CYBERPSYCHOLOGY & BEHAVIOR Volume 8, Number 5, 2005 © Mary Ann Liebert, Inc.

Instructional Set and Internet Use by Low-Income Adults
LINDA A. JACKSON, Ph.D., ALEXANDER VON EYE, Ph.D., FRANK BIOCCA, Ph.D., GRETCHEN BARBATSIS, Ph.D.,YONG ZHAO, Ph.D., and HIRAM FITZGERALD, Ph.D.

ABSTRACT This research examined the effects of instructional set on Internet use by low-income adults during a 16-month longitudinal study. Participants (n = 117) received instructions that focused on either the Internet’s communication tools or its information tools. Internet use was continuously and automatically recorded. Survey measures of computer and Internet experiences, affect and attitudes were obtained to examine their mediational role in the relationship between instructional set and Internet use. Results indicated that instructions focused on the Internet’s information tools led to greater Internet use than instructions focused on its communication tools or only basic instructions about how to use the Internet. Implications for reducing the digital divide are discussed.

INTRODUCTION

H

OMENETTOO is a longitudinal field study designed to examine the antecedents and consequences of home Internet use in low-income families (<www.msu.edu/user/jackso67/homenettoo>). The study began in 2000 when 90 families received home computers, Internet access and in-home technical support in exchange for allowing their Internet use to be continuously and automatically recorded, participating in home visits, and completing surveys at multiple points during the 16month trial. This report focuses on whether instructions about how to use the Internet provided at pre-trial, 1 month, and 3 months influenced subsequent Internet use. Also considered is the mediational role of early computer and Internet experiences, affect and attitudes on the relationship between instructional set and Internet use. Previous research suggests that communication and information are the primary motives for using the Internet, and that communication is generally the stronger of the two.1,2 Indeed e-mail has been referred to as the “killer application” of the Internet.1,3–8 According to researchers at

Carnegie Mellon University: “Despite the hoopla surrounding the WWW, and not to dismiss its power in many domains, it is possible that interpersonal communication is driving the average person’s use of the Internet.”1 In support of the view that communication is the primary motive behind Internet use, national surveys indicate that the overwhelming majority of people who use the Internet at home use it mainly for e-mail.9–12 Even information search is often motivated by the need to communicate with others. The most popular reason for using a search engine is to find other people.13 On the other hand, the monumental success of the World Wide Web attests to the importance of information motives in everyday use of the Internet. The Web is quickly replacing traditional sources of information (e.g., newspapers), especially among today’s youth.2,8,12 With virtually no limit to the number of Web pages in sight, the Web will likely become the primary source of information for everyone who has access to it. In the HomeNetToo project we examined whether focusing instructions about how to use the Internet on either its communication or informa-

Department of Psychology, Michigan State University, East Lansing, Michigan.

465

Following the tutorial the participant was guided through the process of sending an e-mail to himself or herself and project staff. the 16-month trial. relocation to another city (two families).19–28 few studies use longitudinal designs that permit a consideration of cause-effect relationships. medium-sized urban community in the mid-western United States. Michigan. Internet and e-mail. where he or she selected one of three online tutorials about how to use e-mail. Online tutorial options are presented in Table 1. To participate in the project a family had to have home telephone service for at least the previous 6 months. and handouts prepared by project staff about how to use the computer.5 h. tion tools would influence Internet use in the months that followed. the only restriction being that all members of the same household participate in the same condition (i.. Instructional set Also examined in this research was whether computer and Internet experiences. net household income). Lansing. Participants were randomly assigned to instructional set condition. a group neglected in much of the research on Internet use. Only four families did not complete the 16-month trial. save and delete an email.8.4 MATERIALS AND METHODS Participants Adult participants in the HomeNetToo project were 117 residents of a low-income. Participants were recruited at meetings held at their children’s middle school and at the Black Child and Family Institute.8. Based on previous research the following hypothesis was formulated: Instructions on Internet use that focus on its communication tools will result in greater Internet use than instructions that focus on its information tools. During this activity the technology facilitator pointed out how new e-mail was indicated on the screen. how to open. female (80%). and difficulties monitoring teenager’s use of the Internet (one family). The majority of participants reported having some college education or earning a college degree (62%). At one month the participant. Both types of instructions will results in greater Internet use than only basic instructions about how to use the Internet. including computer and Internet experiences and attitudes.” a facilitator-guided demonstration of how to turn on and off the computer and how to logon to the Internet and use email. repeated the process . attributable to death (one family). Invitations were extended to all parents whose children were eligible for the federally subsidized school lunch program. Home visits lasted about 1. Participants in the project were low-income adults.14–18 Lowincome adults are of particular interest not only for testing the generalizability of previous findings but also for identifying barriers to Internet use that may be unique to this group. Although there is abundant research on the correlates of Internet use.6 years old.466 JACKSON ET AL. and agree to completing surveys and participating in home visits at multiple points during Communication condition. and how to view contents of the sent folder. Participants in the communication condition were first given basic instruction about how to use the Internet. together with the visiting technology facilitator. and earning less than $15. households were randomly assigned to conditions). Fewer still use continuously and automatically recorded measures of Internet use. Demographic characteristics obtained at pre-trial indicated that participants were primarily African American (67%).17 Average age of participants was 38. A total of 90 families participated in the project. how long it takes (typically) between sending and receiving an e-mail. Basic instruction consisted of verbal instructions from the visiting undergraduate “technology facilitator.e.7. At pre-trial participants in the communication condition. indicating that our sample was better educated than is typical of low-income samples. Approximately 140 children of the 117 adult participants discussed in this report also participated in the project but completed different surveys. navigated the project’s web site to an Online Activities drop-down menu. affect and attitudes mediate the relationship between instructional set and Internet use. Also necessary was that participants had never before had home Internet access.000 USD annually (49%. never married (42%). with guidance from the visiting technology facilitator. Procedures Three instructional set conditions were created by varying the Internet activities that participants engaged in during home visits at pre-trial. and 3 months. and all adult family members were required to participate in them.3. 1 month. consent to having their Internet use continuously and automatically recorded.

htm Learn the Net E-mail: http://www.INSTRUCTIONAL SET AND INTERNET USE 467 that occurred during the pre-trial home visit. and information tools often involve communication. the participant. except that the topic of instruction was newsgroups.learnthenet.com/english/section/ Mailing lists How To Use Mailing Lists: http://www. At three months. number of domains visited (per day) and number of e-mails sent (per day).newbie.learnthenet.learnthenet. e-mail. We categorized Internet tools as communication or information based on their primary use and on how personal they appear to the user. Although mailing lists provide information. they received verbal instructions from the visiting technology facilitator. and emphasizes information rather than communication. Web search and newsgroups as information tools.com/searches. Participants in this condition received only basic instructions about how to use the computer and Internet. the process was repeated again.gov/global/explore.webnovice.imaginarylandscape.html Navigating the Internet: http://www.actden.bizland.learnthenet.com/searches. Internet communication tools often serve information functions.com Talkcity.deja.htm Explore the Internet: http://lcweb. 1 month.org/reference/e-mail. their mode of delivery and posting is e-mail. At 3 months.com iVillage. and 3 months. The online tutorial options for mailing lists are indicated in Table 1. their mode of delivery and posting is impersonal.bexta. except that the topic of instruction was search engines. together with the visiting technology facilitator. Internet and e-mail. ONLINE TUTORIALS E-mail Newbie Guide to E-mail: http://www.telecom.com/english/html How NewsGroups Work: http://www. at pre-trial. At 1 month the process was repeated. To examine TABLE 1.com: http://www. and the Web.htm Search Engine Fundamentals: http://www. Thus.loc. and handouts prepared by project staff about how to use the computer.com/english/animate/forums.com Chat http://www.com/usenet . identical to those received by participants in the communication condition. the same process was repeated except that the topic of instruction was Internet chats (Table 1). Thus. number of sessions (logins per day).co. Internet and e-mail.com/search. mailing lists and chat were categorized as commu- nication tools. Although newsgroups provide a means of communicating.com Web search Search Engines & Directories: http://www.nz/about_telecom/tutorial Newsgroups Learn The Net: Newsgroups: http://www. Information condition.html Microsoft E-mail: http://www. at pretrial.htm Telecom Tutorials: Searching: http://www2. Participants in the information condition first received basic instructions about how to use the Internet.com: http://www. Basic instructions condition. the quintessential communication tool of the Internet. navigated the project’s web site to the Online Activities dropdown menu where he or she selected one of the online tutorials about the WWW (Table 1).net/ Chats Yahoo Chat: http://chat.yahoo.com/skills2k/e-mail. Then. except that the topic of instruction was mailing lists rather than e-mail.neolink.html Exploring Usenet Groups: http://www. Internet use Four measures of Internet use were automatically recorded for 16 months for each participant: time online (minutes per day).com/english/animate/ Electronic Mailing Lists: http://www. a facilitator-led tour of how to use the computer.com/helpweb/ Index of Mailing Lists: http://paml.com/chat/ WWW Webnovice.webnovice.talkcity.ivillage.

02. time 2 (4–6 months). Univariate statistics revealed main effects of instructional set for time online. previous experience/skills using the Internet (three items. Using computers helps children to do better in school.09. Internet affect (10 items). Mode = 1. Overall. How anxious (nervous) do you feel when you are doing. 49) = 0. time 3 (7–9 months).g. F(2. number of sessions. 5 = very good skills) and demographic characteristics (race.e. and number of domains visited. Participants in the information condition engaged in more Internet ses- RESULTS Preliminary analyses Before examining the effects of instructional set on Internet use we examined whether there were . how would you rate your skills at using the Internet? 1 = no skills. 5 = a great deal of experience). changes over time each of these measures were divided into five time periods. Because of high variability and skewed distributions. e.99. 5 = strongly agree). negative. Computer attitudes (four items.. Survey measures Surveys administered after the three-month home visit included the following measures. 5 = strongly agree). Means for the instructional set groups are presented in Table 2.g. number of sessions. M = 2. Getting an “error message” on a computer? 1 = not at all anxious. 1 = strongly disagree. Md = 1.. which did not differ from each other. 1 = strongly disagree. negative affect (five items. age. e. log transformations of Internet use measures were used in all analyses.468 JACKSON ET AL. e.71. e. positive affect (five items. Internet attitudes (nine items..25) and previous experience/skills using the Internet (alpha = . 1 = strongly disagree.70).01. Using the Internet helps children to do better in school. technophobia) and Internet affect (positive and negative) were computed.g.53). technophobia (11 items. previous experience/skills using the Internet was the measure used in subsequent analyses.. p < .67. How successful have you been in using your computer? 1 = very unsuccessful. time online. number of e-mails sent) during each time period (i.72.14.48.g. Internet experiences (five items. indicating good reliability. Composite scores (arithmetic averages) were computed for previous experience/ skills using computers (alpha = . Participants in the communication condition had more Internet experience/skills (M = 2. 1 = strongly disagree. number of domains visited. p < .g. gender.. marital status. but not for number of e-mails sent. sd = 1. F(2...85) or basic instructions conditions (M = 1.. e. Md = 2.g. 5 = strongly agree). p < . previous experience/skills using the Internet was included as a covariate in the analyses examining the effects of instructional set on Internet use.. One-way analysis of variance (ANOVA) indicated differences among the instructional set groups in their previous experience/skills using the Internet. Therefore. 49) = 4. 5 = strongly agree).15.. p < .91.g. three that corresponded to survey administration points plus half-year and 1-year points.05. 5 = very successful). I enjoy using the Internet. e.001. I feel tense when I use the Internet. F(2. Using a computer is fun. time 5 (13–16 months).49) = 3.g.05. time 1 to time 5).03).05 (Wilks’ Lambda).g.22.. income. 5 = very successful).80 to .110) = 7. pre-existing differences among the instructional set groups in their experience/skills using computers and the Internet. A significant multivariate effect of instructional set was obtained at time 2.g.e. The time periods were as follows: time 1 (1–3 months). Previous experience/skills using the Internet was included as a covariate.89. negative affect (five items. Because these composite measures were highly correlated (r = . Instructional set and Internet use Multivariate ANOVAS were used to examine the effects of instructional set on Internet use (i.80. e. e. education). How successful have you been in finding information on the Internet? 1 = very unsuccessful.46) = 2.92. 5 = very anxious).g. Pre-trial measures were obtained of previous experience/skills using computers (three items. Computer affect: positive affect (five items. p < . F(2. Composite measures (arithmetic averages) of computer affect (positive..42) than did participants in the information (M = 1. and because the focus of this research was on Internet use rather than computer use. sd = 1. 5 = strongly agree). or thinking about doing each of the following? e.29 Computer experiences (three items.84. How would you rate the extent of your experience with computers? 1 = no experience. 5 = strongly agree). F(2. time 4 (10–12 months). I get nervous when I use a computers.. F(2.94. Coefficient alphas for these measures ranged from . p < . M = 1. e. 1 = strongly disagree. e. Mode = 1. adapted from previous research:3. 1 = strongly disagree. Participants in the information condition spent more time online than did those in the communication or basic instructions conditions. 49) = 4.

INSTRUCTION SET AND INTERNET USE Information (n = 43) Mean 56. Results indicated that neither e-mail success nor attitudes about the importance of computers to children’s school success predicted Internet use.41 0.103) = 3.103) = 3. affect and attitudes. sd. HomeNetToo participants seldom used e-mail.69 2. Evidence presented elsewhere indicated that chat and instant messaging were similarly infrequent activities for both adults30–32 and children in the project.44 0.INSTRUCTIONAL SET AND INTERNET USE 469 TABLE 2.53.05). Consistent with predictions.72 sd 127. Although differences did not reach statistical significance for the other time periods. affect and attitudes Given the evidence that instructional set influenced Internet use. sending an average of only two to three e-mails per week throughout the 16-month trial. but not compared to those in the basic instructions condition (M = 3.82 13. instructions that focused on the Internet’s information tools resulted in greater Internet use than only basic instructions about how to use the Internet. with previous experience/skills using the Internet as a covariate. and results of analyses of survey measures obtained at nine months and post-trial may be found in Jackson et al.11 0. contrary to predictions. F(2.30 DISCUSSION Instructions about how to use the Internet provided to low-income users influenced their Internet use in the months that followed. Participants in the information condition used the Internet most.33 Internet use was automatically recorded.15. 4.09 sd 74. they would be viable candidates for mediating the relationship between instructional set and Internet use. Next we examined whether e-mail success and attitudes about the importance of computers to children’s success in school predicted Internet use at time 2.25. sions and visited more domains than did those in the basic instructions condition.24) than did those in the information condition (M = 2.33 Thus. p < .56 . then these measures may mediate the relationship between instructional set and Internet use.82 11.50). Instructional set influenced computer experiences and attitudes. If so. If so.89 1. Thus. respectively) than did participants in the basic instructions condition (Ms = 4.96 12. standard deviation. regardless of which measure of Internet use was considered.52 0. Participants in the communication and information conditions believed that computers were more important to children’s success in school (Ms = 4. instructions that focused on the Internet’s communication tools did not result in greater Internet use than basic or information-focused instructions.05). sample size.01. controlling for previous experience/skills using the Internet. we examined whether instructional set also influenced computer and Internet experiences. Regression analyses were performed separately for each measure to predict Inter- net use.70 0. However.66 0. There were no differences among the three instructional set groups on pre-trial measures of computer and Internet affect or attitudes.21 sd 70.12 21.23 Communication (n = 38) Mean Time online (min) Number of sessions Number of domains visited Number of e-mails sent 42. and those in the basic instructions condition used it least.12. MANOVAs were performed for each set of measures. Participants in the communication condition reported more success using e-mail (M = 3.12. Values are per day at time 2 (4–6 months) n.18 0.31 0.55 7. e-mail success and beliefs about the importance of computers to children’s school success cannot explain the relationship between instructional set and Internet use. they were consistently in the same direction. use of these . Mediational effects of computer and Internet experiences. contributing to e-mail success. F(2. p < .18 Basic instruction (n = 35) Mean 29. One explanation for the failure of communication-focused instructions to increase Internet use may lie in the infrequent use of the Internet’s communication tools. Results of analyses to examine the effects of demographic characteristics on Internet use and survey measures.46 0. while instructions about how to use the Internet’s communication tools were clearly helpful.99 14.

ACKNOWLEDGMENTS This research was supported by a grant to L. Chat. As alternative explanation for the infrequent use of the Internet’s communication tools observed in our sample is that using such tools tends to require more complex typing and writing skills than are required to simply find information. questions asked during the instructional sessions) Participants made good use of the Internet’s information tools (i. Although attitudes about the importance of computers to children’s school success did not influence parents’ Internet use.. Yet another possibility is that the instructions provided in the information condition were somehow “better” than those provided in the communication condition. This is not to say that communication functions should be ignored. evidenced in our ethnographic data. Their families and friends were likely to be poor. then the digital “use” divide may be widening at the very same time that the digital “access” divide is narrowing. and the extent to which parents encourage their children’s computer and Internet use. These findings underscore the importance of considering socio-economic status in research on Internet use.3 e-mail may not be the driving force behind Internet use for all socioeconomic groups. Moreover. 085348). Our findings suggest that factors influencing early Internet use may become less important over time.A. HomeNetToo adults were not employed in occupations that required or encouraged e-mail communication with colleagues. Instructional set influenced participants’ computer experiences and attitudes during the first three months of home Internet access. it is not surprising that HomeNetToo participants made so little use of the Internet’s communication tools. Instructions focused on the Internet’s communication tools resulted in greater success using e-mail than instructions focused on information tools. HomeNetToo participants were poor.31 Findings from the HomeNetToo project underscore the importance of longitudinal studies to understand everyday Internet use. Taken together with a general reluctance to communicate with strangers. While we cannot rule out this explanation entirely. tools was too infrequent to contribute to overall Internet use. in the absence of family. it may still be the case that contrary to previous claims. If low socioeconomic groups are leaving the Internet at more rapid rates than are higher socioeconomic groups. Thus. Both required about the same amount of time to administer. Instant Messaging and online group communication may be particularly demanding and complex modes of communication for which our participants were not yet socialized. Both included online tutorials that were chosen specifically to have similar difficulty levels. Instructions that focused on how to use these tools resulted in greater Internet use than when only basic instructions about how to use the Internet were provided. .30. they may influence the extent to parents encourage their children to use computers and the Internet. Emphasizing the Internet’s information tools may be especially important in introducing the Internet to low-income groups. Poor people do not typically have home Internet access. Findings obtained for one socio-economic group may not generalize to other groups.470 JACKSON ET AL. Nor was there any indication in the behavior of our participants that they found one instructional condition more difficult to understand than the other (e. Thus. Instructions focused on either communication or information tools resulted in stronger beliefs about the importance of computers to children’s success in school than did basic instructions. Factors influencing later Internet use are seldom considered in cross-sectional studies that examine correlates. the Web). Indeed every effort should be made to encourage low-income users to use the Internet’s communication tools so that they too may benefit from the social connectedness that such use seems to foster in higher-income users. it is again unsurprising that participants made so little use of the Internet’s communication tools. friends and co-workers online. from the National Science Foundation (Information Technology Research no. accounting for the greater overall Internet use in the former condition. More research is also needed on Internet “churn” – the cessation of Internet use over time—and the implications of differential churn rates for the digital divide. but not causes of Internet use.. However. one implication of our findings is that more emphasis should be given to the Internet’s information functions rather than its communication functions in introducing the Internet to low-income users and in encouraging Internet use.g.34 more research is needed to understand how computer and Internet attitudes influence computer and Internet use by parents. the design of the two instructional sets argues against it. Why did HomeNetToo participants make so little use of the Internet’s communication tools? An answer to this question may be so obvious as to be easily overlooked. who typically have fewer resources with which to access information. Given the complex relationship between attitudes and behavior.J.e.

New York: Routledge. Rosen. 16. Jackson. (1999). Los Angeles. L.. REFERENCES 1. The digital disconnect: the widening gap between Internet savvy students and their schools.doc. K. CA: California State University. et al. 3. Hamburger. 6. A nation online: how Americans are expanding their use of the Internet. UCLA Center for Communication Policy.INSTRUCTIONAL SET AND INTERNET USE 471 15. (2003). (1992). L. U.ucla. A. S-J.. Boulder. McCreadie. UCLA Center for Communication Policy. 14..). E. UCLA Center for Communication Policy. 4. B. Mukhopadhyay.S. Personality. Katz.. Surveying the digital future: year 1. In: M. L.gov/ntiahome/ fttn99/-FTTN.. (2001). R. The racial digital divide: Motivational. 11. Jansen. Jackson. & Rice. Kraut. Available: <http://www. National Information and Telecommunications Agency. . J. Jackson.ntia. 23. Work.. et al. University of California.S. & Saracevic. UCLA Internet Report. Los Angeles. Available: <http://www..doc. G. V. Florida. (1998). Communications of the Association for Computing Machinery 47:43–47. Surveying the digital future: year 3.. (2002). (2000). (2002). Motivations for and barriers to Internet usage: results of a National Public Opinion Survey.gov/pdf/fttn00. UCLA Center for Communication Policy.edu>. (2003). Scherlis. von Eye. Journal of the American Society for Information Science 49:1101–1114.ccp.ccp. Y..E. Dominguez Hills. (2001). 18. CO: Westview Press.S. (2004). B. 26. & Weil. The functions of Internet use and their social and psychological consequences. 12. Department of Commerce.A. & Bosnjak. UCLA Internet Report. B.org>.. Internet paradox revisited. friendship and media use for information exchange in a networked organization.ucla.A. National Telecommunications and Information Agency.edu>.ucla. 17. C. and cognitive correlates of Internet use. A. (1998). Searchers. (1996). E.. Ervin. & Wellman. B. 31. Communities in Cyberspace. National Information and Telecommunications Agency. Journal of Applied Social Psychology 31:2019–2046. Falling through the Net: toward digital inclusion. Internet paradox: a social technology that reduces social involvement and psychological well-being? American Psychologist 53:1017–1031. S. UCLA Internet Report. MD. MA: MIT Press. Wellman. Biocca. Surveying the digital future: year 4. C.. An electronic group is virtually a social network. L. IT & Society 1:219–244. UCLA Internet Report. (2002). & Haythornwaite. Katz.. von Eye. University of California. Culture of the Internet. Ervin. 24. (2003). Department of Commerce. Wellman. Department of Commerce. A. The social impact of Internet use on the other side of the digital divide. & Chang. P. Boneva. T. MA: MIT Press. Implications for the digital divide of Internet use in low-income families.E. (1998). B. Understanding differences in Web usage: the role of need for cognition and the five factor model of personality. Sex Roles 44:363–380.ntia.. 30. 9. R. (1999). Gardner. 7. S. Kiesler (ed. (ed. (1998). CyberPsychology & Behavior 4:723–743. UK: Blackwell. Smith & P. the subjects they search. Measuring technophobia. & Aspden. Virtual communities as communities: net surfers don’t ride alone. Swiss Journal of Psychology 62:79–90. M.pewinternet.ccp. 179–205.The relationship between extraversion and neuroticism and the different uses of the Internet.doc. Wellman. Pew Internet and American Life Project. U. (2002).. (2000). (2002). U. Communications of the Association for Computing Machinery 39:55–65. Available: <http://www. Baterman. Tracking online life: how women use the Internet to cultivate relationships with family and friends. Internet & Intranet.. University of California. M. Available: <http://www. B. Social consequences of Internet use: access. Available: <http://search. University of California.. Cambridge. A. demographic characteristics and Internet use—findings from the HomeNetToo project. 20. Available: <http://www. et al.doc.ntia. The HomeNet field trial of residential Internet services.. Accessing and browsing information and communication.pdf>. Patterson. Rice. cognitive style. et al. 10. (1999).. Networks in the global village: life in contemporary communities.M.. 21.. M. pp. Cambridge. 13. T. Available: <http://www. (2002). Wellman. and sufficiency: a study of a large sample of Excite searches. Weiser. von Eye.). (1997). Orlando. Spink..D. et al. A.. 19. 8.edu>. (2000). B. Surveying the digital future: year 2. The Internet in everyday life. Available: <http://search. L.pewinternet..D. Gardner.. T.pdf>.A.).. Gender and the Internet: women communicating and men searching.S.L. Pew Internet and American Life Project. affective. National Telecommunications and Information Agency.. 28. A. A. Kiesler.).gov/ pdf/fttn02. L. 22. et al. 5.ccp.. Available: <http://www. Falling through the Net: defining the digital divide. In: S. Kraut. 25.ucla. Social Behavior and Personality 29:391–398.. (2001). Jackson. Mahwah. P.E. M. Presented at the WebNet98–World Conference of the WWW. (2000).. Lundmark..pdf>.A. et al. involvement and expression. NJ: Lawrence Erlbaum Associates. (2001). Falling through the Net: new data on the digital divide. Kraut. Jackson. (2001). G. Journal of Social Issues 58:49–74. B. Los Angeles. Barbatsis. et al. 27. Department of Commerce. 29. Presented at the 24th Annual Telecommunications Policy Research Conference. R.ntia.. & Ben-Artzi. A. Tuten. W. Kollock (eds. M. 2. Haythornthwaite. Oxford. F. J. Available: <http://www.gov/ntiahome/ net2>. R. D. P. R...L. U. (eds. (1997). & Gulia. Los Angeles. Computers in Human Behavior 16:441–449.edu>.org>.. J.S. Solomon’s Island. (2002). Barbatsis. K.

JACKSON ET AL. S. F. Jovanowich. L. Brace. The psychology of attitudes. H.edu . Eagly..472 32.. et al. Santiago.. et al. 34. Jackson.. FL: Harcourt. Does Internet use influence the academic performance of low-income children? Findings from the HomeNetToo project. L. von Eye. Internet attitudes and Internet use: some surprising findings from the HomeNetToo project. A. (2003).. Chile. Orlando. A. Jackson Department of Psychology Michigan State University East Lansing. F. Biocca. Address reprint requests to: Dr. (2003). A. International Journal of Human Computer Studies 59: 355–383. A... von Eye. & Chaiken. 33. Linda A. Jackson. Inc. A. Biocca. MI 48824 E-mail: Jackso67@msu. A. (1993). Presented at the First Latin American World Wide Web Conference: Empowering the Web.