The growing importance of computers in society has been noted by a number of researchers (Shapiro, 1995, Walther & Burgoon, 1992). As Nickerson (1982) noted "The implications that information technology has on our lives are beyond doubt very great". Carpenter (1983) stated: "When you're building a computer system, you are building a social system."

For, "in the next millennium, we will find that we are talking as much or more with machines than we are with humans" (Negroponte, 1995, p. 145). As a result of such dramatic transitions, the very nature of what communication is and what it is not may be significantly altered in the future. Levy (1995) reported that:

The revolution has only just begun, but already it's starting to overwhelm us. It's outstripping our capacity to cope, antiquating our laws, transforming our mores, reshuffling our economy, reordering our priorities, redefining our workplaces, putting our constitution to the fire, shifting our concept of reality and making us sit for long periods in front of computer screens while CD-ROM drives grind out another video clip.(p. 27).

Years ago, computers used up large areas of space. Today, desktop computers are capable of handling large quantities of information, and of providing the user with easy to use, interactive technology. Interactivity describes possibilities for the user to make choices which influence the program units. When interactive technology incorporates different channels, it is called Multimedia.

The term Multimedia describes the concurrent use of different sets of electronically stored data, such as image, video, animation, and sound, in one coherent context which is mediated by a computer. Multimedia applications are currently used for entertainment (e.g., computer games), information systems (e.g., on airports), education, and art.

They allow the user to determine her/his own learning speed, define special areas of interest, and have contents displayed in a way he/she can work with most easily. Multimedia systems allow the user to learn from experiencing, watching, and listening rather than from reading alone.

However, due to novelty effects and information overload, users might feel uncomfortable and experience higher stress levels when they are working with a computer. Promoters of these systems, on the other hand, argue that Multimedia applications facilitate easier learning tailored to each student's specific needs and that therefore stress should be decreased.

This study indicates whether college students experience higher stress levels when they are working with computer applications as opposed to being instructed by "traditional" methods.
 

History of Interactive Systems in Education

Interactive learning strategies have been known for 2,500 years, when Socrates taught students solely by asking them questions to make them find the message in their own answers. Later, the introduction of multiple choice strategies offered a low quality of interactivity. The introduction of personal computers in the late 1970s facilitated the gradual realization of interactive personal systems. Throughout the 1980s, a growing movement toward interactive technology could be viewed. New software applications and advanced hardware made it possible to utilize pieces of electronically stored information from different channels for programs that let the user determine what he/she wanted to see or hear next.

For the first time, the user could control the pace of the flow of information in the program and pick those pieces of information which were most interesting for her/him.

Currently, steps are being taken to include new communication technologies in our education system: The recent passage of the "High Performance Computing and Communications Initiative" by the U.S. Congress will facilitate the development of the "National Information Infrastructure" (NII), a Gigabit-Network that will make it possible to transport and share data quickly among U.S. education institutions such as schools and Universities.(Erickson, 1993, cited in Wexelblat).

Learning Environments in Education

Technology often contributes to the complexity faced by humans in their daily lives. However, it can also provide means to make the difficult more manageable (Weir & Alty, 1991).

Two strategies are extremely important to make difficult or abstract information tangible for students (Erickson, 1993, as cited in Wexelblat): Selective emphasis, on the one hand, highlights certain features of the data and suppresses others in order to emphasize important parts, while contextualization, on the other hand, provides a visual framework within which the data are displayed.

Arguments for the use of Computers for Education

Learning theorists, from Benjamin Bloom to more recent ones, have argued that there is evidence that every student can learn in the right learning environment. In his article "Learning in the twenty first Century: Interactive Multimedia Technology", Alfred Bork suggests that Multimedia systems have an immense potential for making massive improvements in learning.

"These new educational systems can be vastly superior for most of our students to the systems they use today" (Bork, 1992, as cited in Giardina). Bork's studies show that students who are involved in an interactive learning process are more likely to stay at difficult tasks. Learning Materials using many media have been shown to be more successful than materials using only one medium, and individually paced learning has been shown to be more successful than learning at an predetermined speed (Bork, 1992, as cited in Giardina). Several studies showed that students who received information in the context of problem solving were much more likely to use it as a basis for creating new sets of plans.

Findings of other studies (Clark & Craig, 1992, as cited in Giardina) suggest that measured learning gains can be attributed to interactivity. As Nix and Spiro (1992) noted: "Computers in education...can have a significant impact".

Studies about Computer Assisted Instruction

Studies (Bricken & Byrne, 1993, as cited in Wexelblat) demonstrated that students are basically open to and not afraid of using new communication technologies for education. A study about contextual enrichment by videodisc conducted by Gildea, Miller and Wurtenberg (1990) indicated that videodisc systems indeed proved successful for instruction. A study conducted by Chomsky (1990) showed a "Book on Disc", with which students could interact in various ways, to be accepted well. The interactive, easy to use computer program allowed flexible user control. Carol Chomsky suggested similar systems to be used for teaching.

Hanson's and Padden's (1990) touch-screen software and videodisc "Hands On" for deaf children was found to be successful in teaching children to translate from sign-language to writing by telling them children's stories both in the American Sign Language and in English writing.

Guthrie and Dreher (1990) tried to incorporate cognitive aspects in the design of new computer programs used for education. Their study proved their "Cognitive Model" to be successful in enhancing learning performance for various tasks.

Standards for Instructional Software

Previously conducted studies indicate that the design of a computer application to a large portion accounts for its success. Learning environments should consider individual learning strategies in order to encourage the exchange of relevant information between the interactive environment and the user. In addition, computer systems should be easy to operate and should offer the user a pleasant experience.

"The Multimedia learning environment should not represent a passive object that only contains or assembles information, but should become the place where the learner reflects and where he or she can play with and access information and try to interpret it, manipulate it and build new knowledge" (Giardina, 1992).

According to Giardina (1992), designers must ensure that the learner feels satisfied at each stage and experiences a relationship based on interaction with the environment. According to Don Nix, the goal of computer-assisted education is to enable the user to be creative and self-expressing. Active lerner participation should be encouraged. Developers of Multimedia systems today agree that these systems should help students, instead of merely memorizing facts, to reason, solve problems, make decisions, evaluate, and think effectively. A new technological environment of this kind, however, demands new organization of visual, aural and textual information (Giardina, 1992) to ensure its educational effectiveness.

Problems associated with the Design and Implementation of Instructional Technology

The design of computerized learning environments is still more intuitive than scientific. "The result of largely ignoring cognitive issues is demonstrated by a design approach where all the information is presented with equal availability, and little or no effort is put into integrating or prioritizing information".(Grant & Mayes, 1991, as cited in Alty & Weir).

In addition, Nix (1990) questions the level of satisfaction users derive from systems used for education." Many of the ways computers are used in education pose a threat to the dignity of the users. This is a problem". What Nix refers to is the problem of many systems for Computer-Assisted Instruction to merely use unambiguously defined "right" and "wrong" categories. This often results in the learner's experience that he or she as a person is irrelevant to the process of learning. Referring to these concerns, Nickerson (1988) stated "This is very much an undeveloped area......but much more needs to be done in this area".

Computers also influence social settings. "The PC has created a a direct link between science and the individual affecting the language and the lifestyle of the average American". (Chesebro and Bonsall, 1989). These changes could diminish satisfaction in a learning situation. Gratz and Salem (1984) argued that the time a user spends with a computer is a nonsocial experience that promotes an undue emphasis upon self reflective rather than social and cultural conceptions and extensions of the self.

The importance of empathic understanding and social learning, however, has been emphasized by many researchers. The use of computers for instruction is assumed to decrease empathy and social learning in the classroom, because each student works by herself/himself. Stress levels may be increased as a result of the use of computers in the classroom.

Use of Computers and Stress

From the ubiquitous telephone to the gobs of data floating around on the Internet, faxes, electronic mail, and 500 channel television, information washes over us constantly. There's far more of it than we can possibly cope with. (Hafner, 1995, p.76)

Situations involving information overload have been proven to cause stress and impair performance (Keizer, 1991). Information overload might, if not properly handled, lead to increased levels of physiological arousal on the part of the user. Large increases in physiological arousal have been proven to be experienced as unpleasant and as causing fear and disorientation, and stress or even panic reactions are a common response to overarousal (Capella and Greene, 1982, Mehrabian, 1976).

Stress as a function of computer usage by novices has been described as a serious problem in a learning environment (Keizer, 1991). Stress has the potential to impair accurate decoding of others' communication (Keeley-Dyreson, Burgoon, Bailey, 1991), which might be even more problematic in a situation, in which human-human communication is substituted by human-computer communication.

H1: College Students will experience higher levels of stress when they are working with computers than in traditional instructional settings.

In addition, novel communication stimuli or situations have been identified to be factors which can cause changes in arousal.

Therefore, it is assumed that students who work with computers mre regularly experience lower levels of stress than relatively unfrequent users.

H2: Students who use computers frequently experience lower levels of stress when they are working with a computer than unfrequent users.

Since males have been shown to have a better technical understanding than females, it is also hypothesized that males experience lower levels of stress than females when working with a computer.

H3: Males experience lower levels of stress than females when working with a computer.

It is also possible that good students find it easier to adapt to a new learning environment than students with lower learning capacities, who may experience higher levels of information overload and physiological arousal.

H4: The higher a student ranks in class, the lower will the level of stress be which he/she experiences in computer assisted instruction.

Whether the applications available today indeed match the requirements stated, or if they cause stress in the students and thereby impair learning has not been tested so far (Nix, 1990). No research regarding children's levels of stress while working with these applications has been found.
 

Methodology

Subjects

This study constitutes a follow-up to a study conducted in 1994, comparing stress levels of 12 year-old Austrian secondary school students. For this study, Austrian college students were tested, in order to gain results from a different age group. In response to findings from the earlier study, the methodology was modified.

Because it was not possible to draw a random sample for the purpose of this study, a convenience sample was used, consisting of the students in an undergraduate course in Communication at the University of Vienna, who were already used to be instructed with such systems.

Of the 25 subjects, 12 were male and 13 were female. All students had at least some computer experience. About 50% of the students were classified by the instructor as displaying "average scholastic achievement", with about 25% each falling above and below this range.

Even though there has always been concern about findings drawn from students, these concerns are somewhat mitigated by two factors: First, most previous CMC research has also used students; therefore, the replication value is increased, and second, CMC in the "real world" is most frequently used in ways which are very similar to student CMC use (Walther, 1994). Because convenience sampling was used, and the number of respondents was small, this study only has exploratory character, but cannot provide results which are representative and generalizable to the whole population. This is a limitation to the study.

Procedure

The data were collected during a regular class meeting at the University of Vienna. Students were observed while they were retrieving information and exploring certain topics on the World Wide Web, a hypertext approach to providing information on the internet. Two coders coded the students' facial expressions on forms. Thereafter, the same students' facial expressions were coded in a traditional classroom situation in the same course.

After the session, the students were asked to complete a short questionnaire, and subsequently, the instructor was asked to rate each student's performance in class. Upon completion of the questionnaire, students were reimbursed for their participation with candy bars and soft drinks.

Measure

In most of the studies that have been reviewed, stress has been measured by either measuring physiological symptoms such as galvanic skin reflex or heart rate, or through self evaluations (measurement of perceived stress). The most common example for the latter is the "Computer Technology Hassles Scale" developed by Hudiburg (1991). This scale, however, measures the nature and frequency of "hassles" and identifies the most common problems people experience when working with a computer, and does not measure the level or valence of the experienced stress.

The natural setting, in which this study took place, and the fact that 25 subjects were be present at the same time, did not allow for the more complicated and obtrusive method of measuring physiological signs for stress, or the utilization of the FACS, a scale which measures nonverbal stress indicators by analyzing close-up videos of the subjects' faces. Therefore, nonverbal stress indicators were analyzed.

Because in a previous study conducted by the author, out of the nonverbal cues used to assess stress levels (body posture, facial expression, attentiveness, and self adaptors), only facial expression had produced significant results, the other items were excluded from this study, and only facial expression was coded in order to evaluate stress levels. It was assumed that stress levels can be measured accurately by observing the students' facial expressions.

In this study, stress is defined as negatively valenced physiological arousal of high intensity. It has been proven that, to measure experiencd arousal, nonverbal behaviors can serve as useful overt indices (Burgoon, Kelley, Newton, Keeley-Dyreson, 1989). Nonverbal manifestations of emotional arousal depend on whether the emotion is pleasant or unpleasant, intense or nonintense. As arousal increases from moderate to high levels, it has been argued, performance is impaired. In order to be able to measure both intensity and valence of the students' arousal, a continuum scale was employed, categorizing the facial expression, one of the nonverbal signs identified by Burgoon et al (1989) to indicate arousal.

Each student's performance in each setting was coded according to this scale, and values 1-5 were assigned to the variable. Coordination of movements and coordination of speech have been left out, since in the classroom setting, students were not expected to either talk or move enough to make these behaviors conclusive for measuring their stress levels.

Instrumentation

The instrumentation for this study consisted of:

A) A questionnaire which asked the students how hard they thought it was to work with the computer and about their frequency of computer use. On the questionnaires, the gender of the student and his/her performance in class (provided by teacher and collapsed into categories good - average - poor) were also noted.

B) A scale for the coding of facial expressions for both the computer- and the teacher sessions.

Statistical Analysis

The data were analyzed using the SPSS computer program. It was assumed that stress levels would be different when different ways of instruction are employed. Ratings of facial expression were used to determine stress levels. The ratings provided by coder 1 and coder 2 were highly correlated and were collapsed into a new "stress" variable for each the computer session (Alpha = .8) and the traditional session (Alpha = .918).

Average stress levels for both sessions were compared using one-tailed t-tests of paired samples. Further, means of stress levels for different ranges in class, gender, and frequency of computer use were determined in order to find out whether intelligence level, interest in computers or sex play a role in the stress level caused by the comuter application.

Results

The result for the t-test was highly significant (p = .001), indicating that there was a significant difference in the means for stress levels in the computer situation and the regular class session, and further indicating that the effects were in the hypothesized direction (t = 5.16). Therefore, hypothesis one was confirmed.

Hypothesis two, stating that frequent computer users would experience lower levels of stress, was also confirmed. Stress was significantly lower for daily computer users than for the other groups, and slightly lower for people who used computers 3-4 times a week as compared to less frequent users.

Hypothesis three was confirmed, too. Males experienced lower levels of stress than females on average, but the effect was slight, and variances were comparatively high, especially for female respondents.

Similarly, range in class influenced stress levels in the proposed direction, and hypothesis four was confirmed, but, again, variances were high.

Conclusions

Even though all four hypotheses were confirmed, and the significance of the t-test results was high, the results of this study should be viewed with caution, because of the small number of subjects and the high variance for hypotheses 2-4. However, results clearly indicate, at a very high significance level, that stress levels are indeed higher when computers are used for instruction. This can be explained by findings from stress literature, as cited above.

Regarding the frequency of computer use, the study indicates that frequent computer use indeed lowers stress levels for the use of computers in class. This is not surprising, since novelty effects and initial fears are among the strongest predictors for stress related to computer use. The fact that males were found to find working with computers somewhat less stress producing than females confirms that males' technical understanding seems to diminish initial fears and problems.

For range in class, the findings revealed an interesting fact: It seems that both good students and average students experience low levels of stress. However, students with poor scholastic achievements seem to experience working with computers as very stressful. This could indicate that less gifted students might need special attention when they are working with computers. In addition, the harder the students found the application to be, the more stress was measured, indicating that self evaluations can provide accurate outcomes in this context.

Suggestions for Future Research

With only 25 subjects tested, this study cannot provide more than identify tendencies. Therefore, it is suggested that larger samples are tested in order to acquire generalizable results.

Another deficiency of this study, which might question the accuracy of the outcomes, is the use of only facial expressions in order to assess stress levels. It proved to be extremely hard for the coders to distinguish especially between attentive and stressed facial expressions; a sincere look could often indicate both stress and attentiveness.

Therefore, to ensure that indeed stress, and not partially attentiveness, is measured, other stress measures should be included in future studies, for it is conceivable that the more somber, sincere, and "hard" facial expressions we noticed in the computer situations, might actually be a result of the one-on-one environment which required more immediate responses from the students and may have increased activation levels. Even though we believe that this was not the case and what we measured was indeed high arousal caused by the use of the medium, our study might theoretically be seriously flawed if we were wrong. Therefore, more than one measure should be used.

Future studies should test different computer applications in different settings and different age groups in order to determine advantages and problems related to the implementation of computers for learning.

Of the research conducted so far, only a small portion has been conducted by independent researchers whose objectives have not been closely related to commercial goals. Scientific Research might help create new systems that are tailored to the needs of the millions of people whose everyday life will be increasingly determined by them over the next decades.

The question whether we will find many of these applications in various fields in the near future will largely depend on the amount and the results of systematic research. This study could contribute to the ongoing process.
 

References

Burgoon, J.K. (1985). Nonverbal signals. In M.L. Knapp & G.R. Miller (Eds.) Handbook on interpersonal communication. Beverly Hills, CA: Sage.(pp. 344-392).

Capella, J.N. & Greene, J.O. (1982). A discrepancy-arousal explanation of mutual influence in expressive behavior for adult and infant-adult interaction. Communication Monographs, 49.

Chesebro, J.,& Bonsall, D., (1989). Computer Mediated Communication. Tuscaloosa, AL: University of Alabama Press .

Cotton, R. & Oliver, R. (1993). Understanding Hypermedia. From Multimedia to Virtual Reality. London, England: Thaidon Press Ltd. Cutler, R.H. (1994, November). Distributed presence and community in cyberspace. Paper presented at the annual convention of the Speech Communication Association, New Orleans, LA.

Duffy, T. & Jonassen, D.H.Instructional Principles for Constructivist Learning Environments, Hillsdale, NJ.: Lawrence Erlbaum Associates, 1993.

Erickson, T. (1993). Data Visualization Environments: Problems and Prospects. In: Wexelblat, A. (Editor): Virtual Reality. Cambridge, MA: Academic Press Professional.

Giardina, M. (1991). Interactive Multimedia Learning Environments. New York, NY: Academic Press.

Gratz, R.D., Salem, P.J. (1984) Technology and the Crisis of Self. In: Communication Quarterly # 32, p. 98-103.

Kasschau, R., et al. (1982). Information Technology and Psychology Houston, TX: Praeger Publishers.

Larkin, J., & Chabay, R. (1992). Computer-Assisted Instruction and Intelligent Tutoring Systems. Hillsdale, NJ.: Lawrence Erlbaum Associates.

Lipton, M. (1994, November). Forgetting the body: Cybersex and identity. Paper presented at the annual convention of the Speech Communication Association, New Orleans, LA.

Mehrabian, A. (1976) Public Places and Private Spaces: The Psychology of Work, Play, and Living Environments. New York: Basic Books Inc.

Nickerson, R.S., & Zodhiates, P.P. (1988). Technology in education: Looking towards 2020. Hillsdale, NJ: Lawrence Erlbaum Associates.

Nix, D. & Spiro, R. (1990). Cognition, Education, Multimedia. Hillsdale, NJ.: Lawrence Erlbaum Associates.

Papert, S. (1980). Mindstorms. New York, N.Y.: Basic.

Piaget, J. (1929). The child's conception of the world. New York: Harcourt, Brace and Company.

Quarterman, J.S. (1989). The Matrix: Computer networks and conferencing systems worldwide. s.l.: Digital Equipment Corporation.

Robin, H. (1992). Die wissenschaftliche Illustration. Basel, Boston, Berlin: Birkhaeuser.

Shapiro, E. (1995, February). A functional analysis of computer mediated comunication in a software development team context. Paper presented at the convention of the Western States Communication Association, Portland, OR.

Vaske, J. & Grantham, C. (1990). Socializing the Human-Computer Environment. Norwood, NJ: Ablex Publishing.

Walther, J.B. (1994). Anticipated ongoing interaction versus channel effects on relational communication in computer mediated interaction. Human Communication Research, 20, 473-501.

Weir, G. R. & Alty, J.L. (1991). Human-Computer Interactions and Complex Systems. San Diego, CA.: Academic Press.

Wexelblat, A. (1993). Virtual Reality: Applications and Explorations. Cambridge, MA: Academic Press Professional.

Wyshynski, S. & Vincent, J.V. (1993). The Vivid Approach, in: Wexelblat A. (Editor): Virtual Reality. Cambridge, MA: Academic Press Professional.

Appendix
 
 
 

Scale for the Measurement of Facial Expression

Facial Expression:

2 - 1 - 0 - 1 - 2

Concerned - neutral - pleasant
 
 
 
 

Questionnaire

1. How hard did you think it was to work with the computer?

Very difficult ___

A little difficult ___

Really simple ___

2. How often do you usually use computers?

Every day ___

3-4 times a week ___

twice a week or less___

3. Range in Class: (Provided by teacher after questionnaires had been collected).

Good Student ___

Average Student ___

Poor Student ___

4. Gender (coded after session):

Male ___

Female ___
 
 

Table 1

T-Test for Paired Samples

___________________________________________________________________

Stress with Computer 2.36

Stress without Computer 1.36

              t= 5.16     p = .001

___________________________________________________________________

Table 2

Frequency of Computer Use and Stress Levels

Frequency Stress w. Computer (Mean) SD Cases

Table 3

Gender and Stress caused by Computer

Gender Stress w. Computer (Mean) SD Cases

Male                                            2.25     .91     12

Female                                        2.46     1.29    13

Table 4

Range in Class and Stress Levels

Range in Class Stress w. Computer

(Mean) SD Cases

Table 5

Perceived Difficulty of Application and Stress Levels

Perceived Diff. Stress w. Computer

(Mean) SD Cases

D.I.E. 2000, 2001