A few days ago, the headlines blared something akin to this Washington Post article:

Americans are far more socially isolated today than they were two decades ago, and a sharply growing number of people say they have no one in whom they can confide, according to a comprehensive new evaluation of the decline of social ties in the United States.

Naturally, the article goes into some detail about how sad it is that according to this survey, Americans are all alone in the world. They have no one they confide in, and their network of close relationships has dropped from about 3 to 2 in 20 years.

The study itself (which appeared in the June 2006 issue of the American Sociological Review) though, presents a far more complex picture where the researchers qualify their findings a lot more than the news accounts of the study do. For instance, the study authors acknowledge this finding is not consistent with other recent research into this issue (done by respected organizations such as the Pew Internet & America Life project). (The authors of the new study kindly suggest — naturally! — that their research is better than these other studies, but I take such boasting with a grain of salt.)

When new research conflicts with old research, the default assumption is not to trump the new research as somehow being “better” or more “definitive” in answering the question. Rather, researchers suggest this new finding needs to be further studied and, in the meantime, the results should be viewed as what they are — a tentative datapoint.

News organizations and the researchers themselves should do right by the public and say exactly this, but they gloss over the problems with the current study and its contradiction it presents with other research in this field. Until this discrepancy is resolved in future research, however, nothing much more can be said.

Are Americans really discussing personally important matters with less people today than in 1985?

Absolutely, it’s possible, since so many more modalities of communication and sharing have occurred in the past two decades. For instance, if I share my life in a blog, I can’t say I’m discussing personally important matters with X number of people — I could be doing it with thousands. But does it serve the same or a similar purpose for the person doing it? Absolutely. So while the researchers may have indeed found a reliable result, the conclusions they (and others) draw from that result could be astoundingly off.

Americans are not becoming more socially isolated, as the headlines blare. They are far more interconnected today, just in ways that researchers aren’t quite measuring adequately or accurately.