Thesis (M.A.)--Brigham Young University. Dept. of History. / Includes bibliographical references (leaves 128-129).

Thesis (M.A.)--Brigham Young University. Dept. of History. / Includes bibliographical references (leaves 128-129).

Revision of the author's thesis (Ph. D.--Beijing da xue, 2000) / Includes bibliographical references (p. 337-352).

Thesis (Ph. D.)--Harvard University, 1992. / Includes bibliographical references (leaves 371-385).

Thesis (Ph. D.)--University of North Texas, 2000. / Includes bibliographical references (leaves 284-208).

Thesis (M.A.)--Brigham Young University. Dept. of History. / Electronic thesis. Includes bibliographical references (leaves 128-129). Also available in print ed.

Thesis (Ph. D.)--University of Wisconsin. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 289-293).

Thesis (Ph. D.)--University of Texas at Austin, 2005. / Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of California, San Diego, 2007. / Title from first page of PDF file (viewed June 4, 2007). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references (p. 116-125).

A key expectation of elections is that voters hold elected officials to account for their performance in office. A contradiction exists in the literature between those who believe voters are incapable of exercising accountability and those who believe voters are successfully able to do so. This thesis addresses this contradiction and presents four empirical case studies in which I test the conditions under which voters exercise accountability. In order to explore these conditions I focus on situations in which partisan voters in the US receive a performance signal that directly contradicts their partisan loyalty. I refer to these voters as cross-pressured partisans. In the thesis I examine who these voters are and how they behave at election time. I find that partisans are responsive to signals about the quality of economic management of the incumbent government and that candidate competence matters at the local level. I find that the electoral response to corruption is more mixed. Although some partisans punish their party in light of a corruption scandal, corruption also depresses turnout and a lab experiment shows that voters do not punish candidates who embezzle funds if the candidate shares their group identity. The thesis finds that three factors moderate the extent to which partisan voters hold their own party to account: (1) individual voter characteristics; (2) the electoral context and; (3) the type of government performance. First, partisan voters with strong levels of affiliation to a party are less likely to exercise accountability whereas those with moderate levels are most likely to do so. Second, if the partisan stakes of an election are high, democratic accountability is low. Third, although partisan voters exercise accountability in the domain of economic management and candidate competence, the electoral response to corruption is more mixed. The thesis shows that there are many partisans who are conflicted between their loyalty to their party and their evaluations of incumbent performance. Under certain conditions this conflict incites behavioural change at election time. Partisans are therefore not incapable of holding their preferred candidate or party to account for their performance in office, but neither should they always be expected to do so.