Home   |   Register               Web Search: by Google
channel navigation

   ONLINE EXTRAS/What Americans Think
Variables.ucactualname/Political News

 Political News
 The Issues
 Federal Page
 Columns - Cartoons
 Live Online
 Online Extras
 Early Returns
 Political Junkie
  What Americans Think
 Photo Galleries
 Video - Audio







A Nation of Strangers: The Decline of Socializing
Political Junkie

By Richard Morin
Washington Post Polling Director
Monday, June 12, 2000

Bob Putnam is back with a new book with a familiar title: "Bowling Alone." Putnam is the Harvard government professor who became famous five years ago with the publication of an essay with the same name in the Journal of Democracy. It propelled him instantaneously into the media spotlight--and made him a favored target of a number of academics, who sharply challenged his contention that Americans are disengaging from civic life.

Central to his new book is an extraordinary survey data set, discovered after publication of the essay, that will be controversial among survey methodologists. It is the DDB Lifestyles survey, which the international advertising agency has conducted each year since 1975, accumulating more than 87,000 cases.

Embedded in these surveys is a block of questions that measure what Putnam and other researchers call social "dark matter"--the informal contacts and associations that tie people to their families, neighbors, friends and larger community. Specifically, the survey asked, "How many times during the past year did you. . . " and listed 50 activities. The list includes almost everything that people can do with other people, from attending church to using a condom; from going on a picnic to going to a bar; from renting a video to attending a classical concert; from gambling in a casino to entertaining friends at home. It even asks people how often they "gave the finger to another motorist." (This measure of road rage is going up, Putnam reports.)

The DDB data are important because they allowed Putnam to expand one of his core arguments. He contended in his 1995 essay that people aren't joining volunteer groups like the PTA, League of Women Voters or bowling leagues any more. With DDB data, he could test whether informal interpersonal contacts were also in decline--and whether, more broadly, America was becoming a nation of strangers.

These data allowed him to answer one of the most significant challenges to his argument, which went something like this: True, people don't join civic organizations or social clubs any more. But so what? There's lots more informal contact these days, from water cooler conversations to Internet chat rooms, that serve the same function. Social capital isn't declining, it's merely changing form.

The DDB data were unequivocal: Across the board, personal interconnections were in rapid and sharp decline. One example: Between 1974 and 1998, the frequency with which Americans "spend a social evening with someone who lives in your neighborhood" fell by about one-third--from about 30 times a year to about 20 times a year among married people and from about 50 times a year to about 35 times a year among single people.

The data also included a question asking participants whether TV was their "primary form of entertainment." That allowed Putnam to test his theory that TV was partially responsible for declining social contact. Again, he struck gold.

The half of all Americans who said TV is their major source of entertainment volunteer and work on community projects less often, attend fewer dinner parties and fewer club meetings, spend less time visiting friends, entertain at home less, picnic less, are less interested in politics, give blood less often, write friends less regularly, make fewer long distance calls, send fewer greeting cards and less e-mail, and express more road rage.

Case closed. Or is it?

DDB is not based on a random sample of the population. It is based on a mail panel in which participants agree to take part in the survey. It is a "self-selected sample"--a dirty little thing, in the eyes of most serious survey methodologists. That's because the methods used to obtain such samples offer no assurances that the people who agree to participate are generally representative of the population as a whole. To the contrary, there's good reason to assume that they're probably not, increasing the chances that results may be misleading and not representative of the whole population.

Putnam argues that they are not. He sifted through other, methodologically strict random sample surveys to find questions similar to those asked by DDB. A comparison of results revealed few meaningful differences. For example, the DDB survey results suggest that 70 percent of all adults in 1991 "went to the movies" the previous year. The General Social Survey, conducted annually by the National Opinion Research Center at the University of Chicago, reported that 72 percent "went out to see a movie in a theater." The DDB survey found that 56 percent "attended a sporting event" in the previous 12 months-identical to the GSS figure.

There were some differences. Nine percent in the DDB survey reported they "went to an auto race" while 16 percent told GSS interviewers they had "attended [an] auto, stock car, or motorcycle race." Overall, differences were slight.

"I spent six months full-time doing nothing but making sure the (DDB) data set was reliable: finding comparable questions, looking to see if we could find any difference in either levels, change, or in underlying patterns of causation. The results looked essentially identical," Putnam says.

The DDB data have other problems, he acknowledges. "It's not a great sample for the urban underclass, or for young people. That's because those sorts of people move frequently and are less likely to be on the lists from which the samples are drawn. . . . [But] I am convinced that for the purposes I use it for in the book, it is a pretty reliable data source."

He fully expects some thoughtful colleagues to remain unconvinced by even this sophisticated analysis of mere marketing data. "That is going to happen," he acknowledges with a wave of his hand.

His says his arguments are strongly supported, with or without the DDB data. "Before I wrote the book, I asked myself this question: Suppose you took out all the DDB data. How much of the book would you still have evidence for? Basically there are some interesting bells and whistles to the argument. . . . The fundamental things would not be affected."

Richard Morin is director of polling for The Washington Post. "What Americans Think" appears Mondays in The Washington Post National Weekly Edition. Morin can be reached at morinr@clark.net .

© Copyright 2000 The Washington Post Company

Post Archives

Advanced Search

Politics Where
You Live

Enter state abbrev.
or ZIP code

Related Links

Archive: What Americans Think

Polls Section

Home   |   Register               Web Search: by Google
channel navigation