Is Society Becoming Less Civil?

Online polls frequently get on my nerves because, instead of asking for pure opinions, they ask people to hazard uneducated guesses on questions that are ultimately empirical. Case in point is a CNN.com Quick Vote poll from earlier today, which asked:

“Is society in general becoming less civil?”

Respondents, who for the purposes of this post I will assume were predominantly American, overwhelmingly answered “yes.” I am skeptical of this assessment. My cynicism about the role of deliberation in liberal democracy aside, situating the current state of American discourse in historical perspective gives us no reason to believe we have become less civil. Certainly the tenor of the health care debate, taken as a whole, is not more contentious than the discourse that characterized the Iraq debate, or even Bill Clinton’s health care plan. Going back two more decades, it may even be safe to say we’ve made serious improvements since the Vietnam controversy and the civil rights debate in the 60s.

Fortunately, we have relatively objective measures of civic attitudes and behavior. The World Values Survey has compiled data on political and social attitudes with hundreds of survey questions to hundreds of thousands of people in dozens of countries since the early 1980s. I’ve compiled summary data for a few indicators (many of which are commonly employed in political behavior research) of civic attitudes in the US that were available for multiple years. A casual analysis of the results below provides some interesting, if ultimately inconclusive insights about variation in civic attitudes in America. First, interpersonal trust, one of the most widely used indicators of civic-ness in comparative survey research, has stayed remarkably constant since 1995. Second, the proportion of people that identified members of another race as people they would be uncomfortable having as neighbors has stayed relatively constant as well. Third, the proportion of people identifying with their local geographic unit, as opposed to the country as a whole, has not changed significantly. As anomalies, active membership in charitable organizations seems to have decreased, as has the proportion of people that report a high level of life satisfaction (although this effect could be from my aggregation of the responses). In short, although none of these indicators directly measure the vitriol of discourse, they do suggest that political and social behavior in the US has changed little, at least since 1995. I suppose it’s always possible that this particular health care debate brings out the worst in people, but I think that’s a tough case to make. (By way of caveat, I have not performed any statistical tests on these data, and I use “significant” in the colloquial, and not statistical, sense.)

	1995	1999	2006
Total Sample	1542	1200	1249
Interpersonal Trust
Most People Can Be Trusted	543(35%)	431(36%)	491(39%)
You Can’t Be Too Careful	968(63%)	757(63%)	750(60%)
Member of charitable/humanitarian organization
Not a member	895(58%)		847(68%)
Inactive member	233(15%)		170(14%)
Active member	399(26%)		198(16%)
Would not like to have as neighbor- members of different race
Mentioned	110(7%)	97(8%)	48(4%)
Not mentioned	1432(93%)	1103(92%)	1192(95%)
Life satisfaction
1-3	68(4%)	38(3%)	51(4%)
4-7	513(33%)	436(36%)	537(43%)
8-10	953(62%)	726(61%)	653(52%)
Geographical identification
locality	486(32%)	384(32%)
region	147(10%)	128(11%)
country	606(39%)	405(34%)

That said, the CNN poll is interesting because it can be used as an indication of people’s perceptions. If relatively objective indicators of civic attitudes have not changed significantly, a really interesting question is why people perceive behavior to have done so. By way of one speculative hypothesis, the internet is a non-costly form of communication that facilitates the distribution of extreme discourse, even if the absolute number of people that harbor those opinions is relatively unchanged. Other thoughts?