Cultural vs. ideological cognition, part 1
Tuesday, December 20, 2011 at 8:11AM
Dan Kahan

In our study of cultural cognition, we use a two-dimensional scheme to measure the group values that we hypothesize influence individuals’ perceptions of risk and related facts. The dimensions, Hiearchy-Egalitarianism (“Hiearchy”) and Individualism-Communitarianism (“Individualism”), are patterned on a framework associated with the “cultural theory of risk” associated with the work of Mary Douglas and Aaron Wildavsky. Because they are cross-cutting or orthogonal, they can be viewed as defining four cultural worldview quadrants: Hierarchy-individualism (HI); Hierarchy-communitarianism (HC); Egalitarian-individualism (EI); and Egalitarian-communitarianism (EC). 

Often we are asked why we don’t just use the more familiar political measures like “liberal-conservative” ideology or Democratic-Republican party affiliation.  I am going to give a three-part answer to this question in a sequence (likely continuous) of posts.

Part 1: Two dimensions dominate one

We started this project as an effort to cash out the cultural theory of risk, so not surprisingly the first part of the answer is just an elaboration of the argument that Aaron Wildvasky made for using Douglas’s scheme rather than liberal-conservative ideology as a measure of individual differences in political psychology. Wildavsky conjectured that Douglas’s two dimensions would explain more controversies, more coherently, than a one dimensional left-right measure. 

Our work and that of others seems to bear that out. It’s true that Hierarchy and Individualism are both modestly correlated (in the vicinity of 0.4 for the former and 0.25 for the latter) with political conservatism. But the cross-cutting Hierarchy and Individualism dimensions can often capture divisions of belief that evade the simple one-dimensional spectrum of liberal-conservative ideology (or of Republican-Democrat party identity), particularly where conflicts pit the EI quadrant against the HC one:

In addition, the explanatory power of political party affiliation and ideology tends to be very sensitive to individuals’ level of political knowledge or sophistical. They work fine for those who are high in knowledge or sophistication (as political scientists measure it) but not for those are moderate or low.

Wildavsky was aware of this and surmised that the culture measures would do a better job, because cultural cues are more readily accessible to the mass of ordinary citizens than are argumentative inferences drawn from the abstracts concepts that pervade ideological theories.

Our work seems to bear out this part of Wildavsky’s argument, too. The culture measures, we have found, explain divisions even among individuals who are relatively low in sophistication when ideology and party can’t.

The goal is to generate a reasonably tractable scheme that explains and predicts risk (and related facts), and generates policy prescriptions and other interventions that improve people’s ability to make sense of risk. A one-dimensional scheme — like liberal-conservative ideology — is very tractable, very parsimonious, but we agree with Wildavsky that the greater explanatory, predictive, and prescriptive power associated with a two-dimensional cultural scheme is well worth the manageable level of complexity that it introduces.

part 2

part 3

Article originally appeared on cultural cognition project (
See website for complete article licensing information.