About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline

LOG IN
LOG IN


58 Am. J. Pol. Sci. 1 (2014)

handle is hein.journals/amjorpolisci58 and id is 1 raw text is: Widowhood Effects in Voter Participation William R. Hobbs University of California, San Diego Nicholas A. ChristakiS Harvard University James H. Fowler University of California, San Diego Past research suggests that spouses influence one another to vote, but it relies almost exclusively on correlation in turnout. It is therefore difficult to establish whether spouses mobilize each other or tend to marry similar others. Here, we test the dependency hypothesis by examining voting behavior before and after the death of a spouse. We link nearly six million California voter records to Social Security death records and use both coarsened exact matching and multiple cohort comparison to estimate the effects of spousal loss. The results show that after turnout rates stabilize, widowed individuals vote nine percentage points fewer than they would had their spouse still been living; the results also suggest that this change maypersist indefinitely. Variations in this widowhood effect on voting support a social-isolation explanationfor the drop in turnout. ecent experimental studies highlight the impor- tance of social mobilization and demonstrate that efforts to increase turnout may not only con- vince those directly contacted to vote but also their po- litical partners, friends, and family members. For ex- ample, Nickerson (2008) conducted a get-out-the-vote (GOTV) field experiment with two-person households and showed that 60% of the increased propensity to turn out resulting from the treatment was passed from one household member to the other. Gerber, Green, and Larimer (2008) also conducted a GOTV experiment in which they promised to show neighbors whether or not a person voted. This raised turnout by 8%-one of the largest effects ever observed for aby-mail treatment. Bond et al. (2012) conducted a 61-million-person experiment on Facebook and showed that a GOTV message not only increased turnout among the recipients but among the recipients' friends as well. Observational work by Cutts and Fieldhouse (2009) provides evidence that the occu- pants in two-person households have substantially larger effects on each other's turnout propensity than other ge- ographically proximate electors. The basis of many of these studies is a theory of social dependency. That is, forms of political participation that are stronglymotivatedbyinterpersonal influences mayin turn be strongly dependent on them. This dependency, and a lack of (community-based) social connectedness and mobilization in recent decades, may account for vari- ation in political participation rates (Gerber, Green, and Larimer 2008; McPherson, Smith-Lovin, and Brashears 2006; Putnam 2001). Importantly, however, these studies do not directly test whether increased social isolation will have large and long-lasting (negative) effects on political participation. Here, we study political interpersonal dependency directly-specifically, dependency within a spousal rela- tionship. This is a natural starting place for research on so- cial isolation/connection and turnout. Spouses have long been identified as an important source of mobilization and influence (Campbell et al. 1960; Glaser 1959). William R. Hobbs is PhD student of Political Science, University of California, San Diego, 9500 Gilman Dr. #0521, La Jolla, CA 92093 (whobbs@ucsd.edu). Nicholas A. Christakis is Professor of Sociology, Harvard University, 180 Longwood Ave., Boston, MA 02115 (Nicholas'christakis@harvard.edu).James H. Fowler is Professor of Political Science, University of California, San Diego, 9500 Gilman Dr. #0521, La Jolla, CA 92093 (jhfowler@ucsd.edu). We thank Andre Blais, Kent Jennings, Alex Verink, the Human Nature Group, the UCSD Comparative Politics Workshop, panel participants at the 2012 MPSA conference, and the anonymous reviewers for helpful comments on earlier versions of this article. This research uses three publicly available data sources: the Social Security Death Master File (distributed through the National Technical Information Service), the California Voter Record (distributed through the California Secretary of State's office), and the U.S. Census zip-code data. The Social Security Death Master File, relatively complete at the time we obtained the data, no longer includes many state records. More complete death-record information can be obtained by application to the California Department of Public Health. American Journal ofPolitical Science, Vol. 58, No. 1, January 2014, Pp. 1-16 92013, Midwest Political Science Association DOI: 10.1111/ajps.12040 1