This study explores changes in British Columbia electoral patterns during the twentieth century, and relates these changes to dimensions of societal and political modernization. It focuses on swing, the percentage point shift in a party's support between two successive elections, and examines constituency- arid sub constituency-level results in provincial elections between 1903 and 1975. The thesis tested is that development of the province's communications infrastructure was a central cause of the electoral developments which took place. The first part of the study clarifies the electoral developments by tracing changes in the level of swing uniformity and the degree of swing patterning. It begins with evidence that swings became much more uniform as the century progressed. Analyses of electoral shifts in constituencies and nonmetropolitan communities both show that swings of parallel direction and magnitude were much more likely in elections after 1952. This trend to swing uniformity is taken to indicate a decline in the importance of local electoral forces. It is hypothesized that the twentieth century communications revolution contributed to this decrease in electoral localism by facilitating the establishment of locality-arching patterns of political influence. After demonstrating the increase in swing uniformity, we examine three developments which could explain the trend. The premise underlying this part of the study is that increased patterning of swing by the characteristics of constituencies or communities may account for increased uniformity. Tests for cross-election changes in the explanatory power of three variables—electoral competitiveness, socio-economic composition, and region—show that the overall decline in swing variance was not accounted for by increased patterning. The trend to uniformity was unpatterned; voter aggregates with different characteristics and locations were simply more likely to produce parallel swings in later elections. The second part of the study explores the reasons for these developments. The communications development interpretation is tested and alternative interpretations are considered. The communications interpretation argues that improvements in communications infrastructure contributed to an increase in the uniformity of electoral forces operating on dispersed constituencies and communities, and thus helped to bring about increased swing uniformity. Chapters 7 and 8 test four propositions which are derived from this interpretation. These state: (a) that there should be a detailed correspondence between the pace of communications development and the trend to swing uniformity; (b) that the appearance of intense regional communications patterns should predict the regional swing patterns which marked the 1969, 1972 and 1975 elections; (c) that regional differences in the timing of the trend to uniformity should be explained by differences in the pace of communications development; and (d) that communications isolation should explain the tendency of some contemporary communities to swing in ways which indicate that they are insulated from prevailing electoral forces. The results of these tests enhance the credibility of the communications interpretation. In speculating about alternative interpretations we acknowledge that a complete causal map would have to grant other factors an important place. But the evidence supporting the test propositions, and the fact that the most plausible alternative interpretations complement the communications interpretation, argue that communications change was a principal cause of the provincialization of British Columbia electoral politics. Communications modernization altered the relationship between geography and the spatial distribution of electoral results. / Arts, Faculty of / Political Science, Department of / Graduate