Cyber-Research: Metamorphosis on the Internet

ABSTRACT - This paper discusses the possible influences of the Internet on researchers’ and respondents’ objectivity. Cyberspace is a new environment for consumer behavior research because it exists only in the mind of the participants. The author reflects on three years of e-mail interviewing and participant observation in interest groups on the Internet. What emerges is the possibility of a researchers’ metamorphosis into someone more like their subject. This can be a positive aspect of research because much can be learned by becoming a more like one’s subjects.

Citation:

Charles A. McMellon (2001) ,"Cyber-Research: Metamorphosis on the Internet", in E - European Advances in Consumer Research Volume 5, eds. Andrea Groeppel-Klien and Frank-Rudolf Esch, Provo, UT : Association for Consumer Research, Pages: 240-244.

European Advances in Consumer Research Volume 5, 2001      Pages 240-244

CYBER-RESEARCH: METAMORPHOSIS ON THE INTERNET

Charles A. McMellon, Hofstra University, U.S.A.

ABSTRACT -

This paper discusses the possible influences of the Internet on researchers’ and respondents’ objectivity. Cyberspace is a new environment for consumer behavior research because it exists only in the mind of the participants. The author reflects on three years of e-mail interviewing and participant observation in interest groups on the Internet. What emerges is the possibility of a researchers’ metamorphosis into someone more like their subject. This can be a positive aspect of research because much can be learned by becoming a more like one’s subjects.

Consumer behavior on the Internet is attracting the interest of academic researchers who use traditional ethnographic methods (e.g., Moore 1995; Rheingold 1995; Turkle 1995). These researchers are observing and interviewing consumers in their natural setting (i.e., the Internet). Because the Internet is a new environment and appears to be different than the real world environments where research usually takes place, consumer researchers may need to reexamine some of their assumptions about methods, especially the level of objectivity they maintain while conducting depth interviews or participant observations. This new Internet environment-cyberspace-has not yet been clearly defined. It has been described as a consensual hallucination (Gibson 1984), a technological environment where humans can interact (Sterling 1990), and an environment where the individual perceives they are present (Steuer 1992). Certain words in these definitions (i.e., hallucination, perceives, interact, and technological) suggest that this environment may be different in some ways from the traditional research environments. Thus, Internet researchers should proceed with caution when observing or interviewing consumers.

Cyberspace is hypothetical. In reality, individuals become an integral part of cyberspace when they turn on their computer, stare at their web browser (e.g., Netscape), and send their electronic messages at the speed of light back and forth across the wire grid that is commonly called the Internet. While individuals interact in this manner, they may perceive they are in cyberspace; yet all that is cyberspace is in their computes, electronic messages, and wires. One explanation for this perception is that connecting up to the Internet may be an activity that allows individuals to escape the physical world to the virtual world. This virtual world is actually in their own minds (Lupton 1995; Turkle,1995). In this context, this paper suggests that cyberspace may be a new type of research environment (i.e., it takes place in the mind) with its own set of influences on those who are active in it. These cyberspace influences consist of the computer mediated effects of diminishing socially desirable responses and increases in candor, the ease, intimacy, and immediacy of electronic communication and, the perception that cyberspace is a real place. The author’s online experience, which consists of a two year period of depth interviewing using e-mail and over fifty hours as a participant observer in an online chat group, is the basis for this discussion.

The purpose of this paper is to discuss the possible effects of online depth interviewing and participant observation on both interviewer and informant. First, virtual participant observation (VPO) and virtual depth interviewing (CDI) are described. Then, to aid in the understanding of virtual interviewing and participant observation, the previously studied effects of computer mediated communication (CMC) are discussed along with suggested new effects that are emerging as a result of an individual’s participation on the Internet. Examples of the effects identified in the author’s previous research are presented. Next, the possible effects are discussed through a metamorphosis that takes place in the interviewer, the informant, and their relationship. Last, this paper discusses the possible implications of a lessening of objectivity during the research process while gaining in understanding.

INTERNET RESEARCH METHODS

Virtual participant observation (VPO) occurs when the researcher joins an online chat group. Chat groups are gatherings of individuals with similar interests (e.g., stamp collecting or genealogy) who "chat" with one another in real time. VPO is similar to traditional participant observation in that interaction takes place in the social milieu (i.e., cyberspace) of the subjects and data is collected systematically and unobtrusively. It differs in that, (a) visual observation is absent, forcing the researcher to rely on observation of the messages that are sent back and forth; (b) all conversations are recorded electronically; and (c) instead of going out into the field, the researcher sits quietly in front of his or her computer screen, gathering the information from cyberspace.

Virtual depth interviewing (VDI) occurs when the researcher interacts, one-on-one in cyberspace, with an informant who has agreed to be interviewed. This electronic interview can take place in real time or in a more extended version using e-mail which involves sending messages back and forth in a timely manner. These e-mail messages are similar to verbal reports used in traditional ethnography. VDI is similar to traditional depth interviewing in that it is conversational in nature and is one-on-one; but it is not face-to-face, it is screen-to-screen. Thus, the researcher loses the opportunity to interpret or use visual cues. For example, unplanned prompts such as raising one’s eyebrow can not be used by the researcher. In addition, the researcher must rely on text for interpretation of the subject’s tone or emotional level which may limit the analysis.

Researchers, whether using participant observation or depth interviewing, are usually seeking understanding in a natural setting from the words or actions of the individuals involved in the phenomena. The collected data, which can take many forms (e.g., verbal reports, field notes, and videotapes), is then analyzed. Virtual depth interviewing and virtual participant observation appear to be similar in some aspects with the traditional ethnographic approaches to these methods but, more importantly, different in some aspects (e.g., no visual clues).

METHOD

The author’s involvement in two previous research projects form the basis for the current discussion. One project consisted of nine depth interviews conducted over the Internet for a period of two years. This project was a pilot study to develop potential variables that might have an influence on the allocation of Internet time. The second project involved participant observation in 50 hours of "chat room" discussions with a group of consumers on a major commercial service provider. This study examined word-of-mouth on the Internet.

The virtual depth interviews in this study were conducted using e-mail. This approach allows the researcher and the informant the ability to answer at their leisure in a more thoughtful manner. Real time interviewing, although capturing the moment, appears to be problematic for some informants because they do not want to spend the continuous time online in an interview because of online cost or lack of sustained interest. Adopting the long interview approach of McCracken (1988), a prolonged, systematic e-mail communication between researcher and informant was initiated. Informants were solicited to participate by e-mail. If they agreed to participate, the researcher began a series of e-mails with each individual that involved asking questions, prompting them on their comments, and responding to the informant’s questions. The subject matter for this two-year study was an exploration of their involvement in computers and the Internet.

Virtual participant observation was accomplished by the author joining a retirement oriented forum (i.e., chat group) on a major commercial service provider. Membership in this forum is open to anyone. The forum met twice a week in real time on the Internet for at least an hour, sometimes longer, to discuss topics as health, weather, politics, or retirement living. Participation in a real time session requires some typing skills and a knowledge of how to send your message so that everyone in the group sees it. The individuals involved can just watch the discussion (i.e., lurking) or participate. Just watching can be difficult as the other participants will sometimes address the lurker (everyone knows who is attending), trying to engage them in the conversation.

COMPUTERS AND INTERNET MEDIATED EFFECTS

The level of autonomy in communication can influence socially desirable responses (Nederof 1989; Paulhus 1984). This may be true for computer mediated communication (Kiesler and Sproull 1986; Lautenschlager and Flaherty 1989; Martin and Nagao 1989). The most common effects studied are that socially desirable responses of the individual diminish and that candor increases. In addition to these previously studied effects, this paper identifies other effects that may be emerging as more individuals use the computer as a means to communicate on the Internet. This paper adds to the discussion of computer mediated communication (CME) by introducing two new possible effects: (a) the perception that cyberspace is real and, (b) the ease, intimacy, and immediacy of electronic communication.

Social Desirability and Candor

Socially desirable responses involve two components: impression management and self- deception (Paulhus 1984). Impression management is when an individual consciously reports behaviors in a manner that emphasizes desirable behaviors and under emphasizes undesirable behaviors. For example, a conversation occurred between the author and an interviewee about Gallic humor. As an example of the humor, this informant emphasized a desirable behavior about himself.

"I’m 64 years old but my performances are the same as 20...What I hope...at 90 years old I would be murdered with my 20 years old 'paramour,’ killed by her jealous husband (;-)."

Self-deception is when respondents report honestly, but unconsciously skew their answer to be more positive than reality (Booth-Kewley, Edwards, and Rosenfeld 1992). Although, the researcher may not be able to identify the self-deception aspect of socially desirable responses easily because of its unconscious nature, both components, impression management and self- deception, may influence responses in electronic research. Other researchers have also found examples of impression management on the Internet. For example, Shaw (1997) reports a conversation with a gay male who states "You’ve heard of IRC inches?...somebody says eight and you know its probably five."

Candor may also fall under the domain of social desirability. When communication is by computer, candor appears to increase because of perceived anonymity (Bradburn and Sudman 1979; Nederhof 1984; Wiseman 1972). As a result, sensitive subjects are more easily discussed and respondents are more open in their remarks. While social desirability focuses on what the individuals say about themselves, candor appears to include more on what individuals have to say about others. Candor appears to increase during communication in cyberspace where there may be a breakdown in social politeness. One of Moore’s (1995, p. 93) informants confirms this notion, "When you go online...you go a little bit farther, you’re a bit more frank and open. There are some things I would say...that I would never say to my own friends." The openness of those who communicate with candor about others may be the result of the anonymity of cyberspace. The individual is not in a face-to-face situation and is not subject to immediate feedback or any other visual aspect of the situation which might influence his or her responses in a socially desirable manner.

In another case, the author recruited Internet users via e-mail. Most who were contacted responded in a socially desirable manner, but not all. In some individuals candor appeared to increase as demonstrated with these e-mail responses to the author: "Go fuc you self (sic);" and "You got this far, look me up in the yellow pages." It appears that these individuals, who may have been influenced by the anonymity of electronic communication, felt free to challenge the researcher.

Increased candor also appears in real time chat groups. When the author was participating in a chat group discussion about gardening, one of the group apparently became bored and responded with: "ZZZZZZZZZZZZZZ." Another time while discussing Texas’ succession from the Union a member said: "If at first you don’t succeed. Suck once more." The individual who wrote this responded almost immediately with an apology and commented that "It just popped out." Most participants are polite, but it appears as if the Internet will facilitate those who wish to make less socially desirable responses.

While this paper suggests the influence of computers and the Internet on levels of candor, other research results have been mixed. Kiesler and Sproull (1986) found fewer socially desirable responses in electronic versus mail surveys. Also Rosenfeld, et al., (1991) found no effect when comparing computerized surveys to pen and pencil surveys. They suggested impression management may not be an influence in electronic communication because the self-monitoring aspects of the individual’s personality are muted by the lack of visual and verbal feedback cues which, in face-to-face situations, increases socially desirable responses. These findings are understandable because they are computerized surveys. The findings in the current discussion are explained in that they were the result of interaction on the Internet where there is a higher level of anonymity and the messages were conversational in nature.

The Perception That Cyberspace is Real

In addition to the lessening of socially desirable responses and increases in candor, the individual who goes online may also be influenced by the environment where they interact (i.e., cyberspace). Turkle (1995) suggests that computer users develop relationships with their computer. In many ways, they define themselves by how they use the computer and the Internet (i.e., by whom they meet, where they go, and how they act). The computer and the Internet can become an extension of the self. Online individuals name their hard drives, decorate their computer with human accessories like Looney Tunes mouse pads or screen savers with family photos (Terenzini 1997). Some even talk to their computer. As the individual interfaces with their computer, identity can become more fluid. In discussing the cultural aspects of Internet use, Turkle suggests that users abdicate authority to the simulation. They don’t really care much about how the computer works; they assume that what appears on the screen is real. It is, as she says, "the seduction of simulation." For example, one respondent while being interviewed by the author, began talking about the guilt that arose when cheating on a computer game. This suggested a possible deeper relationship with his computer. He states,

"Playing scrabble against the computer does seem as if you are playing another person...sometimes if the machine gets ahead of me.....I cheat.....Ridiculous isn’t it?"

Moore (1995) also supports the notion that cyberspace influences behavior when he suggests computer users confuse virtual reality with the physical world in the games they play and the places they visit online. This confusion may contribute to a situation that allows the user’s deepest fantasies to surface because it is safe online and the virtual reality personality you create can receive little direct criticism or retribution. Over time, individuals may begin to confuse their online personality with their real world personalities believing that who they are online is the same person as who they are off-line.

In addition to the influence of the cyberspace environment on individuals, the individual’s attitudes toward technology, especially when taken to an extreme, may also be a factor. Tse, et al. (1995, p.441) identified something that they called "...the intangible quality and mystique associated with electronic communications and advanced technology." This attitude of mystique also appeared in the words of informants. The words "astounding" or "fascinated" which might be characterized as code words (Arnould and Wallendorf 1994) appear many times in the interviews conducted by the author and the chat discussions attended when discussing technology. These code words can be interpreted as the mystique of technology.

"I was fascinated by the concept of computers,..."

"..the process became so clear, I was astounded"

"Yes, I am amazed at what technology can achieve..."

"As far as I can see, technology will bring us higher and higher..."

Because of this strong positive attitude toward the Internet, the individual’s perception of virtual reality can become biased and influence their interpretation of events or texts that appear on their screens. That is, they begin to believe everything they see on the Internet.

Intimacy, Ease, and Immediacy of Electronic Communication

Many individuals, when logged on to the Internet, sit in the quiet of their office or study. They position themselves before their computer, undisturbed and undistracted, focusing on the screen while they type messages. Their thoughts appear on their screen as they think them. This environment of the Internet user suggests a sense of intimacy between the user and the screen. It can be a sense of oneness with the screen. This intimacy can influence objectivity. The user can become more friendly, more attached to the communications which appear on the screen. In addition, the design of most e-mail software allows for an ease and immediacy of communication. That is, response to an e-mailed question requires typing an answer. Then, with the click of the send button, the message is electronically sent over the Internet, painlessly and efficiently. Moore (1995, p. 45) speculates on why this ease and immediacy in e-mailing is important:

"...it can be sent so quickly, with a push of a button, and second later it arrives at its destination. This surely encourages the impulsive nature in the writer. Am I sure I want to say that? Oh hell, just send it."

THE METAMORPHOSIS

This paper suggests that the method of communication and the environment used in virtual research (i.e., Computers and the Internet) will influence the informant, the researcher, and their communication. As the relationship between the researcher and informant or group progresses during the research process, a metamorphosis of the researcher can take place. The genesis of this metamorphosis begins early in the research process because cyberspace informants can easily refuse to be interviewed or terminate the interview before any meaningful information can be collected. The potential informant has little face-to-face guilt problems in saying "no" to the Internet researcher. Thus, to develop a successful research interaction, the interviewer must quickly find a level of cordiality and trust to entice and hold the informant’s interest. The interviewer can not just ask questions because of the ease in which informants can end the interview. Cordiality can be quickly accomplished, but not without the loss of some objectivity because the researcher becomes more personal. If the researcher is successful in building cordiality to hold the informant’s interest, a relationship begins to develop and over time trust between the researcher and informant develops. As with most relationships neither side is completely passive. Without being asked, informants begin to ask their own questions of the researcher. For example,

"Charles, you know now a lot about me, I’d be glad to read a little about you by return. Please tell me shortly what is the kind of your studies and a little of your life"

"I’ve a couple of questions for you. About how many people have you surveyed in this project of yours? Are you concentrating on a certain age or other demographic group, are you trying to at representable sample of people online? Hope you are keeping cool. It’s pretty hot here right now."

"Charles, before I answer your latest complex question, I have one for you."

In each of these cases, the author felt it necessary to answer the informant. As for discussing the study with informants, the author was able to delay comment until the depth interview process was completed. The questions about the author’s personal history could not be delayed because of the need to build trust. Thus, the author began e-mailing messages about his own personal history to the informant. As a researcher shares their personal history, the relationship deepens and a lessening of objectivity may occur.

In addition, informants, because they are free to write what they want in the safety of their own home, begin to discuss matters other than originally planned by the researcher. When this occurs, the researcher can not easily ignore the informant’s request without jeopardizing the completion of the interview. For example, a British informant e-mailed the author:

"To change the subject completely - I wonder if you have any contacts in the USA publishing world? Two years ago I had open heart surgery and have had four bypasses. I wrote a short book which my wife, who is a commercial artist, illustrated... Any ideas?"

In this case, the author contacted a friend in publishing and wrote back some recommendations on what this informant should do to publish his book in the United States. For some informants, as trust grows and the effects of social desirability lessen, discussion sometimes became more personal. The informant appears to forget that an interview is taking place. One international informant began asking why none of his American friends had responded to the Thanksgiving Day greeting cards he had sent. The informant wanted to know if he had done something wrong. To maintain the relationship, the author discussed this issue with the informant although it had nothing to do with the research. In another example, one respondent began telling his life story without any prompting from the researcher. Most of the comments revolved around his current situation, which had made him "a prisoner in my own home." This respondent stated that his mother, who he was care giver for, had Alzheimer’s disease, his daughter had cancer, he had lung disease, and his son had left the house because he couldn’t take the pressure any more. Although these statements appear to be over generalizations reflecting the informant’s current situation, they brought the researcher and informant closer together and understanding increased. His messages were personal and frank suggesting increased candor:

"Your’s is the first message, I will catch up, but not now. I am tired, hungry, and getting drunk."

"..I’ve just have been in a funk lately. Some days I just don’t get out of bed."

Of special interest is the finding that the interviewer is also influenced by the effects of CMC and the Internet. If candor rises in the informant, it may also rise over time in the researcher. For example, a prospective informant, who had received an e-mail from me asking him to participate in a survey, wrote back " go fuc you self (sic)," the author immediately e-mailed a sarcastic reply, thanking him for the polite message and suggesting that correct grammar and spelling might be more effective. This individual replied again, this time with the correct spelling and grammar. The message was all caps too, suggesting an increase in emotional tone of the message. The researcher, like those individuals in Turkle’s (1995) work, sometimes becomes lost in the screen. This mental state, when combined with the ease, intimacy, and immediacy of e-mailing appeared to facilitate responses when objectivity might dictate that no response is necessary. In another case, an informant accidentally sent an e-mail to the author, which was meant to be sent to a close relative. In the e-mail, the respondent appeared disoriented and physically ill. In response to this, the author telephoned the informant to inquire about his health. This incident further bonded the two although, traditional ethnography might suggest distance between the researcher and informant. In a third case, the author received a cynical and paranoid e-mail refusal to a request to participate in the study. The author wrote back to the individual who had refused to participate in a survey suggesting the message writer not be so cynical and that there are many people on the Internet, including the researcher, who are honest and not out to cheat them. In this case, the author was applying some impression management to the situation. These are examples that show that the relationship that develops between the researcher and subject along with the effects of diminished socially desirable responding (i.e., impression management and candor) and, the ease, intimacy, and immediacy of the technology support the notion that a metamorphosis may have taken place as the researcher evolves into a less objective participant.

The bonding and honesty of the relationship that develops between the researcher and informant also influence the relationship. Researchers might find themselves straying from the original research objectives and beginning unrelated conversations. For example, when a sudden death of a friend occurred, the author shared his grief with informants. All e-mailed quickly with comments to console the author. Again, there was a bonding that allowed the author to better understand the deeper meanings of informant’s behaviors. After two years of interviewing, the informants and researcher know much about each other. The researcher’s detailed knowledge of informants and understanding of their online behavior could not have been attained without this sharing of the researcher’s personal history.

This paper suggests that, within the context of prolonged virtual research, either depth interviewing or participant observation, a metamorphosis can take place. A metamorphosis driven by the honesty and bonding nature of the relationship, by the lessening of socially desirable responses and increases in candor, by the ease, intimacy, and immediacy of the communication process, and by the effect of cyberspace itself. In this metamorphosis, the researcher becomes part informant/part participant, and manufactured distance begins to dissolve. One explanation for this is that, because of these influences, researchers on the Internet may be reinventing themselves as the idealized interviewer or participant because they are released from many of the real world constraints placed upon them when they proceed through the cyber-research process.

As the researcher loses some of his or her objectivity and manufactured distance diminishes, the interview or observed event may evolve to a form dictated more by the research subjects or the effects of the computer and the Internet mediated communication and less that of the researcher.

DISCUSSION

Participant observation and depth interviewing have been used in this study to discuss potential influences on the respondent and researcher. They also demonstrate the potential for using ethnographic methods on the Internet. They extend the scope of these methods into a virtual reality of the Internet and challenge the notion of personal contact as a necessary component.

McCracken (1988) suggests two issues limit qualitative interviewing technique: the time scarcity of informants and their concern for privacy. Virtual interviewing begins to resolve these two limitations. Informants can respond at their leisure and in the privacy of their home or office with some degree of anonymity. For participant observation, Arnould and Wallendorf (1994) point out that it is key for the researcher to gain access to backstage areas if they want to develop the "...complex, textured interpretations of culturally constructed behavior" (p. 486). This may be difficult in many traditional consumer research situations because of socially desirable responses and lack of candor among informants.

Virtual research may aid in overcoming these obstacles by partly metamorphsizing the researcher into an informant. The loss of some researcher objectivity allows for a deeper bonding between researcher and subject. In addition, experiencing a phenomena first hand (e.g., by the influence of increased candor) can allow insights that might be overlooked by the more objective observer. Previous researchers have already used first hand experience in their research either through introspection (e.g., Gould 1991) or actual experience (Hirschman 1990).

Virtual research can be used in market ethnography because it meets the goals suggested by Arnould and Wallendorf (1994). It is systematic in data collection, recorded in a natural setting (albeit virtual), can produce credible interpretations, and is one of the possible sources of data. The market ethnography approach also suggests the necessity of a manufactured distance between researcher and informant which may not be the case in virtual research. Manufactured distance is consistent with the traditional ethnographic approach which suggests that the personal experiences of the researcher, while necessary in the participant observation and depth interview processes, should be excluded from ethnographic writing because they are not scientific (Clifford 1986). Contrary to this point of view, a sub-category of ethnographic writing has emerged that discusses the subjective experiences of researchers (e.g., violence, desire, confusion, struggles with informants). The argument this sub-category makes is that personal experience is an integral part of the research process and should not be excluded as irrelevant (e.g., action research, participatory research). To understand a scientific phenomenon more fully, the researcher should have access to all the information that might lead to greater understanding, not just those deemed relevant by objectivist science.

This paper suggests that, like Gibson’s (1984) cyberspace cowboys, we jack into the Internet, connecting up with respondents in a bond unknown to traditional face-to-face interviewers or those who participate to observe. If the computer is an extension of the self as suggested by Turkle (1995) and computer and Internet communication has an influence, then researcher and informant alike may be influenced by its mediating effects. On the Internet, socially desirable responses are lessened and candor increases. In addition, the environment itself (i.e., cyberspace effects), and the ease, intimacy, and immediacy of electronic communication play a role in the virtual research process.

This influence should not be looked upon as diminishing the effectiveness of the research process. Personal experience may be able to bring us to a deeper understand of consumer behavior. Barbara McClintock, the Nobel prize winning biologist, when asked how she saw cellular structures that no one else could see, said she found them by "thinking like" them (Keller 1983). McClintock’s point of view is worth noting. What understanding might be overlooked in exploring cyberspace if subjective insights are removed from discussion?

REFERENCES

Arnould, Eric, J. and Wallendorf, Melanie (1994). Market-oriented ethnography: Interpretation building and marketing strategy formulation. Journal of Marketing Research, 33(November), 484-5-4.

Booth-Kewley, Stephanie, Edwards, Jack E., and Rosenfeld, Paul. (1992). Impression management, social desirability, and computer administration of attitude questionnaires: Does the computer make a difference. Journal of Applied Psychology, 77(4), 562-566.

Bradburn, Norman M., and Sudman, Seymour (1979). Improving interview methods and questionnaire design. San Francisco: Jossey-Bass.

Clifford, James (1986). Introduction: Partial truths. In James Clifford and George E. Marcus (Eds.), Writing Culture. The poetics and politics of ethnography (pp. 1-26). Berkeley: University of California Press.

Gibson, William (1984). Neuromancer. New York: Ace Books.

Gould, Stephen, J. (1991). The self-manipulation of my pervasive, perceived vital energy through product use: An introspective-praxis perspective. Journal of Consumer Research, 18(September), 194-207.

Hirschman, Elizabeth, C. (1990). The day I almost died: A consumer researcher learns some lessons from a traumatic experience. In Research in Consumer Behavior, Vol.4, ed. Elizabeth C. Hirschman. Greenwich, CT: JAI, 109-123.

Keller, Evelyn Fox (1983). A Feeling for the Organism. The Life and Work of Barbara McClintock. San Francisco: W. H. Freeman and Company.

Kiesler, Sara and Sproull, Lee S. (1987). Response effects in the electronic survey. Public Opinion Quarterly, 50, 402-413.

Lautenschlager, Gary J., and Flaherty, Vicki L. (1990). Computer administration of questions: More desirable or more social desirability?, Journal of Applied Psychology, 75(3), 310-314.

Lupton, Deborah (1995). The Embodied Computer User, in Mike Feathers and Roger Burrows (eds) Cyberspace/Cyberbodies/Cyberpunk: Cultures of Technological Embodiment. Thousand Oaks, CA: Sage Publications, Inc.

Martin, Christopher, L. and Nagao, Dennis H. (1989). Some effects of computerized interviewing on job applicant responses. Journal of Applied Psychology, 74 (1), 72-80.

McCracken, Grant (1988). The Long Interview: A four-step method of qualitative inquiry. Newbury Park: Sage Publishers.

Moore, Dinty W. (1995). The Emperor’s Virtual Clothes. Chapel Hill: Algonquin Books of Chapel Hill.

Nederhof, Anton, J. (1984). Visibility of response as a mediating factor in equity research. The Journal of Social Psychology, 122, 211-215.

Paulhus, Delroy, L. (1984). Two-component models of socially desirable responding. Journal of Personality and Social Psychology, 46(3), 598-609.

Rheingold, Howard (1995). The virtual community. Homesteading on the electronic frontier. Reading, PA: Addison-Wesley Publishing.

Rosenfeld, Paul, Giacalone, Robert A., Knouse, Stephen B., Doherty, Linda M., Vicino, S. Mitchell, Kantor, John, and Greaves, Jean. (1991). Impression management , candor, and microcomputer-based organizational surveys: An individual differences approach. Computers in Human Behavior, 7, 23-32.

Shaw, David F. (1997). Gay men and computer communication: A discourse of sex and identity in Steven G. Jones (Ed) Virtual culture: Identity and Communication in Cybersociety. Thousand Oaks, CA: Sage Publications, Inc.

Sterling, B. (1990). Cyberspace (TM), Interzone, 41.

Steuer, Jonathan (1992). Defining virtual reality: Dimensions determining telepresence. Journal of Communications, 42(4),

Terenzini, Caroline (1997, February 6). User friendly: As computers become a workplace mainstay, employees try to add their own personal touches. Centre Daily Times, pp. 1C, 2C.

Tse, Alan C. B., Tse, Ka Chun, Yin, Chow Hoi, Ting, Choy Boon, Yi, Ko Wai, Yee, Kwan Pui, and Hong, Wing Chi (1995). Comparing two methods of sending out questionnaires: E-mail versus mail. Journal of the Market Research Society, 37(4), 441-446.

Turkle, Sherry (1995). Life on the screen. New York: Simon & Schuster.

Wiseman, F. (1972). Methodological bias in public opinion surveys. Public Opinion Quarterly, 36, 105-108.

----------------------------------------