With Chapter 6, the volume shifts its focus to the role of the media in creating an informational environment that affects voters’ ability to hold elected officials accountable. Opening Part II, Brandice Canes-Wrone and Michael Kistner exploit variation across districts and over time in the congruence between local newspaper markets and House members’ districts. Using this variation, the authors estimate the effect of media coverage on the link between candidate ideology and election outcomes. For incumbents, who have track records of roll- call voting in Congress, differences in coverage only modestly affect the relationship between incumbent ideology and election outcomes. For challengers, however, reduced coverage is associated with a substantial reduction in both the penalty for ideological extremity and the reward for ideological moderation. The authors also find, consistent with the decline of local media and the rise of the internet, that the effect of local newspaper congruence may have decreased over time. Overall, media coverage and information are important in accountability, but in surprisingly subtle ways.

Rounding out Part 1, Kathleen Bawn, Knox Brown, Angela X. Ocampo, Shawn Patterson, Jr., John L. Ray, and John Zaller report results from one of the largest “on the ground” studies of candidate selection ever undertaken. Focusing on fifty-three potentially winnable open seat House races in the 2013–14 election cycle, the authors interviewed local participants and observers to probe the processes behind candidate selection in primary elections. The extensive fieldwork reveals that groups, which may include local party organizations, are the central political actors in the selection process. Voters rely on signals from the groups, whose primary objective is to minimize uncertainty about a candidate’s commitment to particular policy goals. Because this objective may lead groups to promote nominees who are not closest to the primary electorates’ ideological preferences, the candidates who win may be more ideologically extreme than even the primary electorates are. The chapter therefore highlights how analyses that focus on the general election, such as Chapters 2 and 6, are dependent on the candidates that emerge from the primary selection process.