do not influence political trust and political efficacy among the foreign born as much as they do among the U.S. born. We observe limited differences in these relationships between men and women. We also find substantial variation among ethnoracial groups; for example, we find that partisan homogeneity is positively associated with political trust for Blacks and Latinos, but not for Whites or Asian Americans.

Introduction

In this chapter, we introduce the theory, situate the research within the literature, preview the results, and provide an overview for the rest of the book. We argue that individuals with varying social positions have political discussion networks that are composed differently, and as a consequence their discussion networks exert distinct effects on their political behavior. We assert that this book makes three central contributions: (1) expanding the scope of the political discussion network literature by providing a comparative analysis across ethnorace, nativity, and gender; (2) demonstrating how historical differences in partisanship, policy attitudes, and engagement are reflected within groups’ social networks; and (3) revealing how the social position of our respondents affects the impact that networks can have on their trust and efficacy in government, political knowledge, policy attitudes, and political and civic engagement patterns.

The Impact of Political Discussion Networks on Civic and Political Engagement

Chapter 6 examines the association among our network characteristics - partisan homogeneity, discussion frequency, and network size - and, civic engagement, self-reported political engagement, and validated voter turnout across our various social groups. Of course, we cannot causally distinguish between selection and social influence with our data. However, we uncover important associations nonetheless. We find that regardless of social position, across all groups, those who discuss politics more frequently are more politically engaged, but network size and partisan homogeneity are not uniformly associated with engagement. With regards to validated turnout, we observe that network size and discussion frequency are positively associated with voting for U.S.-born respondents, but none of the network characteristics are associated with voting among our foreign-born respondents.

Conclusion

In this concluding chapter, we summarize our core findings and connect them to our broader argument. We discuss some of the limitations of this book with an eye toward future research. Our analyses reveal the wide-ranging potential explanatory power of political discussion networks for U.S. political behavior. At a minimum, our results show how focusing solely on the individual level as the main driver of political actions misses the important ways in which individuals interact with their social context. Such interactions have the potential to explain more fully patterns of political behavior and how that interaction is mediated by individuals’ experiences of marginalization and privilege.

The Composition and Determinants of Political Discussion Networks

This chapter represents one of the core contributions of this book. In this chapter, we present extensive descriptive statistics on how political discussion networks vary by ethnorace, nativity, and gender. We examine variation in the social composition of the networks (e.g., whether individuals report discussing politics with family members, friends, coworkers, etc.), the partisan composition and extent of partisan homophily in the networks, the size of the discussion networks, and the frequency of discussion in the networks. We conclude by examining which individual-level characteristics are most strongly associated with discussion network attributes, and how these relationships vary across the groups. We uncover several important patterns, such as U.S.-born respondents having larger and more homogeneous discussion networks than foreign-born respondents, Whites being least likely to report that they did not know their discussants’ partisanship, and non-whites discussing politics less frequently than do Whites.

Empirical Shortcomings

This chapter reviews some of the empirical reasons why so little is known about non-white networks and describes the original survey designed by the authors to address this lacuna in research. The survey data on a diverse sample of California voters is the basis for most of the analyses presented in this book, supplemented by in-depth interviews with canvassers from three local grassroots organizations. Detailed descriptions of the methodological approach follow, including the sampling strategy, survey design, and respondent characteristics, as well as model selection and variable operationalization. Also covered is the qualitative interview protocol, which involves speaking with a diverse selection of canvassers about current political issues and events. Methods of measuring network characteristics, which include network size, partisan homogeneity, and discussion frequency, are explained in detail.