Using a cross-national perspective covering all free and partly free countries, this thesis addresses two questions: What factors are associated with levels of gender representation, ethnic group representation, and ideological representation? And what are the relationships between levels of gender, ethnic group, and ideological representation? Ideological representation regards policy positions in different issue domains, whilst gender and ethnic group representation are concerned with the inclusion of women and ethnic groups in parliament. The representation of ethnic groups is approached in a multivariate cross-national analysis for the first time. Cultural rather than institutional factors seem to be the best predictors for the different levels of gender representation and ethnic group representation. Cultural attitudes are measured with survey questions on attitudes to women as political leaders, and tolerance of marginalized groups in society. The thesis finds that on average quotas for women and ethnic groups are not associated with higher levels of representation, perhaps because of issues regarding how quotas are implemented. Broadly speaking, little effect of the electoral system on any form of representation could be observed. Looking at levels of ideological representation, in line with some recent studies, the thesis suggests that the electoral system is not associated with different levels of ideological representation. I show that this is the case across various policy domains. Furthermore, the thesis finds no evidence for a direct relationship between levels of gender representation and levels of ethnic group representation, but levels of gender representation may be associated with levels of left–right representation. The relationship between different forms of representation might be shaped by the salience of ideological domains and awareness of under-representation of ethnic minority groups. Overall, the thesis argues that cultural attitudes are central to understanding levels of political representation, a factor often neglected in the literature.

Que traduit la hausse de la participation aux élections présidentielles françaises de 2007 et américaines 2008 ? À partir d’enquêtes électorales française, américaine et canadienne, notre étude met en lumière les changements d’attitudes politiques des citoyens survenus ces dernières années et la pratique croissante d’une participation intermittente et sélective. Ces changements qui résultent de logiques d’individualisation ont permis aux individus de s’extirper de contraintes d’appartenance sociale et de s’affranchir d’allégeances partisanes. Leur jugement se fonde désormais sur une évaluation des critères de l’offre électorale. L’individualisation des valeurs en politique se traduit notamment par une évolution du sens du vote et une hiérarchisation des scrutins. Les électeurs ne se délaissent pas les élections, ils choisissent simplement de participer autrement. Plus instruits, plus autonomes et plus critiques, ces citoyens prescrivent des changements profonds de l’offre. Afin de rallier le plus grand nombre de suffrages, les candidats se voient contraints d’adapter leurs dispositifs stratégiques aux évolutions des comportements politiques des citoyens : structuration des partis, modernisation des campagnes électorales, militantisme à « la carte », nouvelles formes d’engagement politique, usage de l’Internet. Tout est fait pour que le citoyen individualisé trouve dans ces manières de faire la politique les bonnes raisons de voter. / What does the increase of the voter turnout in the 2007 French and the 2008 American presidential election mean? With the French, American and Canadian surveys, we highlight the changes in the political attitude of the citizens occurred in recent years. The voters are now practicing an intermittent and selective voting. This changing political behavior reflects a growing individualization values in politics. They are making their mind up from an assessment of the electoral issues positions, an evolution of the meaning of the vote and, a priorization of elections. Voters are not leaving the polling stations, they are only choosing to participate in different way. More educated, autonomous and critical, the citizens require deep changes in the political evaluations. So, by winning number of votes, candidates have to adapt their strategical plan of actions by including the individualized behavior of the citizens : structuring the political party, modernizing the electoral campaigns, promoting a membership “à la carte”, developing new forms of political engagement, and using the Internet. Everything is done to increase the individualized citizens vote.