(iPoster) Does Party Closeness Organize Democratic Norms? Evidence from Afrobarometer

• In Event: Poster Sessions: Public Administration & Political Organizations and Parties
  In Poster Session: Poster Session: Political Organizations and Parties

Abstract

Survey research on mass belief systems faces a measurement problem: low cross-item constraint may reflect weakly organized attitudes or attenuation from item-specific interpretation and response noise. Using pooled Afrobarometer Round 9 data, we study party-system democratic norms measured by three items- rejection of one-party rule, support for multiparty competition, and support for electoral turnover. We analyze the battery using two standard approaches: Attitude Coherence (Converse Index) and Party-System Norms (Graded Response Model Latent Trait Score). Across specifications with country and country-by-region fixed effects (N approximately 49,503 to 49,863) and across expanded covariate sets and inference procedures, we find a stable result: party closeness is substantively negligible for both outcomes. Using a pre-specified smallest effect of interest, the confidence intervals fall entirely within bounds that rule out even modest within-country associations- larger than 0.02 on the coherence scale and larger than 0.05 standard deviations on the latent norms score. By contrast, education, news exposure, and political discussion predict the latent norms score, while contact with party officials is modestly negative. The findings show that short batteries can yield partially structured latent summaries while producing low response coherence, and they do not support a general within-country relationship between partisan attachment and these party-system norms.

Author

• Joseph Ofori Acheampong, University of Illinois Urbana-Champaign