Three Papers on Culture, Time, and Attitudes

This dissertation uses the lens of cultural sociology to understand variance in people’s attitude reports over time. Across three studies, I use a variety of panel surveys and statistical approaches to understand how and why people change their attitudes and adjudicate theories of culture. Study 1 uses data from the 2006 to 2014 rotating panels of the General Social Survey to adjudicate between a settled dispositions model, in which changes in attitudes are temporary and people return to a settled baseline, and an active updating model, where changes persist. Study 2 explores heterogeneity within the settled dispositions group, asking whether people’s attitude reports should be thought of as temporary constructs drawn from stored considerations or whether they represent durable opinions. It quantifies the prevalence of these opinion behaviors for 544 items from 10 panel data sets. Study 3 seeks to predict variance in attitude responses over time. Using data on religious, moral, and family structure beliefs in the National Study of Youth and Religion, I use Latent Class Analysis to deduce a set of constraints that should shape people’s response patterns over time. I test these constraints on people’s subsequent attitude reports. Taken together, results of these studies suggest 1) people’s attitudes are stable, on average, over the long term; 2) this average stability often masks high levels of instability in the short term, though some proportion of the population is stable on any issue; and 3) both this stability and instability are somewhat predictable based on a person’s pattern of beliefs. Findings suggest a model of culture where people internalize a diverse set of considerations when they are young and are shaped, in the short-term, by environmental influences. But durable cognitive structures, likely formed when people are young, limit the power of changing social circumstances to induce durable change.