UC Berkeley

Partisan polarization spreads through more and more of American society over time. Republicans and Democrats held different opinions toward partisan-valenced institutions like the presidency and a few economic institutions (e.g. big business) in the 1970s, but today almost all societal institutions – whether governmental or ostensibly apolitical – are understood by many Americans as liberal or conservative, made up mostly of one partisan group or the other. As a result many partisans trust only institutions they believe are made up of people who are similar to them ideologically.This dissertation uses attempts to answer these questions: When did this polarization of trust emerge in each institution? Why did this polarization occur? What effect does it have on regular Americans? Chapter 1 documents trust in institutions using 50 years of existing public opinion data and original public opinion measurement. Chapter 2 uses campaign donation data and machine learning to establish a possible cause of polarization of trust: the real composition of institutions. Chapter 3 demonstrates one reason this polarization is important: if it is provided, Americans will use partisan information about doctors to make decisions about health care. There is already evidence of individual-level partisan differences in trust in doctors, as well as aggregate-level evidence of differential outcomes between Republicans and Democrats.