Internet, Survey Research About

Organizations Researching the Internet through Social Surveys

The main organizations engaged in concerted programs of Internet research using the method of the social survey at the national level are described below.

• 1.
  The Pew Center for the Internet and Daily Life provides perhaps the largest continuing amount of survey content related to the Internet, with monthly samples of 2,000 national respondents during March 2000 and March 2001, and surveys on a variety of topics on a more irregular schedule since then. Some surveys have included youth under 18 years of age. A complete account of their many study topics, and instructions for accessing their raw data, can be found at http://www.pewinternet.org. The related Pew Center for the Study of Public Opinion (http://www.pewcenter.org) conducted pioneering studies of Internet use in 1995 and 1998, largely focused on the relation of Internet use to use of broadcast and other media, thus making the Pew Centers perhaps the longest continuing series of Internet-related surveys available.
• 2.
  The National Telecommunications and Information Administration (NTIA) surveys are the source of the well-known “Digital Divide” reports, which document trends in the diffusion of the Internet (particularly among various minority groups). Its surveys are conducted as part of the Current Population Survey (CPS) of the U.S. Census Bureau, with all members of 50,000 households across the country being canvassed. While it thus represents the largest and highest quality of national samples available, the NTIA data are limited by their use largely of dichotomous (“yes” or “no”) questions that do not provide insight into the extent of user behavior in relation to the Internet.
• 3.
  The General Social Survey (GSS) began collecting Internet data from its in-home sample of more than 2,800 respondents in the year 2000; and the Internet module of five to twenty minutes worth of questions which they used was repeated in the year 2002. The unique value of the GSS is its large collection of opinion questions from a high-quality sample to correlate with usage; the year 2000 GSS respondents were reinterviewed as part of a panel study in the fall of 2001.
• 4.
  The UCLA “Surveying the Digital Future” surveys began in 2000 with a national telephone survey of just over 2,000 respondents aged twelve and older, and were repeated in 2001 and 2002 with a panel component. Major topics addressed included types of Web sites visited, consumer activity, and use of other media. A unique feature of the UCLA project is its large number of foreign partners (over twenty countries) who have conducted surveys using its core questions.
• 5.
  The worldwide Clemson University/University of Toronto surveys, which were done in the fall of 1998 in conjunction with the National Geographic Society, included an extensive set of questions and a large multinational sample of more than 20,000 respondents. In 2001, they were revised and repeated in multilingual format with a sample of 5,000. Because respondents volunteered to fill out the questionnaire, the sample is self-selected; however, it was supplemented in 2002 with a cross-sectional survey.
• 6.
  The Rutgers University Project Syntopia surveys have been conducted in 1995, 1996, 1997, and 2000 with national samples of 557 to 2500 respondents, focusing on issues of political and community involvement.
• 7.
  The SIQSS (Stanford Institute for the Quantitative Study of Society) surveys at Stanford University began in 2000, examining self-reported changes in social and media activity as a result of Internet use, and in 2001 and 2002 expanded to include all daily activity in the form of complete daily time diaries. Sample sizes have ranged between 3,000 and 8,000 respondents aged eighteen and older. In order to draw its panel, SIQSS interviews large samples of respondents on a regular basis using random-digit dialing (RDD) methods.
• 8.
  The widely-cited studies at Carnegie-Mellon University have relied on smaller, local samples in the Pittsburgh, Pennsylvania area, largely examining patterns of interpersonal communication. These studies are distinguished by their use of careful experimental panel designs to draw conclusions about Internet impact.
• 9.
  Several national studies conducted at UC Santa Barbara tracked increases in both Internet usage and political usage in 1996, 1997, 1998, and 2000.

Interestingly, the topic of Internet access has not been much examined by the major polling organizations of Gallup, Roper, Harris, Zogby, or Yankelovich—or by the media polls conducted by television networks or major newspapers. Undoubtedly, much Internet research has [Page 785]been conducted by commercial polling firms for their many clients, but almost none of it is publicly available or even publicized. Much data on specific usage of the Internet has been collected by ratings firms, like Media Metrix and Comscore, but the largest ongoing tracking now seems done by Neilsen NetRatings, with its national RDD-based survey of 50,000 respondents whose usage is tracked electronically from their own PCs. Changes across time and analysis by lifestyle segments are conducted by the Forrester research company, while Nua, which surveys Internet trends, compiles much international usage statistics at http://www.nua.com.

...