Skip over navigation

Journal Issue: Children and Computer Technology Volume 10 Number 2 Fall/Winter 2000

Who's Wired and Who's Not: Children's Access to and Use of Computer Technology
Henry Jay Becker


  1. In the national survey, Teaching, Learning and Computing: 1998—A National Survey of Schools and Teachers (TLC-1998), data were gathered about teachers' pedagogy and practices and their use of computers. These data were gathered from 2,251 teachers in more than 600 schools in a national probability sample (a weighted representative sample), as well as from 1,800 other teachers in more than 500 schools specially selected because they had a high density of computer technologies or were systematically involved in instructional reform. These specially selected schools included those participating in 50 major national educational reform programs. Descriptive statistics from the TLC-1998 study calculated or reported for this article are based on the representative sample only; more analytic statistics, such as the relationship between computer use practices and conditions of use for teachers in specific subjects, also include the sample from the specially selected schools. Of all schools surveyed, 75% participated; 68% of teachers selected from the participating schools responded. See the project's homepage at
  2. The U.S. Census Bureau's Current Population Survey of U.S. Households' October 1997 supplement included data on computer use for 23,026 children ages 6 to 17 (plus 18- and 19-year-olds still attending high school). The December 1998 supplement included data on Internet use for 23,337 children of similar ages. The CPS methodology is described online at For the NTIA report of the 1997 CPS survey, see McConnaughey, J.W., and Lader, W. The digital divide: A survey of information "haves" and "have nots" in 1997 (also known as Falling through the Net II: New data on the digital divide). Washington, DC: U.S. Department of Commerce, National Telecommunications and Information Administration. For the NTIA report of the 1998 survey, see National Telecommunications and Information Administration. Falling through the Net: Defining the digital divide. Washington, DC: U.S. Department of Commerce, July 1999.
  3. Anderson, R., and Ronnkvist, A. The presence of computers in American schools. Teaching, learning and computing: 1998—A national survey of schools and teachers, report #2. Irvine, CA: Center for Research on Information Technology and Organizations, University of California, Irvine, and University of Minnesota, June 1999. Available online at
  4. Executive Office of the President of the United States, President's Committee of Advisors on Science and Technology. Report to the President on the use of technology to strengthen K–12 education in the United States. Washington, DC: U.S. Government Printing Office, March 1997.
  5. Internet access in U.S. public schools and classrooms: 1994–99. Stats in brief. Publication no. NCES 2000–086. Washington, DC: National Center for Education Statistics, February 2000. This is largely the result of schools having made the transition from systems of individual computers connected via dial-up telephone lines to high-speed direct Internet connections spread to many rooms simultaneously through a local area network of cables linking school computers and printers. NCES data indicate that the percentage of schools with dial-up connections decreased from 39% in the fall of 1996 to only 14% by the fall of 1999.
  6. "Low-SES" schools are defined here as schools with 40% or more of students eligible for assistance under the Education for the Disadvantaged program, also known as Chapter 1; "high-SES" schools are defined as schools with fewer than 10% of students eligible for such assistance. In a comparison of low- and high-SES schools, the TLC-1998 data indicate the percentages having a high-speed ("T1") Internet connection were respectively 23% versus 42%; the percentages having an Internet-connected computer-to-student ratio of 1 to 12 were 17% versus 44%; and the percentages having half of the school's classrooms with a high-speed Internet connection were 12% versus 29%. These are figures from prior to the government's "E-rate" program, which provides funding for schools and libraries to install wiring. See Appendix A by Roberts in this journal issue.
  7. Becker, H.J., Ravitz, J.L., and Wong, Y.T. Teacher and teacher-directed student use of computers and software. Teaching, learning, and computing: 1998—A national survey of schools and teachers, report #3. Irvine, CA: Center for Research on Information Technology and Organizations, University of California, Irvine, and University of Minnesota, revised December 1999, p. 7. Available online at
  8. The pattern of frequent computer use in middle schools and high schools is similar. The small differences in proportions reflected in the data are due, in part, to the fact that fewer applied courses are offered in the middle school than in the high school curriculum.
  9. For more detailed data on high-frequency computer use in middle and high school classes by school SES level and subject, see Becker, H.J. Who's wired and who's not. Teaching, learning and computing: 1998—A national survey of schools and teachers. Irvine, CA: Center for Research on Information Technology and Organizations, University of California, Irvine, 2000. Available online at
  10. Studies of instructional uses of school computers conducted in the 1980s and the early 1990s found that, except for practicing basic math skills and playing learning games as a supplementary "reward" activity, most school computer experiences during this period focused on making students "computer literate." Typical experiences included learning computer-specific skills such as programming (primarily prior to 1985) or word processing (primarily since 1985). However, systematic opportunities for students to apply such computer skills to academic tasks, such as developing or demonstrating an understanding of course subject matter, were much more rare. In 1992, 11th-graders spent only 31% of their school computer time for work in academic classes. Of their total computer time at school, 45% was in computer education activities, 17% in business education or vocational classes, and the remainder was recreational use. See Becker, H.J. Analysis and trends of school use of new information technologies. Paper prepared for the Office of Technology Assessment, U.S. Congress, under contract no. K3-0666.0. Irvine, CA: University of California Department of Education, 1994, Table 4.1.
  11. College Board trend studies of high school students taking the Scholastic Aptitude Test (SAT) for college admission support the data collected in the TLC-1998 study. As of 1997, some 75% of students taking the SAT had used computers for word processing; 45% had used them in some way in their English courses. However, only about 25% had used computers for math problem solving, and even fewer had used them for science (13%) or social studies (8%). On the basis of the trend data collected between 1987 and 1997, nearly all university-bound students will be using word processing in high school by 2000, but only a minority will use computers for problem solving in other classes: under 30% in mathematics, 20% in science, and 10% in social studies. See 1997 college-bound seniors: A profile of SAT program test takers. Princeton, NJ: College Board, 1997.
  12. By 1998, more than 24% of U.S. teachers were using the World Wide Web with students on a weekly basis. More than 66% of all teachers, across all subjects, were Web users themselves, primarily for identifying information resources for use in lessons or for use by their students. Becker, H.J. Internet use by teachers. Teaching, learning and computing, report #1. Irvine, CA: Center for Research on Information Technology and Organizations, University of California, Irvine, 1999, p. 4. Available online at
  13. To the extent that boys and girls have different patterns of choosing elective courses, the genders may experience computers in school differently; however, the evidence for such gender differences is mixed. Traditionally, girls have been underrepresented in vocational education classes, but they have been overrepresented in business education classes, where computers and increasingly sophisticated software are used a great deal. The TLC-1998 data do not include information about the different ways in which boys and girls use computers. Most data regarding gender differences in computer use derive from about 15 years ago, when systematic computer use was just beginning in American schools. See, for example, Becker, H.J., and Sterling, C.W. Equity in school computer use: National data and neglected considerations. Journal of Educational Computing Research (1987) 3:289–311. For a well-done ethnography of gender issues in the use of computers to teach a single subject (geometry) based on observations from 1985–87, see Schofield, J.W. Computers and classroom culture. New York: Cambridge University Press, 1995.
  14. Experiencing the newer and more creative uses of computers only in nonacademic classes is probably not as meaningful or constructive as experiencing intellectually powerful applications in core academic classes. However, curricular demands and structural constraints (especially) on high school academic teachers—as well as teachers' own objectives and priorities, discussed later in this article—have so far greatly impeded the integration of more demanding computer tools into teaching practice.
  15. In comparison, more than 80% of teachers of computer education classes and 67% of business education teachers had enough classroom computers to provide a ratio of at least one computer for every four students enrolled. Differences in classroom access to computers account for most of the differences between school subjects in how likely it is that students get to use computers on a frequent basis (discussed earlier).
  16. More specifically, 55% of classes with five to eight classroom computers used them frequently, compared with 21% of classes using computers in labs.
  17. About one-third of science teachers with five or more classroom computers had their students use spreadsheets or database programs on at least 10 occasions, compared with only 6% of science teachers whose classes had access to a large computer lab instead. Math teachers were less likely to have their classes use those types of software, but again, those with five computers in their classroom were much more likely to do so than those whose classes used a lab (12% versus 1%).
  18. Social studies teachers also tended to use more sophisticated software when five or more computers were in the classroom, but there were too few social studies teachers with such access in the TLC-1998 study (N=14) to include in the analysis.
  19. More specifically, 50% of classes with four or more Internet connections in the classroom used computers frequently, compared with 25% of classes with a single Internet-connected computer.
  20. See note no. 7, Becker, Ravitz, and Wong, p. 33.
  21. In addition, the extent of students' school-related use of computers outside the classroom was found to vary depending on the teacher's objectives for use of the technology. For example, by far the highest level of student computer use outside of class was reported by teachers whose objective was to help students present information to an audience. Well-above-average school-related use outside of class was also reported by teachers whose objectives were to help students improve writing, communicate with other people, and find out about ideas and information. Students reportedly used computers outside of class less often if their teachers' objectives were to reinforce skills, remediate skills, or help students learn to "work independently" (although it appears that some teachers may have used the phrase "to work independently" to mean that the students worked quietly rather than independently). See note no. 7, Becker, Ravitz, and Wong.
  22. See note no. 12, Becker.
  23. Findings concerning the difference between professionally engaged teachers and private practice teachers in terms of general pedagogy and computer use practices can be found in Riel, M., and Becker, H. The beliefs, practices, and computer use of teacher leaders. Teaching, learning and computing, special report. Irvine, CA: Center for Research on Information Technology and Organizations, University of California, Irvine, 2000. Available online at
  24. For the complete set of differences between high- and low-achievement-level classes, see snapshot no. 8 at the TLC-1998 Web site at
  25. In fact, teachers have received contradictory messages about how they should treat classes with different levels of prior preparation. On one hand, they are urged to individualize their instructional approach to fit students and classes. On the other, they are warned not to channel less demanding activities to their lower-performing classes and students or to limit the presumably more interesting work to their higher achievers.
  26. This has been true for many years. See note no. 13, Becker, pp. 289–311.
  27. For detailed data on the correlation between school-level SES and teachers' objectives for student computer use, see snapshot no. 7 at the TLC-1998 Web site at
  28. Excluding the effects of ethnicity (that are independent of measured socioeconomic factors), and excepting the slight disadvantage of students in southern states, other demographic differences in children's access to home computers—such as for one-parent households or homes in central cities or nonmetropolitan locations—can be accounted for almost entirely by the two variables of income and education.
  29. The two variables of income and education combined can explain nearly all the differences in children's access to home computers. For example, among children whose parents had less than a high school education and where family income was under $20,000, only 11% had a home computer. In contrast, for children in families with more than a $75,000 income and at least one parent with a master's degree, 95% had a computer at home and 80% had Internet access as well. Multiple regression analysis of these two variables produces somewhat larger beta coefficients for family income than for parent education (.36 versus .26). Regarding only the question of whether a child's household has Internet access, given that it has a home computer, neither family income nor parent education were quite as strong predictors as before, but again, income appeared somewhat more important (.20 versus .16).
  30. A rough indication of the extent of the inequality resulting from residential segregation by SES is provided in the TLC-1998 survey, which asked school principals if they thought the majority of their students had access to home computers. Only 4% of school principals from schools primarily attended by students from low-income families estimated that a majority of students had access to home computers, compared with 70% of school principals from schools primarily attended by students from professional or managerial families.
  31. See note no. 2, McConnaughey and Lader, charts 15a through 15d. Available online at
  32. Only for African-American and Hispanic children in high-income homes with parents having advanced degrees did technology access meet or exceed the level available to non-Hispanic white children from families with similar incomes. See Table 8 at Similar results were obtained through multiple regression analysis: The negative effects of being African American were a little larger than of being Hispanic (for example, betas for having a computer at home = –.20 for African Americans and –.16 for Hispanics). See Table 9 at the same Web site for the results of the multiple regression.
  33. Results from another recent survey indicate that parents' work-related Internet access is the primary determinant of home Internet access among computer-owning households. In that survey, the percentage of parents reporting that they use the Web at work at least every few days was twice as high among Internet-accessing households as among other computer households (41% versus 19%), while neither SES nor parent computer expertise related strongly to Internet access. See Turow, J. The Internet and the family: The view from parents, the view from the press. Philadelphia: Annenberg Policy Center, University of Pennsylvania, May 1999. In contrast, the 1998 CPS survey indicated that, for children with a computer at home, Internet access was more likely if parents had more education and income but not if they used the Internet at work. Differences in how work-based Internet use was measured may be responsible for these conflicting findings.
  34. The 1998 CPS survey data show that 51% of children ages 6 to 8 had a computer at their home, compared to 60% of adolescents (ages 12 to 17, plus those still in high school at age 18 or 19).
  35. According to the 1998 Roper Youth Report, adolescent boys were more likely than girls to have their own computer, by 17% to 10%. See Roper Starch Worldwide. Today's kids—especially teens—are wired to the hilt. Press release. New York: Roper Starch Worldwide, November 24, 1999. Results based on in-home, face-to-face interviews with 1,189 children ages 6 to 17. Available online at However, neither the TLC-1998 nor the Roper study found gender differences in likelihood or frequency of home computer use.
  36. According to 1997 CPS data, fewer than one-third of children whose parents were not high school graduates but who had a home computer had one that met the five criteria employed to measure functionality. In contrast, two-thirds of children in computer-present households where at least one parent had a bachelor's degree had a computer that was broadly functional.
  37. The effect of parent work use of computers was even greater when breadth of computer use at work was measured. Regression analyses indicate that the number of computer applications each parent used at work was as large a factor as education or family income in predicting whether a home computer would have Internet access or broad functionality or would be one of multiple computers in the home. For the multiple regression analysis, see Table 11 at
  38. Children and adolescents are defined here as schoolchildren ages 6 to 17, plus those still in high school at age 18 or 19. The data for most of the information about home computer access and use come from the U.S. Census Bureau's Current Population Survey of U.S. Households, 1997 and 1998 supplements.
  39. See note no. 3, Anderson and Ronnkvist. If K–3 teachers (not surveyed in the TLC-1998 study) had classroom computers in the same proportion as fourth- through sixth-grade elementary teachers, the overall percentage of teachers with at least one computer in their classroom as of 1998 would rise to 59%—slightly higher than the percentage of children's families with a computer at home.
  40. An analysis of CPS data indicates that in 1998, the average person-to-computer ratio in children's computer-owning households was 3.9 to 1. In contrast, the TLC-1998 data indicate that in classrooms with any computers at all, the mean student-to-computer ratio was about 14 to 1.
  41. Friends' homes and libraries are other important nonschool locations for additional computer use. The Roper-Annenberg survey found that 12% of 6- to 17-year-olds used computers in the homes of friends and relatives during the previous month, and 5% used computers in public libraries. See note no. 33, Turow. The National Assessment of Educational Progress (NAEP) provides quite different information about the proportion of adolescents who use computers at the homes of friends and in libraries, however. The 1996 NAEP study indicated that more than 50% of students used a computer at a friend's house, and more than 60% of students (more than 70% at the 11th grade) used a computer at a library. See Campbell, J.R., Voelkl, K.E., and Donahue, P.L. NAEP 1996 trends in academic progress. Report no. NCES 97- 985. Washington, DC: National Center for Education Statistics, September 1997. The contrast probably results from the fact that the NAEP question did not specify a time period (such as "during the previous month"), and the differences in response styles between oral interviews (Roper-Annenberg) and questionnaire self-reports (NAEP).
  42. In 1998, nearly all home Internet access was provided by relatively slow dial-up modems rather than the high-speed, always-connected networks that are becoming increasingly common at schools and in some residential communities.
  43. Although adolescents were still more likely to use the Internet from home, younger children showed the greatest increases in their level of participation during this period. Home Internet use increased from 3% to 10% for children ages 6 to 8, from 12% to 21% for children ages 9 to 11, and from 19% to 29% for children ages 12 to 14. Meanwhile, home Internet use by high-school-age adolescents, although still more widespread than use by younger children, went up by only 5 percentage points, from 27% to 32%. The Roper-Annenberg Foundation study from earlier in 1998 found similar percentages regarding children's home access to the Internet.
  44. See note no. 33, Turow, p. 17.
  45. The 1998 Roper Youth Report, conducted in spring 1998. See Roper Starch Worldwide. Kids favor Internet for homework, chatting and surfing: As in real life, girls are greater virtual socializers. Press release. New York: Roper Starch Worldwide, February 22, 1999. Available online at
  46. The 1998 Roper Youth Report found that the proportion of children and teens who used the Internet at least occasionally (in home, at school, or other) was greater than the proportion of adults who did so, even young adults in their twenties. See Roper Starch Worldwide. America is fastest-growing Internet market, as teens lead the way. Press release. New York: Roper Starch Worldwide, April 26, 1999. Available online at In addition, a panel study of families' home use of the Internet in Pittsburgh (the HomeNet project) found that teenagers used the Internet much more than their parents. See Kraut, R., Mukhopadhyay, T., Szczypula, J., et al. Communication and information: Alternative uses of the Internet in households. Pittsburgh: Carnegie Mellon University, July 1997. Available online at See also Kraut, R., Lundmark, V., Kiesler, S., et al. Why people use the Internet. Pittsburgh: Carnegie Mellon University, April 1997. Available online at
  47. See note no. 46, Roper Starch Worldwide. Among those with online access, 41% of teens ages 13 to 19 reported using online services, compared with 33% of adults ages 30 to 39, 31% of adults ages 20 to 29, and 31% of adults ages 40 to 49. In addition, a panel study of families' home use of the Internet in Pittsburgh (the HomeNet project) found that teenagers used the Internet much more than their parents. See note no. 46, Kraut, Mukhopadhyay, Szczypula, et al.; and Kraut, Lundmark, Kiesler, et al.
  48. The HomeNet project also found that recreation and communication activities dominated teens' use of the Internet. Teens were less likely than adults to use the Internet for getting product information, to purchase items, to read the news, or to view sexually oriented materials. See note no. 46, Kraut, Mukhopadhyay, Szczypula, et al.
  49. See Table 12 at for a summary of the results of the standardized multiple regression (beta) coefficients linking each contextual factor with each aspect of home computer use.
  50. Note that the percentage of all children who used home computers in these various ways is only half the percentage of children in home-computer owning households shown in Figure 9 and discussed in the text.
  51. For a number of reasons, only one variable was examined for each member of a child's family: the parents' use of computers at work, the child's own use of computers at school, and the child's siblings' use of computers at home. It is likely that the attribution of "impact" belongs as much to the people involved as to the context in which their use of computers is measured.
  52. Results from the Roper Youth Report, conducted annually, found similar percentages of girls and boys using computers (not only in the home, but also in places such as friends' homes, the library, and part-time work settings) for the first time in 1998. See Roper Starch Worldwide. Kids computer use stabilizes; gender gap disappearing. Press release. New York: Roper Starch Worldwide, February 16, 1999. Available online at roperweb/news/content/news106.htm.
  53. Gender differences for some types of home computer use were somewhat larger for adolescents—for example, playing games was reported for 68% of teen boys but only 55% of teen girls.
  54. Gender differences in types of home Internet use were generally no greater than two percentage points. Largest differences were in retrieval of news and sports information, with 17% of boys in households with computers, but only 11% of girls likely to use the Web for that purpose.
  55. Attitudinal and stylistic differences between girls and boys are explored in Brunner, C., Bennett, D., and Honey, M. Girl games and technological desire. In From Barbie to Mortal Kombat: Gender and computer games. J. Cassell and H. Jenkins, eds. Cambridge, MA: MIT Press, 1998.
  56. For example, Ford Motor Company provided its 350,000 employees with home computers, color printers, and unlimited access to the Internet for the nominal fee of $5 per month. See Brown, W., and Swoboda, F. Ford offers home PC to every employee; $5-a-month plan reflects firm's focus on Internet. Washington Post. February 4, 2000, at A1.
  57. Word processing is one of several types of computer applications that have replaced skill practice as the primary focus for middle and high school computer use, apart from teaching about computers. Word processing is by far the most common type of computer assignment, not only in English class, but in almost every other subject as well.