Introduction Traditional community based upon location, a history of interlocked relationships, has fallen away. What can replace it? (Cutler, 1994, p. 11) Computer mediated communication (CMC) is increasingly influencing not only organizational communication, but also private interpersonal relationships (Quarterman, 1989). Marriages and divorces have been attributed to online communication, and as early as 1989, computer networks such as the Internet have affected the social and private lives of millions of users worldwide (Quarterman, 1989). As Nickerson (1982) stated, "The implications that information technology has on our lives are beyond doubt very great". Previous research has focused mainly on organizational communication (e.g. Steinfield, 1992), and studies have concentrated on preferences for certain media for certain tasks (e.g. Markus, 1987), group problem solving activities (e.g. Vaske and Grantham, 1990), or contents of electronic mail or computer conference messages (e.g. Walther and Burgoon, 1992). Whenever relationships formed or maintained through network exchange have been researched thus far, a strong emphasis has been placed on single factors such as impression development, or on the success rates these relationships have produced for organizational tasks. This study, by contrast, addresses the issue of interpersonal relationships in CMC. It explores e-mail relationships with regards to the levels of relational satisfaction, intimacy, "inclusion", and interpersonal attraction in a medium that has frequently been described as "impersonal". It focuses on the question whether, due to the nature of computer mediated communication, lower levels of relational satisfaction, interpersonal attraction, intimacy, and "inclusion of other in the self" are obtained in this medium rather than in face to face communication settings. This constitutes one of the first attempts to systematically compare variables measuring the quality of relationships in face to face versus computer mediated situations. Systematic research of this kind might lead to a better understanding of the possibilities and limitations of a medium that could have large effects on the relationships we have in our lives.   Computer Mediated Communication The way you meet people in cyberspace puts a different spin on affiliation: in traditional communities, we are accustomed to meeting people, then getting to know them; in virtual communities, you can get to know people and then choose to meet them. (Rheingold, 1993, pp. 26-27) Cyberspace is the "virtual space" where humans communicate with each other via electronic technology (Lipton, 1994). The creation of a "virtual space" is an example of progress imposing modifications on our environment (Lipton, 1994). In his critical analysis of CMC, Cutler (1994) notes that "new social situations, new relationships, and the new roles that result place pressure on personal identities" (p.3). The term "computer mediated communication" (CMC) refers to a type of interpersonal communication which is facilitated through the use of a computer network or conferencing system. In contrast, "face to face communication" (FtF) describes an interpersonal communication setting in which all interactants share a common space and time context and can engage in communication where all five senses are used by the interactants. The growing importance of computer mediated communication has been noted by a number of researchers (Shapiro, 1995, Walther & Burgoon, 1992). Almost universal access to CMC is expected to be achieved through the implementation of the National Information Infrastructure (NII) as proposed by the U.S. government, and even today, computer mediated communication is changing the nature of communication for millions of users worldwide. However, CMC differs greatly from traditional face to face settings, because all nonverbal cues are filtered out by the "lean" medium.   The "Cues-Filtered-Out"-Approach Early CMC research utilized Social Presence Theory (Short, Williams, & Christie, 1976) and Media Richness Theory (Draft, Lengel, & Trevino, 1987) to develop the "cues-filtered-out" approach (Culnan & Markus, 1987). Media Richness Theory assumes that different media have different capacities for mediating cues; CMC was regarded an extremely "lean" medium. Social Presence Theory hypothesized that the feeling of being "present" in an environment was very low in a medium in which a variety of cues are "filtered out". Empirical studies found CMC to be more task-oriented, less emotional, and less personal (e.g. Hiltz, Johnson, & Turoff, 1986) than face to face communication. Rice and Love (1987) found less socioemotional content to be exchanged in CMC, and Sproull and Kiesler (1986) explained these and similar findings with the lack of "social context cues" (e.g., stationary, furniture, fragrances, nonverbal cues) in the single-channeled medium.   The "Social Information Processing"-Approach Contradictory findings from field experiments (Feldman, 1987; Weedman, 1991; as cited in Walther, 1994), however, were later believed to suggest that media bandwidth was an insufficient predictor of CMC effects on the nature of social interaction (Walther, 1994). In his analysis of impression development, Walther notes that "social cognition may affect how we utilize CMC ... and relate to others in this environment" (Walther, 1993a, p.382). In his "Social Information Processing" theory (1992), he argued that the critical difference between FtF and CMC was a question of rate of impression development rather than capability. Walther expected relational motivators such as the affiliation motive, and impression management and dominance drives, to encourage CMC users' use of "strategic probes" and similar strategies when decoding text-based cues. This should allow for the development of impressions of the dyadic partner which are of equal accuracy to impressions formed through FtF. Because of the single-channeled nature of CMC and the limited amount of message exchange, this process was expected to be retarded in CMC (see Walther, Anderson, & Park, 1994). Walther expected impression accuracy in CMC to slowly but steadily increase until levels comparable to FtF should be reached (Walther, 1993b). However, even though a study about impession development in FtF and CMC conducted by Walther in 1993 provided partial support for his theory, it did not prove it (Walther, 1993a). Another study conducted by Walther and Burgoon in 1992 examined various immediacy cues in CMC messages. Although for several conditions, no significant differences between CMC and FtF could be found, the Social Information Processing perspective on relational development was only partially supported (Walther & Burgoon, 1992).   Anticipated Ongoing Interaction versus Channel Effects In a recent study about the effects of time restriction on social interaction, Walther, Anderson, & Park (1994) hypothesized that time restrictions in previous studies had not given CMC groups enough time for impression development. Later, however, Walther (1994) found that in longitudinal studies, as he had suggested them in order to provide the CMC groups with enough time for impression development, even during the first meeting alone, participants demonstrated more socio-emotional behaviors than participants in one-shot studies did. This could not be explained by his Social Information Processing theory. Therefore, Walther suggested that rather than the duration of the relationship and impression development, the anticipated future interaction (which was higher in the longitudinal experiments) had higher effects on the socio-emotional content in CMC communication. He hypothesized that rather than the actual medium employed (CMC versus FtF), the anticipated future interaction influences relationship variables more than the nature of the medium. In zero-history one-shot studies, the drive to create a positive impression in the other person was supposed to be very low in CMC, and this fact was assumed to account for the small amount of socio-emotional content of CMC messages as found in earlier studies. However, Walther's overall findings, stating that the difference between CMC and FtF was smaller than originally assumed, could have been affected by the fact that he only analyzed the verbal parts of CMC and FtF messages. His analyses could therefore not tap into the crucial difference between CMC and FtF: the lack of nonverbal communication.   Nonverbal Communication, Relational Satisfaction, Intimacy, Inclusion, and Interpersonal Attraction Nonverbal communication is an essential part of human communication. Relationships without touch, for example, might be considered abnormal, even pathological (Andersen, n.d.). Similarily, lack of eye contact can prevent interactants from receiving relevant social information (Andersen & Coussoule, 1980). Perceptions of nonverbal immediacy cues such as forward body leans, smiles, touch, eye contact, direct pody position, etc., and perceptions of nonverbal backchannel behaviors are positively related to both perceived understanding and relational satisfaction (Stelzner, et al., 1994). For the purpose of this study, relational satisfaction is defined as a favorable affective response to the reinforcement of positive expectations in a relationship. The absence of nonverbal cues is expected to cause a decrease in the effectiveness of this reinforcement of expectations, and thereby cause a decrease in the perceived level of relational satisfaction. H1: In non-romantic, long-term, cross-gender relationships, the level of relational satisfaction is lower in CMC relationships than in FtF relationships. "Relational intimacy" describes "feelings and perceptions of union, closeness, interdependence, warmth, and affection" (Andersen, 1989, p.5). Nonverbal communication has been proven to be a crucial part of intimate relationships. Andersen (1985) noted that no function of nonverbal communication is more central to human interaction than the process of communicating warmth or intimacy to one another. Relational satisfaction is correlated to intimacy, and nonverbal behavior is believed to account for as much as two-thirds of the meaning in any social interaction in which such cues are present (Burgoon, 1985). It has, for example, been suggested that love cannot be successfully conveyed in the abscence of touch (Montagu, 1971). CMC lacks many aspects of nonverbal communication. It differs greatly from FtF communication, and it can be assumed that it lacks many of the cues necessary for the foundation of intimate relationships. H2: In non-romantic, long-term, cross-gender relationships, the level of intimacy is lower in CMC relationships than in FtF relationships. Intimacy can also be viewed as a process of an escalating reciprocity of self-disclosure in which each individual feels his or her innermost self validated, understood, and cared for by the other (Aron, Aron, & Smollan, 1992). It is characterized by the central idea of "sharing that which is inmost" with others (McAdams, 1988). Similarly, in the empathy model for explaining prosocial behavior, empathy was found to result in part from a basic confusion between ourselves and others (Wegner, 1980). These findings lead to the concept of a connection of selves in a close relationship, or "inclusion of other in the self". Nonverbal communication is believed to play an important role in this process, too. The level of nonverbal involvement can provide an important indicator of the quality of a relationship, and it seems that the characteristic patterns of nonverbal involvement in relationships can indicate the intimacy level of relationships (Patterson, 1988). Therefore, it is hypothesized that the lack of nonverbal communication in CMC leads to a lower level of relational inclusion of the other's personality in the self. H3: In non-romantic, long-term, cross-gender relationships, the level of inclusion of the other's personality in the self is lower in CMC relationships than in FtF relationships. Likewise, interpersonal attraction has been found to be a facilitator of interpersonal communication, and much interpersonal communication has been found to exist for the primary purpose of enhancing interpersonal attraction (McCroskey & McCain, 1974). In fact, one of the most dramatic ways in which the body communicates is through the degree of attractiveness (Malandro & Barker, 1983), and physical attraction is positively related to communication satisfaction (Zakahi & Duran, 1994). Physical appearance has been found to have great communication potential, and in experiments, subjects have attributed more positive qualities to physically attractive people (Burgoon, Buller & Woodall, 1989). Therefore, it can be hypothesized that the level of attraction is lower when no nonverbal cues are available, as is the case in computer mediated communication. H4: In non-romantic, long-term, cross-gender relationships, the level of interpersonal attraction is lower in CMC relationships than in FtF relationships. For the purpose of this study, it was assumed that the levels of relational satisfaction, relational intimacy, inclusion, and attraction could be measured accurately through self-reports. Likewise, in established relationships, impression development should already have been completed. Asynchronous e-mail relationships were compared with face-to-face relationships. The asynchronous CMC mode was employed for several reasons. First, asynchronous CMC facilitates more flexible and less stressful relational communication, according to Walther and Burgoon (1992). Second, as Walther and Burgoon suggested, "comparing FtF to asynchronous CMC maximizes experimental variance" (as cited in Walther, 1994, p.482). In a recent study conducted by Walther (1994), only asynchronous CMC differed significantly from FtF on several relational variables. This lead Walther to use only asynchronous CMC to test his further hypotheses, "in order to provide the most clear and meaningful hypothesis test" (Walther, 1994, p. 488). Even though Walther has described the fact that "no CMC research to date has examined the nonverbal visual behaviors of face to face groups as part of their total expressive output" (1992, p.63) as a major deficiency of previously conducted comparative FtF/CMC research, the interpretation of nonverbal cues was not considered necessary in this study, because the self-ratings which were used included the subjects' interpretations of their dyadic partners' nonverbal comunication, if availabe. Method Subjects The population from which the sample was drawn consisted of Southern Californian college students. Even though there has always been concern about the generalizibility of 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 only two types of relationships were compared, a sample size of 100 subjects was sufficient to obtain significant results. The study focused on dyadic, non-romantic, long-term, cross-gender relationships; the findings are generalizable only to the population from which the sample was drawn. Because subjects' self-reports were used as a measure for relational satisfaction, intimacy, inclusion, and attraction, only perceived levels could be analyzed, and the views of only one dyadic partner were available. Of the 100 respondents, 57 were male and 43 were female. The average age was 24.46 years, with a range from 18 to 44 years (Table 3). The respondents had been using e-mail for an average of 13.5 months, and 87 percent stated that they liked e-mail, with 13 percent stating they somewhat liked e-mail, and none of them stating they did not like e-mail. The majority of the respondents reported using e-mail between 15 minutes and one hour per week, with eight percent reporting less than 15 minutes of e-mail use, and 21 percent reporting more than one hour per week (Table 6). About half of the respondents had never seen their dyadic CMC partner; those who had, had not seen the other person for at least three months in order to qualify for participation. Procedure The data were collected at a Southern Californian University campus, where potential subjects were approached in front of computer laboratories. Only students who reported both FtF and CMC friendships (with a minimum length of one month for CMC) qualified for participation in the study. They were asked to think of both a CMC and a FtF non-romantic, cross-gender relationship, and to rate their quality on a self-administered questionnaire. For most items, ratings from 1 to 5 were used. The completion of the questionnaire took about 5-10 minutes; upon completion, subjects were reimbursed for their participation with a choice of a treat, snack, or soft drink. Measure After a pre-test of the questionnaire, several items were deleted, and instructions were clarified. The final questionnaire contained 52 items measuring sets of constructs that signal positive reactions towards interpersonal relationships. In order to measure interpersonal intimacy, relational satisfaction, inclusion, and attraction, several different scales were employed. The first section of the questionnaire measured intimacy, using a slightly simplified version of Andersen, Andersen, & Jensen's Generalized Immediacy Scale (1979). The second section of the questionnaire measured relational satisfaction, using a shortened version of Hendrick's Relational Assessment Scale (1988), a measure which is highly correlated with Spaniard's Dyadic Adjustment Scale (Metts & Cupach, 1990). Items from Spanier's Dyadic Adjustment Scale (1976), which were proven to be valid and reliable for measuring the quality of marriage and similar dyads, were also used to measure relational satisfaction. In their study about sources of satisfaction and conflict in long-term relationships, Argyle and Furnham (1983) identified the shared-interest factor as an almost universal source of relational satisfaction. Shared interest was therefore included as an item in the instrumentation. The third part of the questionnaire measured attraction, using modified versions of McCroskey and McCain's Interpersonal Attraction Scale (1974) and Snavely and Clatterbuck's Attraction Scale (1980). The latter scale employs parts of scales developed by Buchholz, et al. (1976) and several other scales. The third part of the instrumentation also contains items which were created in response to findings by Spanier (1976) and Argyle and Furnham (1983). The fourth part of the questionnaire measured the level of inclusion of the other in the self, using Aron and Aron's (1992) "Inclusion of Other in the Self" Scale (IOS), a graphic scale using Venn-diagram-like items. The last section of the questionnaire asked respondents for demographic information (age, gender, e-mail experience, duration of described relationships). Different rates of impression development for CMC and FtF have been reported as a source for inaccurate outcomes by Walther (1993a). Walther, however, concluded that over time, impression development in CMC and FtF are becoming close to equal, even though impression development may take longer in the "lean" medium of CMC. With means of 18.04 months for CMC and 18.14 months for FtF, the relationships which were analyzed in this study were of almost equal average lengths and could be regarded as established relationships (Table 3). Therefore, in both conditions, impession development should have been completed and should therefore not significantly influence the outcome of this study. Statistical Analysis Items measuring relational satisfaction were collapsed into a new variable; a low initial alpha reliability value for FtF lead to the exclusion of two items. All items measuring attraction were recoded into a new variable. Likewise, the five items on the modified "Generalized Immediacy Scale" were collapsed into a variable named "Intimacy" (Table 2). Subsequently, one-tailed t-tests for paired samples were performed to detect differences in the means for relational satisfaction, intimacy, inclusion, and attraction between CMC and FtF relationships. One-tailed t-tests were used because the hypotheses predicted that the means for the variables would be lower for CMC than for FtF.   Results The results for all four t-tests were significant (p<.05), indicating that there was a significant difference in means for all four variables and that the effects were in the directon stated in the hypotheses (Table 1). Therefore, all four hypotheses were confirmed. Hypothesis one concerning relational satisfaction was confirmed (t=-3.16; p=.001). The computed mean for relational satisfaction in FtF was 3.94, while the mean for relational satisfaction in CMC was 3.69. Intimacy was coded on a different scale ranging from 1 (high intimacy) to 5 (no intimacy). The means of 1.83 for FtF and 1.96 for CMC proved to be significantly different in the hypothesized direction, confirming hypothesis two, although the significance was comparatively low (t=-1.80, p=.0375). Similarly, hypothesis three regarding inclusion was confirmed, with means of 2.98 for FtF and 2.39 for CMC (t=-3.54, p=.0005). This hypothesis, as well as the fourth hypothesis, was confirmed at a very high significance level. Hypothesis four stated that the level of attraction was lower in CMC than in FtF. The mean for attraction in FtF was 3.92, compared to 3.68 for CMC (t=-3.39, p=.0005). Therefore, hypothesis four, too, was confirmed. Regarding anticipated future interaction, Walther expected the drive to create a positive impression in the other person to be very low in one-shot-study CMC, and he assumed that this fact accounted for the small amount of socio-emotional content of CMC messages as found in earlier studies rather than actual media differences. In the established relationships which were analyzed in this study, differences in the anticipated future interaction between CMC and FtF were found in the direction Walther suggested (i.e., more participants thought their CMC relationships would be over in six months, and more FtF relationships were believed to get closer within six months). However, the effects measured were small and are assumed to account for only a small portion of the overall findings (Table 4).   Discussion Results of the study provided support for all four hypotheses. Findings from the study provided empirical support for the preference for face to face settings as opposed to CMC for the fulfillment of positive human relationships. Relationships maintained in face to face settings are regarded to be more satisfying than relationships maintained through e-mail. This can be explained by the lower level of communication effectiveness in CMC due to a lack of redundancy of information elements usually provided by nonverbal cues. It has been demonstrated that communication effectiveness is closely related to communication satisfaction (Hecht & Marston, 1987). Even though the significance level for the intimacy variable was slightly lower than those of the other variables in this study, e-mail relationships were found to be less intimate than relationships in face to face settings are. This can easily be explained by the fact that nonverbal cues such as touch have been equated with intimacy itself (Morris, 1971), demonstrating the importance of nonverbal communication for intimate relationships. The same is true for perceptions of "inclusion of the other in the self". Less inclusion is felt in CMC relationships, at a very high significance level, indicating that the single-channeled nature of CMC is inadequately suited to capture the multifacetedness of cues required for the establishment of empathic understanding and feelings of union and closeness. People also feel more attracted to relational partners with whom they interact in real-world settings rather than in virtual environments. This result was also highly significant, and it demonstrates the common-sense truth that there is a higher potential for interpersonal attraction when people can see, feel, hear, smell, and touch each other as opposed to being in a situation in which characters on a screen are all they have at their disposal. In addition, people in relationships which allow for the exchange of nonverbal immediacy cues are also more inclined to be romantically interested in the other person, even if the effect size was small (Table 5). However, more "undecided" answers for the CMC relationships demonstrated that it is difficult for people to decide whether they may be romantically interested in another person if they only communicate with this person through e-mail. Walther's theory stating that impression development in FtF would over time plateau out, converge with CMC, and provide the basis for relationships of equal quality was not confirmed. Rather, the results of this study suggest that significant differences in the quality of relationships exist between FtF and CMC settings even in long-term relationships in which impression development has already been completed. Walther also hypothesized that incomplete impression development in previous studies had accounted for less socio-emotional content of CMC rather than the nature of the medium. Since established relationships of an average duration of 18 months were analyzed for this study, the results found can in fact be ascribed to the nature of the media themselves. In this regard, again, our findings contradict Walther's Social Information Processing Theory. Walther (1994) also suggested that the less personal character of CMC found in earlier studies may to a great extent actually have been due to different levels of anitcipated future interaction, which he found to be lower in CMC. Again, findings from this study contradicted Walther's hypothesis. Even though the anticipated future interaction tended to be slightly more negative for CMC, the effect size was small (Table 4). Therefore, results can be ascribed to the difference in media rather than the perceived future relationship. In these regards, our study confirms earlier findings of studies employing the "Cues Filtered Out" and the "Media Richness" approaches. The lack of nonverbal cues seems indeed to be the reason for the lower quality of relationships maintained through e-mail. In summary, it can be said that at its current developmental stage, CMC lacks both the breadth and depth required for the maintenance of satisfactory, intimate, close interpersonal relationships. This is not to say that CMC is not well suited for other types of communication, such as organizational communication. In addition, future CMC systems may at least in part allow for the mediation of nonverbal immediacy cues.     Suggestions for Future Research While the testing of sources for relational satisfaction would have exceeded the range of this study and was therefore not included, it is suggested for future research in this area to test differences in sources for relational satisfaction in CMC and FtF relationships. Similarly, correlations of demographic characteristics such as age and gender with satisfaction with CMC relationships may indicate which groups of people find e-mail relationships most problematic. It is conceivable that for some groups of people, such as communication apprehensives, elderly people, the physically challenged, or persons living in remote areas, e-mail may in fact be a medium which meets their communication needs; such sub-populations should be identified and researched. While this study focused on non-romantic relationships, further research should include romantic relationships, too (e.g., long distance relationships). Our study found close human relationships to be too vivid, ambiguous, and multi-channeled to be adequately captured by a single-channeled medium. Future research, however, may aid in the evolution of a medium which might at some point be capable of handling human emotions as trivial, complex, and diverse as happiness, sorrow, and love.       References Andersen, P.A. (1989, May). A cognitive valence theory of intimate communication. Paper presented at the International Network on Personal Relationships Conference, Iowa City, Iowa. Andersen, P.A. (1985). Nonverbal immediacy in interpersonal communication. In A.W. Siegman & S. Feldstein (Eds.) Multichannel Integrations of Nonverbal Behavior (pp. 1-36). Hillsdale, NJ: Lawrence Erlbaum. Andersen, P.A., & Coussoule, A.R. (1980). The perceptual world of the communication apprehensive: The effect of communication apprehension and interpersonal gaze on interpersonal perception. Communication Quarterly, 28, 44-54. Andersen, P.A. (n.d.). Arousal and anxiety. Unpublished manuscript. Andersen, J.F., Andersen, P A., & Jensen, A.D. (1979). The measurement of nonverbal immediacy. Journal of Applied Communications Research, 7, 154-180. Argyle, M., & Furnham, A. (1983, August). Sources of satisfaction and conflict in long-term relationships. Journal of Marriage and the Family, 481-493. Aron, A., Aron, E.N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality and Social Psychology, 63, 596-612. Buchholz, S., Lashbrook, W.B., & Wenburg, J.R. (1976). Toward the measurement and processing of social style. Paper presented at ICA, Portland, OR. Burgoon, J.K. (1985). Nonverbal signals. In M.L. Knapp & G.R. Miller (Eds.), Handbook on interpersonal communication. (pp. 344-392). Beverly Hills, CA: Sage. Burgoon, J.K., Buller, D.B., & Woodall, W.G. (1989). Nonverbal communication: The unspoken dialogue. NY, NY: Harper & Row. Culnan, M.J., & Markus, M.L. (1987) Information technologies. In F.M. Jablin, L.L. Putnam, K.H. Roberts, & L.W. Porter (Eds.), Handbook of organizational communication: An interdisciplinary perspective (pp. 420-443). Newbury Park, CA: Sage. 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. Draft, R.L., Lengel, R.H., & Trevino, L.K. (1987). Message equivocality, media selection, and manager performance: Implications for information systems. MIS Quarterly, 11, 355-366. Hecht, M.L., Marston, P.J. (1989): Communication satisfaction and the temporal development of conversations. Communication Research Reports, 4, 60-65. Hendrick, S.S. (1988). A generic measure of relationship satisfaction. Journal of Marriage and the Family, 50, 93-98. Hiltz, S.R., Johnson, K., & Turoff, M. (1986). Experiments in group decision making: Communication process and outcome in face to face versus computerized conferences. Human Communication Research, 13, 225-252. Lipton, M. (1994, November). Forgetting the body: Cybersex and identity. Paper presented at the annual convention of the Speech Communication Association, New Orleans, LA. Malandro, L.A., & Barker, L. (1983). Nonverbal Communication. Menlo Park, CA: Addison-Wesley. Markus, M.L. (1987) Toward a "critical mass" theory of interactive media: Universal access, interdependence, and diffusion. Communication Research, 14, 491-511. McAdams, D.P. (1988). Personal needs and personal relationships. In S. Duck (Ed.) Handbook of personal relationships: theory, research and interventions (pp. 7-22). New York: Wiley. McCroskey, J.C., & McCain, T.A. (1974). The measurement of interpersonal attraction. Speech Monographs, 41, 261-266. Metts, S., & Cupach W.R. (1990). The influence of relationship beliefs and problem-solving responses on satisfaction in romantic relationships. Human Communication Research, 17, 170-185. Montagu, A. (1971). Touching: The human significance of the skin. New York: Harper & Row. Morris, D. (1971). Intimate behavior. New York: Random House. Nickerson, R.S., & Zodhiates, P.P. (1988). Technology in education: Looking towards 2020. Hillsdale, NJ: Lawrence Erlbaum Associates. Patterson, M.L. (1988). Functions of nonverbal behavior in close relationships. In S. Duck (Ed.) Handbook of personal relationships: theory, research and interventions (pp. 41-56). New York: Wiley. Quarterman, J.S. (1989). The Matrix: Computer networks and conferencing systems worldwide. s.l.: Digital Equipment Corporation. Rheingold, H. (1993). The virtual community: Homesteading on the electronic frontier. Redding, MA: Addison-Wesley. Rice, R.E., & Love, G. (1987). Electronic emotion: Socioemotional content in a computer-mediated network. Communication Research,14, 85-108. 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. Short, J., Williams, E, & Christie, B. (1976). The social psychology of telecommunications. London: Wiley. Snavely, W.B., & Clatterbuck, G.W. (1980, November). The impact of social style on person perception and attraction across three relationship contexts. Paper presented at the annual convention of the Speech Communication Association, New York, NY. Spanier, G.B. (1976, February). Measuring dyadic adjustment: new scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 15-26. Sproull, L., & Kiesler, S. (1986). Reducing social context cues: Electronic mail in organizational communication. Management Science, 32, 1492-1512. Steinfield, C. (1992). Computer mediated communications in organizational settings: Emerging conceptual frameworks and directions for research. Management Communication Quarterly, 5, 348-365. Stelzner, M.A., Egland, K.L., Andersen, P.A. & Spitzberg, B. (1994, November). Perceived understanding, nonverbal communication, and relational satisfaction. Paper presented at the annual convention of the Speech Communication Association, New Orleans, LA. 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. Walther, J.B. (1993a). Impression development in computer mediated interaction. Western Journal of Communication, 57, 381-398. Walther, J.B. (1993b). Construction and validation of a quantitative measure of impression development. Southern Communication Journal, 27-33. Walther, J.B. (1992). Interpersonal effects in computer mediated interaction: A relational perspective. Communication Research, 19, 52-90. Walther, J.B., Anderson, J.F., & Park, D.W. (1994). Interpersonal effects in computer mediated interaction: A meta-analysis of social and anti-social communication. Communication Research, 21, 460-487. Walther, J.B. & Burgoon, J.K. (1992). Relational communication in computer mediated interaction. Human Communication Research, 19, 50-88. Wegner, D.M. (1980). The self in prosocial action. In R.R. Vallacher (Ed.) The self in social psychology. New York: Oxford University Press. Zakahi, W.B. & Duran, R.L. (1994). Attraction, communicative competence, and communication satisfaction. Communication Research Reports, 1, 54-57.     Appendix:  Questionnaire   Part 1: Please think of a person of the opposite sex with whom you have been in contact only over e-mail. The relationship should have existed for at least three months. It should be non-romantic, and you should not have seen this person for at least three months. It could be a person you have never met in person. Please answer the following questions:   1. Please place an "x" in each of the following scales to indicate the word that best describes the conversational style of the other person: Involved    __ I __ I __ I __ I __ Uninvolved Cold          __ I __ I __ I __ I __ Warm Unfriendly  __ I __ I __ I __ I __ Friendly Close         __ I __ I __ I __ I __ Distant Pleasant     __ I __ I __ I __ I __ Unpleasant   2. Please indicate your level of agreement or disagreement with the following statements in regard to the e-mail relationship. 5 - Strongly Agree, 4 - agree, 3 - undecided, 2 - disagree, 1 - strongly disagree. __ 1. In general, I am satisfied with this relationship. __ 2. This relationship is good, compared to most. __ 3. I do not regret having gotten into this relationship. __ 4. This relationship met my original expectations. __ 5. There are few problems in my relationship. __ 6. We share many interests. __ 7. I think I can trust this person. __ 8. I feel I know her/him personally. __ 9. It is hard to talk about intimate issues with her/him. __10. This person understands me as an individual.   3. Please indicate your level of agreement or disagreement with the following statements in regard to this person. 5 - Strongly Agree, 4 - agree, 3 - undecided, 2 - disagree, 1 - strongly disagree. __ 1. We could never establish a personal friendship. __ 2. It would be difficult to meet and talk with this person. __ 3. I like to spend time with this person. __ 4. I sometimes wish I were more like this person. __ 5. This person just would not fit into my circle of friends. __ 6. This person is dependable. __ 7. This person makes me feel good about myself. __ 8. This person does not like me. __ 9. This person is concerned about me. __10. This person smiles a lot.   4. Please answer the following questions. 1. For how long have you known this person? __ __ months. 2. Have you ever personally met this person? Yes __ No __. 3. Are you interested in getting romantically involved with this person? Yes __ No __ Don´t know __. 4. What do you think this relationship will be like six months from now? Closer than now __ about the same __ more distant/over __ DK __.   Part 2: Now, please think of another person of the opposite sex, with whom you have a non-romantic relationship. This relationship should not be an e-mail relationship. Rather, think of a person who you see regularly, e.g., a friend from college or work, or a roommate. Please answer the following questions:   1. Please place an "x" in each of the following scales to indicate the word that best describes the conversational style of the other person: Involved   __ I __ I __ I __ I __ Uninvolved Cold         __ I __ I __ I __ I __ Warm Unfriendly __ I __ I __ I __ I __ Friendly Close        __ I __ I __ I __ I __ Distant Pleasant    __ I __ I __ I __ I __ Unpleasant   2. Please indicate your level of agreement or disagreement with the following statements in regard to the e-mail relationship. 5 - Strongly Agree, 4 - agree, 3 - undecided, 2 - disagree, 1 - strongly disagree.   __ 1. In general, I am satisfied with this relationship. __ 2. This relationship is good, compared to most. __ 3. I do not regret having gotten into this relationship. __ 4. This relationship met my original expectations. __ 5. There are few problems in my relationship. __ 6. We share many interests. __ 7. I think I can trust this person. __ 8. I feel I know her/him personally. __ 9. It is hard to talk about intimate issues with her/him. __10. This person understands me as an individual.   3. Please indicate your level of agreement or disagreement with the following statements in regard to this person.   5 - Strongly Agree, 4 - agree, 3 - undecided, 2 - disagree, 1 - strongly disagree. __ 1. We could never establish a personal friendship. __ 2. It would be difficult to meet and talk with this person. __ 3. I like to spend time with this person. __ 4. I sometimes wish I were more like this person. __ 5. This person just would not fit into my circle of friends. __ 6. This person is dependable. __ 7. This person makes me feel good about myself. __ 8. This person does not like me. __ 9. This person is concerned about me. __10. This person smiles a lot.   4. Please answer the following questions. 1. For how long have you known this person? __ __ months. 2. Are you interested in getting romantically involved with this person? Yes __ No __ Don´t know __. 3. What do you think this relationship will be like six months from now? Closer than now __ about the same __ more distant/over __ DK __.   5. Please tell us about yourself. This survey is anonymous, and your answers will be kept confidential. 1. What is your age? _ _ years. 2. What is your gender? Male __ Female __. 3. For how many months have you used e-mail? _ _ months. 4. How much time per week do you usually spend using e-mail? 15 min. or less __ 16-30 min. __ 31-60 min. __ 1-2 hours __ 2-4 hours __ more than 4 hrs. __   5. Would you say that you enjoy e-mail? Yes __ Somewhat __ No __ Don't know __.