Knowledge and collective preferences - A comparison of two approaches to estimating the opinions of a better informed public

This article compares estimates of "informed" public opinion derived from the regression-based approach of Bartels, Delli Carpini and Keeler and Althaus with those from the deliberative polling method of Fishkin on the same sample of respondents. Contrary to low-information rationality perspectives, both methods indicate that across a range of prominent policy domains, level of political awareness has a strong impact on the expressed preferences of individuals. And while self-canceling across respondents tends to translate these individual-level influences into only rather modest effects in the aggregate, on a significant minority of issues, substantial shifts in collective opinion remain. The broad similarity of the estimates produced by these two very different methods, in addition to their convergence with previous studies of information effects, lends some simultaneous support to the validity and reliability of both approaches.