(iPoster) Ethnic Population Distribution and Pseudo Urban Effects on Ethnic Voting

• In Event: Poster Session: Democracy and Autocracy I

Abstract

Comparative political studies of developing democracies often perpetuate a common belief: urban areas inevitably become engines of change, driving democratic transitions. This expectation is particularly pronounced in countries with significant ethnic divisions. The urban effect is theorized to weaken ethnic divisions through two primary mechanisms: first, by fostering culturally diverse urban coalitions that transcend ethnic segregation, and second, by cultivating an economically independent urban middle class resistant to ethnic-based patron-client relationships. However, I argue that we may have overestimated the urban effect by overlooking the geographical distribution of ethnic populations. Specifically, the urban effect may emerge not because urbanization inherently fosters changes in voting behaviors or campaign strategies, but because certain ethnic groups are more likely to reside in urban areas. Using mediation analysis of aggregated election results and logistic regressions based on survey data from Malaysia’s 14th General Election, a landmark event that marked the country’s first ruling party turnover, I demonstrate that urban areas were more likely to vote for change primarily because they housed a higher concentration of Chinese citizens, Malaysia's largest ethnic minority. In addition to presenting these empirical findings, I address several methodological challenges, particularly the limitations of using aggregated election results to conduct robust causal analyses in contexts where high-quality survey data is scarce.

Author

• Tyler Rongxuan Chen, University of Michigan