*****  To join INSNA, visit  *****

This discussion originated on the Association of Internet Researchers
list. But as it is germane to social network analysts, comm-tech
sociologists, and community sociologists -- I am reposting it here.
Absolutely no apologies for crossposting.

  Barry Wellman         Professor of Sociology        NetLab Director
  wellman at

  Centre for Urban & Community Studies          University of Toronto
  455 Spadina Avenue    Toronto Canada M5S 2G8    fax:+1-416-978-7162

  You're invited to visit -- and contribute to -- my new fun website
 "Updating Cybertimes: It's Time to Bring Our Culture into Cyberspace"

---------- Forwarded message ----------
Date: Sat, 24 Jun 2006 10:02:59 -0400
From: Barry Wellman <[log in to unmask]>
To: aoir list <[log in to unmask]>
Subject: overstated media inference from fine study

Re the Washington Post story below (and a similar USA Today story):

To my mind, it's an interesting case of media distortion.

The good news: the accounts are based on a high-quality survey (US General
Social Survey)  by first-class reseachers (Lynn Smith-Lovin & Miller
McPherson) + Matthew Brashaers whom I don't know, in sociology's leading
journal, American Sociological Review. I was a referee on this paper, btw,
and revealed myself at appropriate time to the authors.

The study replicates one stimulus Q from the General Social Survey 20
years ago about who do you have to discuss important matters with. It
finds a mean of about 2 in 2004, down from 3 in 1984 (or was it 1985)?

Here's the problem:

Based on this, the 2 newspapers have created a huge social isolation spin
on this, when it's well known that people have lotsa ties, not just 2 or
3. See our data from Pew Internet, Connected Lives study) + lotsa others.
Indeed, altho we are not as parallel as the US GSS, comparing Connected
Lives (3rd East York study, 2004) with the second East York study (1979
data) shows more active ties now, both intimate and significant ties.

Thus the media spin is a huge inferential leap from a decline in the
super-core ties to saying Americans are socially isolated.

If you read further in the Washington Post and USA Today articles, you'll
see me quoted as suggesting that we now have more ties -- and more contact
with ties -- but that relationships are differentiated. In other words,
the relative decline in discussion partners shouldn't be generalized to
either absolute isolation or even growing isolation.

Of course, YMMV.

  Barry Wellman         Professor of Sociology        NetLab Director

Date: Fri, 23 Jun 2006 13:56:00 -0400
From: Richard Forno
Subject: [Air-l] Americans and social isolation

It would be interesting to see how this fits into the whole "being alone
together" argument within a technology context....for example, someone who
doesn't want to be around people but who socializes happily in a MMORPG
environment or spends hours upon hours on AIM or IRC.

Washington Post story:
Social Isolation Growing in U.S., Study Says
The Number of People Who Say They Have No One to Confide In Has Risen

By Shankar Vedantam
Friday, June 23, 2006; A03

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.

A quarter of Americans say they have no one with whom they can discuss
personal troubles, more than double the number who were similarly isolated
in 1985. Overall, the number of people Americans have in their closest
circle of confidants has dropped from around three to about two.

The comprehensive new study paints a sobering picture of an increasingly
fragmented America, where intimate social ties -- once seen as an integral
part of daily life and associated with a host of psychological and civic
benefits -- are shrinking or nonexistent. In bad times, far more people
appear to suffer alone.


SOCNET is a service of INSNA, the professional association for social
network researchers ( To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.