273. "Debating the Societal Effects of the Internet: Connecting with the World," Public Perspective, Vol. 11, No. 3 (May/June 2000), pp. 42-43.


One can readily sympathize with Professors Norman Nie and Lutz Erbring, the investigator and co-investigator of a recent study on the social consequences of the internet conducted by the Stanford Institute for the Quantitative Study of Society. Like many scholars before them who have conducted extensive surveys, their results at first seemed rather self-evident and dull. They spent much effort and resources to reach 4,113 adults in 2,689 households. They analyzed their data and came up with such findings as the internet is used more for e-mail (90%) than banking (12%), more people use the internet for surfing (69%) than for trading stocks (7%), and those who use it extensively spend less time in traffic (14% of heavy users).

What is one to do about such findings? Some survey masters have been very skillful at coming up with post hoc interpretations of their data. They tease out findings based on factors initially not included in the design of their studies, or they attach compelling concepts to the data without necessarily claiming that they were based on the evidence itself. A well-known example is the "discovery" of reference groups by Robert K. Merton and his associates in the famous Samuel Stouffer data, included in the American Soldier series. Another fine example is the introduction by Paul F. Lazarsfeld of the concept of two-step communication and opinion leaders in his voting studies.

Unfortunately, Professors Nie and Erbring followed a less revered route. They provided a summary of their study to the media, which got front page attention but actually conflicted with their findings. Their major conclusion was that people who use the internet cut themselves off from family and friends, diminish their social lives, and become isolated and lonely. Professor Nie told the New York Times that "the more hours people use the internet, the less time they spend with real human beings." In a press release, he further explained that his study focused on the emotional effects of "more people being home, alone and anonymous." Nie also stated that "The internet could be the ultimate isolating technology that further reduces our participation in communities even more than television did before it." Finally, discussing the ethics of business dealings, Nie said that "When we lived in small communities, the old story was that you said to yourself, 'I'll see this guy and his wife at church on Sunday, so I better be honest with him today.' Now, it's becoming, 'Hell, I won't ever even know this guy's name.'" And on the NewsHour with Jim Lehrer, Nie declared that "as we've moved through the results of many of these technological inventions, the crisis of modernity is aloneness and anomie." These claims landed the study on the front pages of the New York Times and the Washington Post, and on CNN and NPR, among others.(1)

What do Nie's and Erbring's figures actually show? In discussing the findings one must note that they concern two groups of people: those who are not connected to the internet (N=2078) and those who are connected (N=2035). The latter are further divided into light users (less than 5 hours per week; 64% of the "connected" sample) and heavy users (more than 5 hours per week; 36% of the "connected" sample).

Of all users, only 9% said that they spend less time with their families and 9% less with their friends, while nearly ten times more people -- 86% and 87% respectively -- said that they spend the same amount of time with family and friends as before! Moreover, quite a few (6% and 4% respectively) reported that they spend more time with family and friends. The proper headline of their study should have been: "Internet does not significantly affect social life."

The picture does not change much if one focuses on the heavy users. Only 10% of those who spend 5 to 10 hours online per week reported that they spend less time with family and friends, and only 15% of those who were online 10 or more hours per week said so.

The finding that some internet users spend more time with family and friends may at first seem unlikely, but it is hardly so. The study itself shows that by far the largest effect of internet activity for all users is to reduce the amount of time spent watching TV (46%) and shopping (19%). For heavy users, 59% spend less time watching TV, and 25% spend less time shopping. (Obviously it takes less time to order things from eToys or Amazon than to go to a mall or store.) The study did not inquire how the time saved in these ways was used or whether or not some of it was allocated to increasing social life.

Along the same lines, the study found -- as is widely known -- that people's most common use of the internet is communicating via e-mail. This, too, is a time-saving device compared to letter writing and even phone calls. Ergo, the internet readily allows people to spend both more time on the internet and more time socializing. In effect, the fact that people use the internet largely for communication and not shopping or banking, and that most of this communication is with people they are familiar with rather than with strangers, strongly suggests that people relate to one another more rather than less because of the internet.

Surveys often do not include control groups or groups that allow meaningful comparisons. One of the merits of the Nie-Erbring study is that they included about as many non-users (2,078) as users (2,035). However, they did not make the obvious comparisons. Thus, while they show that 21% of all users work more at home because of the internet, we do not know whether those not connected to the internet also work longer hours, which seems to be a recent American trend. And for the same reason, we do not know if some of these people, too, have spent less time with their families and friends in recent years as compared to before. Given that so few of even the heavy users report fewer social contacts, such a comparison would be particularly revealing.

The authors of the study provide a strong reason why they ignored the control or comparison group: the study provided them with a free connection to the internet and e-mail accounts and equipped them with WebTV setup boxes. Most scholars would agree that such a giveaway might well contaminate the non-users. But then the question arises, why include them in the first place? It seems that the study plans to return to the same sample of both users and non-users annually to establish trends, as suggested to me by one of the team members. This would fly in the face of all we learned over two generations, since the famous British fiasco, about repeat visits to the same panel. Whatever the plans, the methodology of the study is puzzling.

The study shows that far from being isolated, a large number of the users -- 24% -- draw on the internet for communication with others, such as in chat rooms or on message boards. These include many thousands of virtual communities run by Yahoo!, Excite and AOL. While some are rather superficial forms of social bonds, others -- for instance those that bond people who share an illness or some other social problem -- are surprisingly affective and effective.(2)

Everyday experience shows that people use the internet to connect in three different ways: to reinforce existing relations among family, friends, and coworkers; to forge new relationships, which can be quite intense, as shown by reports of people falling in love online; and to join or form communities, some of which closely bind people who share an illness or an alcoholic in the family or some other serious challenge life has thrown their way.

The study further disregards that for people who are not mobile, for single parents with children, and for people who fear the streets at night, the internet is often a major way to form or maintain social bonds.

As to a suggestion by Nie that on the internet you do not hear the voice of others and hence miss nuances of communication, voice is rapidly being made transmittable over the internet. Of course, phone calls are already made over the internet (including conference calls) and -- most relevant -- Excite already offers groups that meet on its site the capacity to hear how the group is responding. Moreover, while Nie is right that you will never "share a coffee or a beer with somebody on e-mail or give them a hug," on the internet you aren't walking city streets in fear of falling victim to violent crime, either. I do not mean to resurrect the image of our cities as jungles. But for millions who are infirm, truly old, or otherwise handicapped or anxious, the internet provides a reassuring access to the social realm, day or night, sleet or shine.

True, people often do not display their real selves when first contacting strangers on the internet. But as the saying goes: "No one is a dog on the internet," which means that people are given a chance to try out and develop different selves. And those who have found dates and spouses this way report they like the fact that they first got to know one another before they saw each other. As a result, they say, appearances matter less.

The internet, like other new technologies, changes our lives, and not all for the better. However, claims that it increases our social isolation are wholly unsupported, especially by this study.

1. I rely on statements made to the media because when the findings were initially released Nie and Erbring provided little write-up of their findings on their web page. Hence, one must draw on their statements to the press to establish what they claim to have found.

2. Amitai Etzioni and Oren Etzioni, "Face-to-Face and Computer-Mediated Communities, A Comparative Analysis," The Information Society 1999, Vol. 15, No. 4, p. 241-248, and A. Etzioni and O. Etzioni, "Communities: Virtual vs. Real," Science, July 18, 1997, p. 295. Also see A. Etzioni, "E-communities Build New Ties, but Ties that Bind." New York Times, February 10, 2000, p. E7.

The Communitarian Network
2130 H Street, NW, Suite 703
Washington, DC 20052
202.994.6118
comnet@gwu.edu