This thesis conducts a cross-national analysis of perceived electoral fairness across 80 countries in Africa, Asia, Europe, and the Americas. The main research question is focused on uncovering the determinants of perceived electoral fairness. The thesis analyses two broad groups of variables, with the chapter structures following from this distinction. The first set of variables arises from a meta-analysis of previous studies to ascertain the strongest determinants of electoral fairness. This entails testing as many individual and national level variables as possible across as many countries as possible to reveal any global trends or regional differences. The second set of variables arise from two theoretical models of electoral management body (EMB) design, both of which focus on evaluating the merits of EMB independence or autonomy from the ruling government. Once again, the approach is to examine as many electoral management design variables as possible across as many regions as possible. Results unexpectedly indicate frequent negative associations between EMB independence and perceived electoral fairness. The thesis considers different possibilities for these unexpected negative results, with the most probable cause being the many independent EMBs in authoritarian democracies included in this study. The thesis uses multivariate ordered probit and multilevel mixed-effects regression models to conduct analyses at the national, regional, and global levels. Public survey data comes from the AfroBarometer, ArabBarometer, AmericasBarometer, AsianBarometer, and Comparative Study of Electoral Systems. National level indicators come from the Quality of Government, Administration and Cost of Elections (ACE) Electoral Knowledge Network, and International Institute for Democracy and Electoral Assistance (IDEA).
|Qualification||Doctor of Philosophy|
|Award date||7 Mar 2014|
|Place of Publication||Canberra, ACT|
|Publication status||Published - 1 Sep 2014|