Volume 12 Numbers 1 & 2, 2002

THE EFFECTS OF COMPUTER ANXIETY AND COMMUNICATION APPREHENSION

ON THE ADOPTION AND UTILIZATION OF THE INTERNET

Steven C. Rockwell
 University of South Alabama

Loy Singleton
University of Alabama

Abstract. Even while the cost of Internet access continues to drop, one-third of U.S. homes are expected to reject this technology.  This study suggests that psychological barriers such as computer anxiety and communication apprehension might offer some insight into the reasons for this rejection.  Internet usage patterns of 249 survey respondents were monitored for a one-year period.  Results suggest that those with high levels of computer anxiety are less likely to use the Internet at all while those with high levels of communication apprehension reported that they were less likely to use Internet services that involve interpersonal communication.  Further, the results suggest that experience with the Internet appears to increase the amount of time spent on-line.  The implications of these findings for policy makers are discussed.

Introduction

Psychological barriers to adoption might be suggested as a reason why some individuals would resist adopting and using the Internet even when cost or access are not issues for them.  For example, a psychological barrier that has received considerable scholarly attention is that of computer anxiety, the reported fear or discomfort experienced while using computers (e.g. Artwohl, 1989; Balance & Rogers, 1991; DeLoughry, 1993; Fariña, Arce, Sobral & Carames, 1991; Honeyman & White, 1987; Howard, Murphy & Thomas, 1986; Hudiburg, 1989; Lambert & Lenthall, 1989; Leso & Peck, 1992; Murrell & Sprinkle, 1993; Raub, 1981; Reed & Palumbo, 1988; Rosen & Maguire, 1990).  

As pointed out by Okebukola, Sumampouw, and Jegede (1992), those experiencing computer anxiety will often choose not to use a computer when given a choice.  Since a majority of Internet users currently access the Internet via a computer, those individuals experiencing computer anxiety might choose not to use a computer and, therefore, in turn, be cut off from accessing the Internet. 

However, while the Internet today is typically accessed using a personal computer, and psychological barriers to personal computer use could impede Internet use, the Internet is expected increasingly to be accessed by technology that bears little or no resemblance to a traditional personal computer: cable television set-top boxes, “smart telephones” and pocket-sized, hand-held digital devices. Consequently, conventional computer anxiety might account for only part of the psychological resistance to adoption, especially five years from now.  Resistance to Internet adoption could arguably have some additional psychological basis beyond computer anxiety.

It has been suggested that task anxiety, rather than computer anxiety per se, might account for at least some anxiety people experience when doing various types of work with computers, such as a student using a word-processing program to write a research paper (Reinsch, 1985). That is, nervousness using the computer stems primarily from the task undertaken, such as writing the paper, rather than the use of the computer itself.  This would suggest that perhaps task-related anxiety, especially anxiety associated with a communication event, might be associated with some aspects of Internet use. 

Like the personal computer, the Internet is also a task-oriented technology.  It is a communication infrastructure that uses a variety of communication hardware, software and data transmission protocols to send and receive text, graphics, audio and video. Use of the Internet is essentially the accomplishment of various communication tasks and it has been demonstrated that Internet use can, for some people, be associated with communication-task-related anxiety.  For example, communication apprehension has been demonstrated to predict the likelihood of using on-line services such as the Internet (Scott & Rockwell, 1997).

Apart from the impact both computer anxiety and communication apprehension have on the amount of time spent using the Internet, this study also examines how experience with various computer and communication technologies might mitigate any psychological barriers and lead to increased levels of Internet use.

This study is based on field data collected by tracking time spent on the Internet by individuals categorized as to their self-reported levels of computer anxiety and communication apprehension.   The study period was one year in duration.

Literature Review

Computer Anxiety

Computer anxiety is a term widely used to describe the fear or apprehension experienced by certain individuals when using or considering using a computer.   Much of the research in the area of computer anxiety examines the relationship between computer anxiety and experience level with computers.  Many of these studies suggest that increased levels of computing experience lead to lower levels of computer anxiety (e.g. Fariña, Arce, Sobral, & Carames, 1991; Lambert, Lewis, & Lenthall, 1989; Liu, Reed, & Phillips, 1990; Loyd & Gressard, 1984; Raub, 1981).  Scott and Rockwell (1997) suggest that computer anxiety is a significant predictor of the future use of a wide range of interactive communication technologies, including the Internet.  

However, some studies have failed to support these findings, reporting non-existent or weak relationships between computer anxiety and experience levels (Sears; Marcoulides, 1988).  Others, such as Gos (1996), offer support for the task anxiety model first posited by Reinsch (1985).  This model suggests that it is not a general anxiety toward computers, but, anxiety toward performing specific tasks on a computer, e.g., programming or interpersonal communication. 

Communication Apprehension

Communication apprehension is an individual trait characterized by fear or anxiety when contemplating or engaging in some type of communication with other people (McCroskey, 1978). This research has dealt primarily with oral communication apprehension (McCroskey, 1984).  However, it typically suggests that individuals exhibiting high degrees of apprehension about communication are less likely to engage in any type of communication.  An Internet user can engage in a variety of activities such as viewing a web page, sending or receiving e-mail, engaging in real-time text- or voice-based chat, transferring information from one location to another, participating in a videoconference and viewing or listening to audio-visual materials.   Such tasks involve aspects of interpersonal communication and thus could be associated with some type of communication apprehension.   Scott and Rockwell (1997) found a significant negative correlation between communication apprehension and the likelihood of using any kind of online service.  This would suggest that communication apprehension might affect the use of forms of interpersonal communication beyond face-to-face, oral communication; in particular, the Internet and especially the more immediate forms of Internet communication, such as Internet relay chat (IRC) and multi-user domains (MUDs). 

If communication apprehension can be associated with using the interpersonal communication capabilities of the Internet, then individuals exhibiting high degrees of communication apprehension should be more likely to avoid these activities. Moreover, since it has been suggested that the primary use of the Internet in the home is for interpersonal communication (Kraut, Mukopadhyay, Szcypula, Kiesler, & Scherlis, 1998; Survey, 1999), the communication anxious should spend less time, overall, online.

Experience

Additionally, Scott and Rockwell (1997) illustrated that experience can serve as a predictor “above and beyond” other anxiety/apprehension measures.  Therefore, experience with a technology can lessen the impact of any access issues stemming from a sense of anxiety with the technology.  In the present investigation, support for the experience factor would provide evidence that remediation strategies for the digital divide need to provide more than just access alone.  Training or orientation sessions, by providing experience with computers and the Internet, would help ease any access limitations based on psychological barriers.  

Hypotheses

Based on the above discussion, it would appear that those exhibiting higher levels of computer anxiety would be less likely to feel comfortable using a computer to gain access to the Internet. Since the Internet usage tracked in this study was predicated on the use of a computer, no alternate means of access was available, the following hypothesis was formulated.

H1: Increased computer anxiety will result in decreased amounts of time spent on the Internet.

Moreover, since it is argued that Internet anxiety levels might be a function of communication task anxiety, and once on-line this type of anxiety might better explain differences in the amount of time spent using the Internet, the following hypotheses were formulated.

H2a: Increased communication apprehension will result in decreased amounts of time spent on the Internet.

H2b: Increased communication apprehension will result in decreased reported likelihood to use various services offered on the Internet, especially those involving interpersonal communication.

Additionally, based on findings by Scott and Rockwell (1997), regarding previous experience with a technology as a strong predictor of its future use, offering explanatory power above computer anxiety level alone, the following hypothesis was formulated.

H3: Individuals with prior Internet experience will use the Internet more than those without prior experience, regardless of anxiety level.

Method

Free Internet access service sponsored by the local community was made available to all citizens of a small southeastern city of 6800.  They had a choice of either signing up for Internet access through public terminals in the local library or for a free dial-up account via a modem from their home.  Completion of an account request form was required before an account could be established at either location.  Each registrant was also asked to complete a survey that was attached to the account request form.  Instructions on the survey instrument informed the registrants about the confidential and voluntary nature of the survey.  The forms were made available to registrants at the local library and at City Hall.

All registrants, a total of 249, completed the survey’s general demographic questions, technology apprehension and communication apprehension measures, and questions regarding their level of experience with the Internet and other information technologies and their expectations regarding their Internet use.   Each registrant's actual time spent on line was then tracked for the one-year duration of the study.  Library users were tracked using sign-up logs maintained by library staff; home users through authentication and tracking software installed on the project's remote access server.

Variables

Computer/Technology anxiety was measured with a modified version of Raub's (1981) 10-item measure which has been previously shown to be acceptably reliable with alphas ranging from .81 to .86 (Howard & Smith, 1986; Ray & Minch, 1990; Fariña et al., 1991).  The questions were left intact, but the response options were modified to a five-point, Likert-type scale with response options ranging from strongly disagree to strongly agree.  Communication apprehension was measured with a version of McCroskey's (1970) 10-item personal report of communication apprehension (PRCA), using the same modifications as the anxiety measures.  Overall levels of communication apprehension and computer/technology apprehension were derived by summing responses for each scale and dividing by the total number of questions in each scale. 

Level of experience with the Internet was assessed via a single item where respondents indicated whether they had previously used the Internet or had an Internet account.  To assess respondents’ experience with computers, they were asked whether they had any previous computer experience and whether they owned a home computer.

Respondents’ likelihood to use various communication-related Internet services was assessed using a series of three items asking how likely they were to use e-mail, join an on-line discussion group, or to take part in an on-line chat session.  Other questions dealing with both traditional interpersonal-related communication situations and to technology related, non-communication situations were included in the questionnaire for validity purposes.  These additional included items were: How likely are you, to give a speech to a group of people, use a computer to word process a document, view a program on a VCR, and record a program on a VCR.  Respondents answered via a five-point, Likert-type scale with response options ranging from very unlikely to very likely.   Respondents’ amount of time spent online was tracked for the one-year duration of the study. Additional demographic data were also collected.

Analysis

Hypotheses 1, 2a and 2b were tested via correlation analysis. For hypothesis 1, correlation coefficients were calculated between participants’ computer anxiety scores and their average time spent using the Internet.  For hypothesis 2a, participants’ communication apprehension scores were correlated with their average time spent using on-line.  After the data were collected, it was discovered that of the 249 individuals who completed a questionnaire, only 116 of these actually used the Internet during the one-year duration of the study.  It was then decided to examine the relationship posited by hypotheses 1 and 2a further by examining the relationship between computer anxiety and communication apprehension between Internet users and non-users.  Analysis of variance was used to test these relationships.

For hypothesis 2b, participants’ communication apprehension scores were correlated with the four questions designed to measure likelihood to use various Internet services.  Hypothesis 3 was tested using a two-way analysis of variance with average Internet use serving as the dependent measure and previous Internet experience (none, used it before but not had an Internet account, have or had an Internet account) and participants’ level of computer anxiety (high = top half computer anxiety scores, low = bottom half computer anxiety scores) serving as the independent measures.  Significance for all procedures was set at an alpha level of p<05.

Results

A total of 249 participants requested Internet service and filled out the questionnaire during the study period.  Of these, 116 participants successfully used the Internet while the remaining 133 never attempted to gain access. Participants who used the Internet did so for an average of 3.09 hours per month. One-half (50%) of participants were female with 45% of this group actually using the Internet. Fifty-five percent of the male participants used the Internet during the study period.  The average age of the participants was 42.96 years.  Ninety-one percent of the participants reported having at least some computer experience.  Regarding previous Internet experience, 63% of the participants reported having no previous experience with the Internet, 21% reported having prior experience, but never having their own Internet account, and 16% reported having had an Internet account previous to taking part in the study.

Applying the usual rule of thumb of about 2.5 persons per average household, and assuming that the registrants all came from separate households, then roughly speaking, the registrants represented about 10 percent of the city’s households.  As might be expected from national adoption patterns, the registrants had considerably higher incomes than the city population as a whole.   Registrants had a median income 69% higher than the city median income. Interestingly, this income disparity between registrants and the city population appeared here regardless of the service being offered at no charge.  A free service would presumably help to overcome income as a barrier to adoption and produce a group of registrants with a median income more nearly resembling the city’s median income.  However, even with a free service available, registrants tended to have substantially higher incomes than the city in general.

Registrants were predominantly college educated, while most of the city’s residents typically attain high school diplomas, at most.  Ethnically, compared to the city population, whites were over represented among the registrants and blacks under represented.  Although ethnicity differed, there was not a statistically significant difference in the gender makeup of registrants compared to the city population.  Overall, the income, education and ethnic differences between the registrants and the city population echo familiar patterns documented many times in other studies and in the popular press regarding use of the Internet (Belluck, 1999; Hafner, 2000; Stepanek, M., 1999). 

Participants exhibited a wide range of computer anxiety and communication apprehension scores.  Computer anxiety scores ranged from 1.0 to 4.4 with an average score of 1.8 while communication apprehension ranged from 1.0 to 4.7 with an average score of 2.6.  Scale reliabilities were acceptable for both computer anxiety (alpha = .83) and communication apprehension (alpha = .87).

Hypothesis Testing

Hypothesis 1 asserted that increased levels of computer anxiety would be associated with decreased amounts of time spent on the Internet.  Among Internet users there was a slight negative correlation (r = -.074) between these variables, but this correlation did not prove significant (p = .443).  Computer anxiety, then, did not appear related to the amount of time the Internet was used by those participants who did use it. 

Using analysis of variance, computer anxiety scores of these Internet users were compared to the computer anxiety scores of the 133 participants who had signed up for service but never accessed the Internet during the study period.  Internet users exhibited a significantly lower (F (1,229) = 5.73, p = .018) computer anxiety score (M = 1.73) compared to non-users (1.95).  Although Internet users were not significantly differentiated on the amount of time they spent on-line based on their computer anxiety scores, participants who requested service but did not use the Internet during the study period exhibited significantly higher levels of computer anxiety, compared to their Internet-using counterparts.  Thus, the first hypothesis was partially supported.

Hypothesis 2a maintained that increased levels of communication apprehension would be associated with decreased amounts of time spent on the Internet.  The correlation between these variables (r = -.036) among the participants who actually used the Internet did not prove to be significant (p = .582).

An analysis of variance was also performed comparing these Internet users and the non-users to compare their levels of communication apprehension.  The difference between Internet users (M = 25.71) and non-users (M=26.01) as to level of communication apprehension was not significant (F (1,231) = .078, p = .780).  Hypothesis 2a, therefore, was not supported. 

Hypothesis 2b posited that increased levels of communication apprehension would be negatively correlated with self-reported likelihood to use Internet services such as e-mail and newsgroups that involve interpersonal forms of communication.  As expected, communication apprehension was negatively correlated to these variables (see Table 1).

Table 1.Correlations of Communication Apprehension with Likelihood to Use Internet and Non-Internet Technologies

Technology                                                     Communication Apprehension

_____________________________________________________________

E-mail                                                                                    -.243**

On-line Chat                                                                           -.260**

On-line Discussion Group                                                       -.154*

Word Processor                                                                     -.168*

View a Program on a VCR                                                     -.120

Record a Program on a VCR                                                  -.074

Give a Speech                                                                        -.682**

_____________________________________________________

The strongest negative correlation was between communication anxiety and giving a speech in public (r= -.682), a validation of the intended use of the communication apprehension scale to measure the anxiety felt in oral communication situations.  However, significant negative correlations also emerged for the included Internet related communication situations: likelihood to use e-mail (r = -.243),  to take part in an on-line chat session (r = -.260), and to join an on-line discussion group  (r = -.164).  No other significant correlations emerged in the analysis.  Thus, hypothesis 2b was supported.

Hypothesis 3 stated that individuals with prior Internet experience would use the Internet more than those without prior experience, regardless of anxiety level.  To explore this hypothesis, participants were sorted into one of three groups based on their reported level of experience with the Internet: no prior Internet experience, previous Internet experience without an Internet account, and Internet experience using an account.   An analysis of variance was then performed comparing these groups on the average amount of time they spent using the Internet.  This comparison was significant (F (2,244) = 3.056, p = .049).  In this analysis, only the participants who had an Internet account previous to the study used the Internet significantly more (M = 2.49), on average, than participants with no prior Internet experience (M = 1.15) (see Table 2).  Participants with prior experience, but no account, fell in between the other groups (M = 1.63) and failed to differentiate significantly from either of these groups.  The interaction between computer anxiety level (high, low) and previous Internet experience was not significant (F (2,244) = .781, p = .459) for this analysis.  Thus, hypothesis 3 was supported. 

Table 2. Average Time Spent using the Internet Based on Internet Experience

Experience Level

                        No                       Experience/                 Previous

                   Experience         No Internet Account    Internet Account

                          1.46_a                     1.63_ab                        2.49_b

Discussion

The results in this study would seem to suggest that the ‘digital divide’—at least in this study’s setting—involves two divides or barriers.  The first is the divide between the city population as a whole and the portion that signed up for free Internet service. Few citizens took up the free Internet access offer to begin with.  Those who did were not representative of most of the population.  They were certainly not representative of lower-income, lower educated citizens. This divide is characterized by a difference in income, ethnicity and education between those who signed up and those who didn’t.  Registrants as a group were better educated and had substantially higher income than their city population.  There is other research that suggests that the disproportionately high level of whites to African Americans among the city’s registrants may be an artifact of income (Novak and Hoffman, 1998).  This income and education disparity between registrants and the city population is the same divide that has caught the attention of policy makers nationally and sparked substantial debate. 

The second divide suggested by this study is the divide within the registrant group, between those who actually took the final step of using the service and those who did not.   This divide represents a much different type of barrier than the first divide.  It is not directly predictable by the demographic characteristics that correlate with the first divide.  Rather, this second divide is based in a psychological condition, the degree of fear of, or at least discomfort with, computer technology.  This difference in comfort levels was the only variable that was related to actual use of the Internet to a statistically significant degree.  Each of these divides represents a significant problem for programs and policy-makers intent on narrowing the digital divide. 

The second divide is more subtle, because in a sense, it is self-imposed.  The data suggest that, even among people who are willing to sign up for free Internet service, a substantial number will fail to actually take the next step and put it to use.  These Individuals, typified by a higher degree of computer anxiety, were significantly less likely to use the Internet at all during the study period.    Thus, it appears that computer anxiety can be a rather strong inhibitor of adoption and needs to be carefully considered when assisting individuals exhibiting high levels of this anxiety with their initial ventures into cyberspace.

Further, in this study, previous experience with computers seemed to have little impact on lessening computer anxiety.  Most of the study’s participants (91%) reported having prior computer experience, yet there were clear distinctions between the computer anxiety scores of Internet users and non-users.  This would appear to lend support to previous studies (e.g., Rosen, Sears, & Weil, 1987; Marcoulides, 1988) that also failed to find a relationship between computer anxiety and experience. 

However, previous experience with the Internet did appear to affect the amount of time participants spent using the Internet.  Participants who had previously had an account through an Internet service provider spent more time on-line than those without previous Internet experience did.  The participants’ level of computer anxiety did not seem to affect this relationship.  Since it is logical to assume that participants who previously had Internet accounts had more Internet experience than other participants, it would appear that by increasing exposure to, and thus experience with, the Internet, the barrier to entry created by elevated levels of computer anxiety might be overcome.

Communication anxiety did not offer any explanatory power about differences in the amount of time spent on-line or even whether a participant used the Internet at all during the study period.  However, it did seem to offer some insight into the types of services participants exhibiting high levels of this trait were likely to avoid.  These individuals reported that they were less likely than their low communication apprehension counterparts to use Internet services that involve interpersonal communication.  However, since this study was not able to track what types of Internet services participants utilized, it is impossible to tell if high communication apprehensives avoided these services and spent their time using other services that did not appear to be high in interpersonal communication or simply overcame their reticence once on-line.  Regardless of the explanation of these findings, this study does lend support to the notion that McCroskey’s (1970) communication anxiety scale can be extended to measure this trait in interpersonal communication situations beyond oral, face-to-face communication.           

Conclusions

This study strongly suggests that technology apprehension is a real component of the complex phenomenon of Internet adoption and use.  It might also help to explain why so many residents failed to register in the first place.  It is a worthwhile subject of study in determining why people who have access to the Internet fail to take it up.  Clearly, as access becomes more widely available under various foundation and government supported programs, questions will arise regarding the dynamics of adoption that go beyond the cost of technology.  The presence of the computer technology apprehension phenomenon and its possible role in rejection of the Internet when access is available, suggests that public policy designed to achieve high levels of Internet access and use, based on a technology-push model (“build it and they will come”) will only partially succeed in the goal of bridging the digital divide. 

Because of the necessary constraints on data collection in the design of this project, a precise answer as to how best and most efficiently to overcome the computer technology apprehension barrier must await further research.  Further, because the study only involved a population of volunteer registrants from a small town in the Southeastern U.S., the generalizability of the results to other populations is limited.  If possible, future studies might want to use random sampling techniques from user populations with divergent geographic and socioeconomic backgrounds.  However, the results suggest, at least for the population tested, that orientation and instruction that de-mystifies the Internet and makes it a friendlier environment for first-time users could help ensure that they actually utilize the Internet when it is accessible.  Only then can they begin to derive the educational and economic benefits being touted by Internet access proponents.

References

Artwohl, M. J. (1989).  What research says about keyboarding skills and computer anxiety.  Unpublished Research Report.  ED 312444.

Balance, C. T., & Rogers, S. U. (1991).  Psychology of computer use:  XXIV.  Computer-related stress among technical college students.  Psychological Reports, 69, 539-542.

Belluck, P. (1999, September 22). What Price Will Be Paid By Those Not On the Net. New York Times, p.12.

DeLoughry, T. J. (1993, 28 April).  Researchers say 'technophobia' may afflict millions of students.  Chronicle of Higher Education, pp. A25-A26.

Dukes, L. R., Discenza, R., & Couger, D. (1989).  Convergent validity of four computer anxiety scales.  Educational and Psychological Measurement, 49, 195-203.

Fariña, F., Arce, R., Sobral, J., & Carames, R. (1991).  Predictors of anxiety toward computers.  Computers in Human Behavior, 7, 263-267.

Gos, M. W.  (1996).  Computer anxiety and computer experience: A new look at an old relationship.  The Clearing House, 69 (5), 271-277.

Hafner, K. (2000, March 5). A credibility gap in the digital divide.  New York Times, p.4

Honeyman, D. S., & White, W. J. (1987).  Computer anxiety in educators learning to use the computer: A preliminary report.  Journal of Research on Computing in Education, 20, 129-138.

Howard, G. S., Murphy, C. M., & Thomas, G. E. (1986).  Computer anxiety considerations for design of introductory computer courses.  In Proceedings of the 1986 Annual Meeting of the Decision Sciences Institute (pp. 630-632).  Atlanta, GA: Decision Sciences Institute.

Hudiburg, R. A. (1989).  Psychology of computer use: VII. Measuring technostress: Computer-related stress.  Psychological  Reports, 64, 767-772.

Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T., & Scherlis, W. (1998). Internet paradox: A social technology that reduces social involvement and psychological well-being?  American Psychologist, 53, 1017-1031.

Lambert, M. E., & Lenthall, G. (1989).  Effects of psychology courseware use on computer anxiety in students.  Computers in Human Behavior, 5, 207-214.

Leso, T., & Peck, K. L. (1992).  Computer anxiety and different types of computer courses.  Journal of Educational Computing Research, 8, 469-476.

Liu, M., Reed, W. M., & Phillips, P. D. (1990, April).  Teacher education students and computers: Gender, major, use, occurrence, and anxiety.  Paper presented at the Annual Meeting of the American Educational Research Association, Boston, MA.

Marcoulides, G. A. (1988).  The relationship between computer anxiety and computer achievement.  (2), 7-19

McCroskey, J. C. (1970).  Measures of communication-bound Journal of Educational Computing Research, 4 anxiety.  Speech Monographs, 37, 269-273.

McCroskey, J. C. (1978).  Validity of the PRCA as an index of oral communication apprehension.  Communication Monographs, 45, 194-203.

McCroskey, J. C. (1984).  The communication apprehension perspective.  In J. A. Daly & J. C. McCroskey (Eds.), Avoiding communication:  Shyness, reticence, and communication apprehension (pp. 13-38).  Beverly Hills, CA: Sage.

Murrell, A. J., & Sprinkle, J. (1993).  The impact of negative attitudes toward computers on employees' satisfaction and commitment within a small company.  Computers in Human Behavior, 9, 57-63.

Novak, T. & Hoffman, D. (1998).  Bridging the digital divide: the impact of race on computer access and internet use.  Vanderbilt University, Project 2000.  Available at: http://www2000.ogsm.Vanderbilt.edu.

Raub, A. C. (1981).  Correlates of computer anxiety in college students.  Unpublished doctoral dissertation, University of Pennsylvania.

Ray, N. M., & Minch, R. (1990).  Computer anxiety and alienation: Toward a definitive and parsimonious measure.  Human Factors, 32, 477-491.

Reed, W. M., & Palumbo, D. B. (1988).  The effect of the BASIC programming language on problem-solving skills and computer anxiety.  Computers in the Schools, 4, 91-101.

Reinsch, N. L., Jr. (1985). Technology aversion (with implications for education and training).  Office Systems Research Journal, 4, 9-20.

Rosen, L, & Maguire, P. (1990).  Myths and realities of computerphobia: A meta-analysis.  Anxiety Research, 3, 175-191.

Rosen, L. D., Sears, D. C., & Weil, M. M. (1993). Treating technophobia: A longitudinal evaluation of the computerphobia reduction program. Computers in Human Behavior, 9, 27-50.

Scott, C., & Rockwell, S. C.  (1997). The effect of communication, writing, and technology apprehension on likelihood to use new communication technologies.  Communications Educator, 46, 44-62.

Stepanek, M.  (1999, October 4). A Small Town Reveals America’s Digital Divide. Business Week, p. 188.

Survey: Electronic Mail Leading Reason To Go Online. (1999, September 30). San Jose Mercury News, p. 1.  Retrieved October 4, 1999 from the World Wide Web: http://www.sjmercury.com/cgi-bin/edtools/printpage.p1.