The source:“Social Isolation in America: Changes in Core Discussion Networks Over Two Decades” by Miller McPherson, Lynn ­Smith-­Lovin, and Matthew E. Brashears, and “Trends in Civic Association Activity in Four Democracies: The Special Case of Women in the United States” by Robert Andersen, James Curtis, and Edward Grabb, in  American Sociological Review, June 2006.

When it comes time to let down their hair and talk, Americans have fewer people to confide in than they did just a generation ago. The number of people the average person would consider going to for advice fell from about three to two between 1985 and 2004. Almost half the population now says they can discuss important topics with only one other person or no one at ­all.

The greatest change has come in the decline in intimates outside the family circle. Twenty years ago, 80 percent of Americans who re­sponded to the national General Social Survey had at least one confidante who was not a relative. By 2004, that number had fallen to 57 percent, according to sociologists Miller McPherson and Lynn ­Smith-­Lovin, of the University of Arizona and Duke University, respectively, and Matthew E. Brashears, a Ph.D. candidate at Arizona. The number of people who depend totally on their spouse has doubled, to not quite 10 ­percent.

Better-educated and younger people have larger “discussion networks” than others. Women have slightly more confidantes, statistically, than men, and whites have more than ­non­whites. Intimate friendships between neighbors and fellow participants in civic activities have declined the ­most.

The authors say that their research may have detected another trend. “Shifts in work, geographic, and rec­re­ational patterns” and increasing use of the Internet may be leading to the development of larger, less localized groups of friends than in the past, when strong, tightly interconnected networks were more the norm.

Similar social forces may be responsible for the purported decline in civic engagement in the United States. Sociologists Robert Andersen, James Curtis, and Edward Grabb, of, respectively, McMaster University, the University of Waterloo, and the Univer­sity of Western Ontario, all in Canada, studied civic activity in the Netherlands, Great Britain, Canada, and the United States. They found a decline only in ­America—­and, significantly, only among women. While the lessening of civic involvement in the United States has been blamed on television watching and the fading of the more selfless World War II generation, the authors note that the same fac­tors are at work in the other three countries. Because the decrease in civic involvement is limited to women, Andersen and colleagues suggest that the “greater de­mands” on American women’s free time may be responsible. Women’s child-care duties have increased in the United States, while declining in Canada and the Netherlands, for example. “The larger time commitment American women now make to paid work, combined with their in­creased time for child care, could be the principal explanation be­hind the decline in civic association activity of Americans,” the authors ­say.