I know the view exists, but I slighted it as 'context-free' because I don't know what constitutes 'many people', especially given that commenters are self-selected and thus not statistically representative of the general population. Here in San Francisco, for example, the main newspaper (the moderate-liberal Chronicle) is awash in conservative reader commentary, but this vocal contingent's influence on election results seems to be marginal. So perhaps all the above tells us is that (for example) drivers are more likely to read newspapers as well as owning valuable vehicles, which cyclists are more likely to chat on blogs, in keeping with their lower economic outlay.

I know I'm being pedantic here. Perhaps an interesting freakonomics post would examine different perceptions of few, some, man,y most, all and other such terms, the softness of which is often abused in public discourse.