THE INTERNET AS RESEARCH CONTEXT*

Annette N. Markham, Ph.D.
Department of Communication, University of Illinois at Chicago

This document is a draft of a forthcoming (2003) chapter in the book:
Qualitative Research Practice
Clive Seale, Jay Gubrium, Giampietro Gobo, & David Silverman, Eds.
London: Sage.

© Copyright Annette N. Markham 2002. All rights reserved.

Computer-mediated communication is not just a tool; it is at once technology, medium and engine of social relations.  It not only structures social relations, it is the space within which the relations occur and the tool that individuals use to enter that space.  Steven G. Jones, Cybersociety

What I know of the Internet is like filling a thimble full of water, and saying I hold the ocean in my hands.  Sheol, study participant

Conducting qualitative research in the unfamiliar field is challenging.  Via the Internet, the unfamiliarity of the field can be complicated by the fact that the field is negotiated versus geographic; interviewing is often, and perhaps preferably, an anonymous exchange of text messages.  On the Internet, using acronyms, odd spelling conventions, or referring to personae using pronouns like splat, h**, or spivak is equivalent to learning the language of the culture you’re visiting.  To make it even more strange, you may be sitting on your couch for much of the fieldwork, traveling to multiple cultural venues through your laptop, and interviewing text beings you’ll never see in the flesh.  I find it fascinating.  Others, I’m sure, would shake their heads and gladly trade computer and couch for pencil, notebook, two chairs, and good old fashioned conversation. 

The goal of this chapter is to illustrate several methodological quandaries, apparent in any research context but particularly highlighted by Internet contexts.  Using various excerpts from my research journals, I discuss my own encounters with study participants in the virtual field, primarily in interview settings, to exemplify these quandaries, provide a sense of my lived experience as both researcher and Internet user, and to display the significance of the researcher’s choices on the outcome of the ethnographically informed project.  To give the reader an idea of how users make sense of the Internet, as well as possibilities for studying in this swiftly growing area I also offer a general framework for how the Internet is conceptualized. 

>I ache everywhere.

I've been talking with people for hours,

but my jaw is sore because I haven't spoken.

I feel like my voice is migrating to the tips of my fingers,

which means my brain doesn't have to translate so much.

I suppose this is a good thing...

but I'm a bit troubled by what is happening to my conversational skills.

beginning to prefer fragments

and all lowercase

have trouble completing sente-

I dreamed that I was having a conversation with someone

and I could see the letters

--courier font, probably size 12--

streaming out of my gaping and silent mouth.

Sentences hung for a moment in the air and then wafted up to the ceiling,

..........where they dissolved. 

I resent my body's intrusion on my life online,

and my online life's impact on my body.

***

Sometimes I will take a deep breath and realize

I haven't been breathing......

then I'll feel giddy with the rushing intake of air.

Sometimes I blink,

and realize I must not have blinked in a long time, because it feels so good. 

Then I'll close my eyes for awhile,

enjoying the sensation of not staring

bug-eyed

at the glare of the computer screen. 

After a few hours online, my body is screaming with pain. 

If I don't chew gum, I'll clench my teeth. 

If I don't talk, my throat gets raw and sore. 

My hands take the most punishment.  They ache and throb

because I forget to stop typing.

So why can't I stop? 

I lose track of time, and four, five, six hours will pass

before I realize I haven't stood up, leaned back,

or even taken a sip of water. 

I haven't had hot coffee in months;

it's always cold before I remember to drink it.

My body hurts. 

Yet I'm actively participating in life online. 

I chat with people,

With all the windows open

In one, my mom and I get to be together

In another, I feel the breeze as I walk down the avenue,

and pick up a frisbee

to throw it to my interviewee, Beth Ann. 

In a third, I’m in a chatroom and just met dominOh!, whom I hope to interview.

In this window, I’m writing research notes.

At the same time that I'm engaging in these activities,

my eyeballs are drying out,

I don't speak a word,

and no part of my body is moving at all except my hands. 

In a way, I resent the

encroachment

of my body into my life here. 

This frightens me, sort of....

***

In 1995, I started meeting people who spent a lot of time interacting with others on the Internet.  I also started hearing stories in the popular media warning us about an emerging disorder called Internet Addiction.  In a conversation with a colleague, I learned that he spent almost half his waking hours interacting with others on the Internet. I couldn't imagine what would engage these people for so many hours of every single day.  As a researcher, I couldn’t resist the opportunity to talk with them.  I knew that if I really wanted to understand these people and their online lives, I would have to go online, too.  I figured out how to use different interaction programs, closed myself in my office, dialed up an internet connection, and got started.  What I thought would be a short term project using the Internet for structured and semi-structured interviewing turned into a two-year, exhausting, exhilarating exploration of the lives of a dozen individuals who lived online.  As any good ethnographer would, I had gone to live with them. 

The excerpt from my research journal above is representative of how I felt during my stint in this virtual field.  I spent much of my time sitting in an office chair focusing on a point about 18 inches from my nose.  Yet the point of focus rarely felt only 18 inches away; I was miles and worlds away, right here in a cozy online chat room, talking to a scientist from Sweden, another researcher in South America, or an American college student living somewhere between the Atlantic Ocean and the Pacific.

The stories told to me by the people I met formed the backbone of my book Life Online, and, to a large extent, my professional research life. The issues I confronted while talking with them as a researcher have become the issues I grapple with every day in the classes I teach and my continuing research. Since that first study, I have moved to a different office and learned to look away from the screen every once in a while to spare my vision, but the lessons I learned in conducting this research have never left me.

There is an elegant simplicity to the idea of studying Internet contexts as a social scientist: collecting, analyzing, and interpreting data to build theory and knowledge of this network of social potential.  But from the point of one who works  in this virtual field, the apparent simplicity is an  abstraction fraught with multilayered complexity and paradox as one faces the actuality of trying to know anything about the other, online.

What does it mean to interview someone for almost two hours before realizing s/he is not the gender the researcher thought s/he was?  Does that say more about the researcher or the participant, and how should an ethnographer account for this in both research design and practice? 

What do you do when 500 people respond to your request for interviews, when you imagined you would get about 20? 

How do you draw a sensible boundary around the ethnographic context when the culture you’re studying is a collective living in 55 different countries around the globe and the basic frame of the society is constituted by a network of email messages rather than persons generally located in the same physical space? 

How does the researcher extract him or herself from the culture under study when every word he or she types actively contributes to the ongoing formation of the cultural boundaries?

The dilemmas associated with doing Internet research often arise in the midst of a study, unanticipated and unaccounted for by even the most careful research design.  Inductive and explorative, the potential of the Internet as a tool or context for research is still emerging, particularly as technologies for interaction change.  As I illustrate the complexities of using text to interview, interpret, and represent others in the research report, I hope to encourage researchers to maintain close sensitivity to the context; constantly critiquing one’s own role in co-constructing the cultural spaces of inquiry, and mindfully attending to the premises guiding and shaping the interpretation and presentation of the object of analysis.

What do we mean by the term The Internet?

Now that the dust has settled in the frenetic race to explore, settle, colonize, buy, build, control, and study the Internet, we have the opportunity for calmer, more experienced reflection.  Many of us are still wondering what it all means.  Others of us have set our sights on what comes next.  Regardless of differences in methodological approaches and objects of analysis, most qualitative researchers who consider themselves part of the growing interdisciplinary field of Internet Studies agree that the concept we label The Internet is, both in practice and theory, a multiplicity of cultural phenomena not limited to either a monolithic entity or a universal set of experiences.  The term sustains itself through its ambiguity; surfers, netizens, consumers and researchers can and do interpret it freely, deriving and applying meaning of the concept in countless ways. 

We have a vast range of choices for delineating and studying the Internet.  Even as I argue that the term Internet is not easily encapsulated by a solid set of experiences or approaches, I offer a general framework describing how people experience the Internet as well as how we investigate it in the social sciences and humanities.  In an earlier work (Markham 1998) I posit the argument that users experience computer-mediated communication alternately or simultaneously as tool, place, or way of being.  Extended from computer-mediated communication to the Internet (however one defines it), this framework comprises a useful heuristic. 

One can usefully conceptualize the Internet as a tool for retrieving or transmitting information and connecting with others.  As a medium, the Internet can be seen as a research resource. In my own work, I view the Internet as an umbrella term for those social spaces constituted and mediated through computer-mediated interactions.  As such, the Internet can be seen as a place or a research context.  If one conceptualizes the Internet as a way of being, the focus shifts away from looking at the Internet as a tool or a cultural space and moves toward the ephemeral territory of exploring the ways individuals in a computer-mediated society construct and experience themselves and others because of or through Internet communication.  As everyday life becomes more and more inundated with communication technologies, it is appropriate to focus a critical, analytical gaze on how individuals construct and negotiate their lives in an information saturated environment. 

The Internet engages most users at multiple conceptual levels.  The extent to which we have mixed metaphors when talking about the Internet (highway, frontier, community, net, web) implies both a complex understanding and an effort to understand through a sensible if changing linguistic frame.  The fact that we shift terminology frequently may indicate a familiarity with jargon.  For instance, we talk about going there, doing research there, going to work there and so on.  Even when we use these descriptors, we may not visualize ourselves as being any specific place any more then when we are on the phone.  We commonly talk about being online or on the Internet, but we do not seem to mean this literally. 

Shifting terminology can, however, indicate shifting meanings and shifting experiences.  The change from conceptualizing the Internet as a tool to a place can emerge through the design of the interface, the level of engagement in the activity, the length of time we’ve been online, the depth of the involvement, and so forth.  The deeper distinctions between considering the Internet a place and a way of being seems to depend on the extent to which one integrates technology into one’s concept of being as well as one’s concept of social construction. 

Internet as Tool

The most common frame used to describe the Internet is that of tool.  As such, the Internet is network of electronic connections, a communication medium, a conduit that allows information to flow from one place to another.  Utilized in the framework of a tool, the Internet can extend one’s reach, expand the senses, and complicate traditional notions of time and space.  Whether we’re saving time by shopping online, spending time surfing the latest film reviews, collapsing physical distance to chat with a group of friends in three different countries, or increasing psychological distance by using email rather than walking across the hall, Internet communication is altering the fundamental processes by which we get things done. 

In very basic terms, a researcher can elect to study the tool itself, social interactions afforded by this tool, or use the tool to aid in the research project, all depending on the specific research project, the form of the research questions, the researcher’s epistemological stance, and the researcher’s methodological preferences.  Research in the past decade has delved deeply into the first and second areas; we have learned a great deal about how these connections are made possible, what types of interaction are possible, and what the effects are on individuals, groups, corporate processes and structures, and so forth.  The impact of the Internet in everyday life is both predictable and surprising as individuals, communities, and even nations adopt this technology and use it in their own creative ways (see Miller and Slater (2000) for an example of the Internet in Trinidad).

The third area is of more recent concern.  As Mann and Stewart (2000) note, we have only begun to imagine the ways computer mediated communication can be used to augment our traditional qualitative methods.  Mann and Stewart’s book, in fact, is arguably the first attempt to comprehensively lay out principles and practices for qualitative research using Internet communication as a resource.  This work notably shifts our focus from considering research of the Internet to research using the Internet as a tool. Instantaneous transmission, high speed connections, and inexpensive networks provide access to participants and cultural phenomenon beyond our local reach; software eases the difficulties of transcription and augments our capacity to access, sort, and code data.  Organizational and community artifacts are readily available for easy download, storage, and analysis.  Online interviews can be synchronous or not, designed and timed to satisfy the needs of both participants and researchers.  Observation in ethnographic settings can be less obtrusive.

Of course, as Marshall McLuhan, Neil Postman and other media skeptics note, all technologies come with a double edge.  The potential to enhance is accompanied by the potential to remove, disable, or diminish.  This fact should not stop one from using this tool, but it should be reason for reflection and flexibility.

For example, instantaneous worldwide networks provide a connection to the context or participant that is not physical but virtual; studying them at a distance, often via text based computer mediated communication, I no longer have access to many of the nonverbal cues I normally rely on during data collection in the field.  Although many text-based cues have been invented to simulate embodied nonverbals (Baym, 1995; Witmer & Katzman, 1995), for many users, these emoticons or verbal expressions may not completely satisfy one’s need for embodied knowledge.  In my own experience, the absence of the body doesn’t make the interaction less real, or the “knowing of other” impossible, but it forces an adjustment of perspective; I must be keenly aware of my preconceptions.  I must alter research design, strategies and tactics based on the fact that Internet communication removes previously assumed embodied cues which contribute vital information utilized in understanding what is meant by what is uttered.

Internet communication allows me to collect, archive, and analyze greater quantities of social texts. At the same time, it may also compel me to fall into the delicate trap of continuing to collect data simply because I can and not because I should.  There is a growing trend to collect larger and larger data sets for qualitative analysis of Internet communication.  While I am as yet unwilling to conclude that this is a good trend, I am quite willing to point out that trends can morph subtly into normative rules.  It is worth remembering that larger data sets can require more time and possibly the use of more data analysis aiding software solutions, neither of which may be optimal for all types of qualitative research projects.

Additionally, online interviewing diminishes the difficulties of scheduling and distance because we can meet our subjects in real-time or we can exchange messages over time, each from respective locales.  The removal of bodies from an interview changes the nature of the interaction from orality to textuality, which is not a minor shift.  In addition, there are other problematic issues encountered by researchers:  Interviewing in text requires different pacing. It also requires deliberate attention to providing written examples of conversational markers such smiling or laughing at a story being told, indicating confusion or subtly prompting more information without verbally interrupting the respondent.  Interviewing via text may be more suitable for people who type fast, who are accustomed to the medium, and, depending on the research question, whose personalities come through in the text as clearly as they would face to face.  Interviewing online requires constant rethinking of the definitions of the terms “real” and “authentic.”  It tests our notions of trust and brings new forms of stereotyping to the foreground, such as spelling ability, sentence construction, and depth of vocabulary, not to mention the tendency to assume by default that the other with whom we are interacting is white and educated (Kolko, 2000; Nakamura, 1995; Poster, 1998). 

Observing people’s behaviors in computer-mediated communication contexts is certainly less obtrusive, in that a researcher can lurk and not be noticed.  Participant observation is also easier, in that joining groups is not difficult.  Yet both these capacities afforded by the Internet must be balanced carefully with ethical considerations.  Some public groups perceive their interaction to be private and can be surprised and angered by intruding researchers (Bromseth, 2002).  Other groups know their communication is public but nonetheless do not want to be studied (Gajjala, 2002; Hudson & Bruckman, 2002).  Confidentiality is almost impossible to preserve with the sophistication of search engines (Mann, 2002).  Ethical issues have sparked much debate and disagreement among experts and remain a vital issue for each researcher to consider (see, e.g. the ethics committee of the Association of Internet Researchers and their associated statement (2002)).

As a tool, medium, or conduit, the Internet can be viewed as a portal through which we access and interact with information and other people.  For many users, the Internet is more than this; it also has dimension and meaning as a location for interaction.

Internet as Place

As the quote by Steve Jones at the beginning of this chapter articulates, the Internet is not only a conduit that facilitates the swift and planet-wide flow of information, it comprises the cultural spaces in which meaningful human interactions occur.  There, in a described, imagined, or perceived place, one can spend time wandering, navigating, and otherwise exploring.  One can converse, come to know and love, insult [flame], and otherwise interact with others one meets there.  Although computer-mediated social spaces have no literal physical substance, they can be perceived as having dimension, comprising meaningful, structured places where things happen that have genuine consequences.  In this frame, the Internet is not so much a prosthetic for the senses but a separate environment where the self can travel and exist.

Conceptualized as a place, the Internet becomes a research context, a socio-cultural milieu which can and should be studied in context.  Using basic terms, one can study the space itself, the interactions within these places, and the relationships and communities formed through the interactions. 

Just as the context is defined in multiple ways, boundaries of the culture are sketched not just by the preconfigured design or programmed parameters of interaction but by the interactions of participants.  Borders are thus negotiated processes (Hine 2000) rather than well-defined, static, or geographic.  Also, as I have noted previously (Markham 1998), the researcher’s engagement with people in these contexts influences directly the structure and border of culture quite significantly.  The researcher’s presence and influence presents problematic issues in all arenas of ethnographic inquiry but is accentuated by the highly negotiable feature of boundary in computer-mediated contexts.

As the field grows steadier in its sensibilities and approaches, we learn that computer-mediated environments are both like and unlike physical cultures:  These spaces of interaction can draw on or transcend traditional ways of being with others, reify traditional or create new stereotypes, democratize or marginalize.  These spaces, like the humans constituting and occupying them, are like any social space we see and study in physical environs; I argue that the primary distinctiveness of the Internet lies in the capacity for anonymity and the unique way this technology reconfigures time and space.  It has as much potential and limitation as our imagination.

Suffice it to say, the contexts of Internet-facilitated relationships, communities, and cultures are as multiple as the members.  Consequently, research of these spaces by anthropologists, sociologists, and scholars from multiple related disciplines is both vast and varied.  As the surge to capitalize on this new arena for research slows, our methods of approaching and studying these contexts will improve. 

Internet as Way of Being

The reality of the Internet as a social space is beginning to be taken for granted in this academic field of inquiry.  We also acknowledge and, for the most part, laud the idea that Internet communication can be a very influential mediator/moderator of human experiences.  Through the design, control, and play of information in online contexts, personalized worlds can be created, organized, and enacted.  Though the Internet is quite literally a network of computers, the outcome is a fuzzy mapping of imagined geographies, perceived physicalities, and transcendent forms.  As a means for reinscribing, reconfiguring, or otherwise shifting identity, body and self’s connection with other, the Internet becomes, for some, a way of being.

This third frame involves a more integrated sense of the Internet as a part of the self.  Within this frame, users may not focus on the technology used or occupied but rather on the expression and negotiation of self and other with or through Internet technologies.  Users who have integrated Internet technologies into their lives to a high degree can be seen to incorporate the Internet as a way of being.  Users might spend much of their time as computer-mediated beings, adopting alternate or additional personae in various text and graphic online environments, seeking transcendence from embodiment or a different embodiment, protection from embodied others, or an eventual merge of mind and body with machine.  On the other hand, there are those users whose embodied connection to the technology is powerfully evident, such as those who broadcast daily activities as public display via webcams or even those who feel best when the Internet is practically attached to the body via mobile communication technologies.

Interestingly though, while we might have a sense of “being there,” both the “being” and the “there,” Novak points out, are user-controlled variables.  Objects and bodies are but “collections of attributes . . . assembled for temporary use, only to be automatically dismantled again when their usefulness is over” (1991, p. 235).  In other words, we can create and destroy our various identities and selves at will in cyberspace; our identities can be perceived as having continuous malleability and transformative potential.  Of course, this potential is intertextual; Internet beings dialogically and recursively constitute each other simultaneously or alternately as author and audience, performer and stage, marking and marked.

From this perspective, computer-mediated communication is both process and product, medium and outcome.  Online identities and associated cultural contexts are multitudes of ever-evolving, self-referential sets of texts, influencing and being influenced by readers and writers and the individuals’ willingness to treat these texts and the associated social structure constructs as real.  Within this frame, the focus of research might be to reconsider and reconceptualize certain taken-for-granted aspects of being human with others, to explore the intersections of individual, technology, and identity, and to examine closely how Internet technologies are woven into a participant’s life experience.

The boundaries between these three frames of tool, place, and way of being are permeable, if not artificial.  One can conceptualize and experience the Internet as both tool and place, use the Internet as a tool while integrating it as a gestalt, or land in various categories depending on the time of day, type of technology, person with whom one is interacting, and any other number of factors.  Rather than a taxonomy, model, or theory, this framework is simply one way of making sense of experiences of the Internet as well as the complex growing field of Internet research. 

The remainder of the chapter describes in more depth specific issues one might face when moving from the theoretical to the practical, tactical engagement as a researcher within the cultural phenomenon of the Internet.  I focus here on my experiences interacting with users solely within Internet contexts to illustrate that within even this single area, many design and interpretive challenges must be confronted and negotiated.

Interviewing text-to-text:  Enabling and constraining features for researchers

How much of good conversation is based on reading the other person’s face?  How much of good storytelling relies on the nodding head, strategic pauses, chuckles, gasps, or raised eyebrows?

Online interviewing, at the most basic level, involves the exchange of texts.  Through this exchange, the qualitative researcher is hoping to glean something meaningful about organizational members’ lived experience, attempting to draw out examples, stories, or descriptions that will speak to the depths of experience, the meaning of relationships, and the understanding of identity.  Engaging others in conversation may always involve a great deal of patience, careful listening, and constant interpretation, but face to face, conversational knowing is, for the most part, taken for granted.  We humans, as a rule, trust our senses, particularly vision.  We generally believe in the universality of nonverbal communication and have developed an amazing faith in the naïve notion not only that our messages are understood perfectly by others, but that we truly know what others mean by what they say.  Here, I will provide some examples in my own research when my interactions with participants confronted me with the reality that conversation is an accomplishment, revealed many of my tendencies to categorize, stereotype, and otherwise encapsulate participants, and highlighted the dilemmas of filtering and editing the words and possibly the being of the participants of our studies.  These dilemmas are not restricted to Internet contexts; indeed, the examples I provide highlight the need in any study for ethical reflection on the choices one makes throughout the collection, analysis and presentation of those people we study.

Talking with participants:  The conscious accomplishment of conversation

Beth-ANN smiles

Markham nods understandingly

Beth-ANN says “I think I like it this way because I can just type what commes to mind and not have to think about it as much thinkgs seem to be communicated better through my fingers then my voice.”

I tried my best not to immediately type something back, because I had been running over Beth’s sentences constantly since we started the interview.  I just couldn’t stop myself; long pauses between messages often got interpreted as a dropped connection.  My participants and I frequently interjected the anxious question “Are you still there?” if the pause in our conversation grew longer than 15 seconds.

Sure enough, Beth eventually continued,

Beth-ANN says “that's why I like being on here so much.”

I asked,

Markham asks “do you think that talking with your fingers better than your voice is the major difference between RL and online communication?”  

And then, as an afterthought, I added,

Markham asks “for you, I mean?”

Beth wrote,

Beth-ANN says “I use the Internet for a lot of things even now to find information, chat, look for things just use it for everything but I haven't bought anything off the Internet yet.”

I wondered if Beth meant to type ‘bought’ instead of ‘brought.’  I typed, “What do you mean by brought off...”  I considered for a moment, and then erased the message.  My colleague Bill had just knocked on my office door and was feeling very scattered.  Better to buy myself some time, I thought, and wrote a different message.

Markham asks “Beth, can you hang on a minute while I use the restroom?”

Beth said,

Beth-ANN says “yes it is because I can type what i'm feeling better then I can voice my;m” 

A few seconds passed, then Beth continued,

Beth-ANN says “feelings it just comes a little easier seeing things to answer then hearing and having to answer I like to worrk with my hands alot.”

Hmmm.....good thing I hadn’t pressed on with the question of bought versus; as usual, I was racing ahead of Beth and she was plodding along, answering questions in the order I asked them. 

Beth-ANN exclaims “yes I can!”

Ah-ha.  She means, Yes, she’ll wait while I’m in the restroom.

I quickly typed,

Markham exclaims “thanks!  back in a flash!”

Beth-ANN says “ok taht’s cool”

I sighed with relief, leaning back in my chair to pay attention to the physical person who had just walked into my office.

“Hi, Bill.”  I motioned distractedly to a chair.  My colleague had dropped by to check on the progress of the interview.  We chatted for about a minute about the blessings and banes of multitasking when I noticed a message had appeared on the screen from Beth,

You sense that Beth-ANN is looking for you in Hut X

Beth pages, "is a girl who's interviewing me and it's on here that she's interviewing"

I quickly sent the message,

Markham says “hi again”

followed quickly by a smiley emoticon:

Markham :-)

Beth replied

Beth-ANN says “hi.  your back.  that’s cool.”

Beth-ANN smiles

Markham smiles back

Before email, instant messaging, and other forms of computer mediated communication became ubiquitous in my life (and the lives of everyone I know), I found these media awkward and unwieldy tools for conversation.  Interviewing in these media was even more challenging, in that it forced me to become aware of and monitor my own interaction tendencies.  Indeed, it might be better to simply say that interviewing via the Internet highlights the fact that interviewing, in general, is difficult.  Text based online interaction requires active reflection on and management of very basic elements of conversation, such as taking turns at the appropriate time, nodding, or mm-hmm-ing to imply, “Go on, I’m listening.”  Online, I couldn’t give a questioning glance or wrinkle my forehead or frown slightly to let the other person know I didn’t understand what they were getting at.  I couldn’t smile, chuckle, or laugh spontaneously.  Indeed, if I wanted to react (without interrupting the flow of the story) to something I found amusing, funny, striking, or in some other way noteworthy, I had to type something such as, “emote smiles” or “emote grimaces understandingly.”  Then a message would appear on their screen that read “markham smiles” or “markham grimaces understandingly.” 

Each time I felt compelled to react “nonverbally” to statements the participants made, I had to decide whether or not to risk disrupting their thoughts to let them know I was listening and was engaged in the conversation by verbally signifying a nonverbal behavior.  This issue became less troublesome as the interviews progressed and I became more adroit with this activity, but not less salient. 

Synchronous interviewing online took about twice as long as face to face.  I didn’t anticipate this and had to make many adjustments throughout the process to accommodate different participants writing style and speed.  For example, Beth would answer each question in order, constant and steady, even if I interrupted with other clarification questions.  Sheol, another participant, seemed to be as scattered as I, able to leap back and forth between topics in short phrases and disjunctive ideas. 

I wrote to Beth,

Markham asks “what do you do mostly when you are online?  Where do you go?”

Beth-ANN says “I'm usually on the MOO when I'm in my room.  But I go all over the place I have lots of bookmarks on my computer.

Markham asks “mostly the moo?  or do you irc too?”

Beth-ANN says “I just love to look aroud at everything and anything aplus my teacher my English professor likes us to search for things in class for projects and stuff she's an Internet junky too.”

Considering this an interesting label to give her professor, I asked,

Markham asks “What do you mean by ‘internet junky?’”

Beth continued her response,

Beth-ANN says “I love the Internet and my professor likes it that I like thae Internet because she says it's the wave of the future and there are not enough women on the Internet.  The  Internet is a place we can make the most impact”

The conversation seems disjunctive because our pacing was not synchronized.  I had to learn to slow down to give participants enough time to respond fully to the questions.  When I was interviewing Beth, I would ask a question and wait for what seemed like a long time for her to respond.  Sometimes, if I didn’t see writing on the screen shortly (don’t ask me how I define “shortly”… I’m sure the actual passage of time was much shorter than the multiple minutes I imagined), I would wonder if she had received the message.  Then I would wonder if she was still there.  Then, to make sure she was there, I would send the same message again or another message asking if she got the first one.  At other times, after Beth would send a message, I would ask the next question (a logical enough conversational move, I thought), and Beth’s response would be a continuation of her previous message. 

In effect, I interrupted almost every story she tried to tell.  She would be warming up to the question, getting started on an in-depth answer, and I would abruptly ask a different question.  I couldn’t help myself.  In my discomfort with what I thought was silence, I felt compelled to fill the blank void with more writing.  Meanwhile, Beth was chatting away . . . I just couldn’t see/hear it yet. 

Two considerations are important at this point:  Constant interruption of the participant’s talk can have a significant impact on the flow and content of ideas.  Yet interruption is a primary mode of interaction in text-based spaces and therefore is a necessary skill to be practiced by researchers.  The disjunctive, fragmented, and non-linearity of text-based computer mediated synchronous conversation no longer seems acutely strange for many users because they have grown accustomed to it.  As researchers using this form of communication, we must attend to the details, acknowledge the possibilities and limitations, and be both practiced and flexible to adapt to each situation as it arises. 

For example, because I continually interrupted Beth, I perceived her ideas to be cut off.  To solve this problem, I forced myself to focus on other things.  I started writing a research journal at the same time so that she had time to respond in peace.  Learning to be patient was crucial for another reason as well.  If I asked a question, got a complete response, and still remained silent, she would often fill the empty space with more depth regarding her previous answer or another related story that occurred to her during the silent period.  Most interviewers learn to do this in face-to-face contexts, but here it took concentrated effort to not type.  At the time of this interview, to cope with my own inability to manage this highly fragmented and non linear format, I developed the strategy of literally sitting on my hands, which gave Beth time to catch up with my questions and kept the conversation more sensible for me.  I wouldn’t call this a practiced art, just a desperate strategy at the time.  Still, it highlights the necessity working in a computer mediated environment to develop keen sensibilities about how the interaction works and practice strategies that prompts without interrupting talk in this space. 

Another example illustrates the importance of working with rather than against the disjunctive and text-only form of the interaction.  I realized that small talk was both extremely time consuming and essential to the conversations. Often, side conversations became intimately interwoven into the series of answers to questions I had constructed prior to the interview.  

All interviews have elements of small talk, comic relief, or subtle shifts in new conversational directions; these features allow the participant to relax, encourage a conversational mode, and allow the conversation to guide the questions rather than the other way around.  These elements are so natural in face-to-face contexts, they often remain unnoticed.  Particularly, the nonverbal features of interaction can be communicated without interrupting, acknowledged even as the recipient continues his or her statement. For the interviewer, these taken for granted elements of communication are unobtrusive yet vital means to encourage, prompt, shift, verify, validate, confirm, question, and so forth.  In a textual environment, however, even clarifying a participant’s response took time to write and several messages back and forth to complete.  In short, the basic elements of good conversation seem to steal precious time from what I had been taught was the heart of the interview, the set of protocol questions.  Flexible adaptation allows one to make better use the situation as well as the predetermined questions.  In my interview with Beth, dismissing the protocol and shifting to an unstructured interview was necessary to remain sensitive to the context.

In general, interviewing online took more planning, more time, and more self control than I thought it would.  There is, however, a significant tactical advantage of online interviewing:  Time to devise and deploy dynamic, context-driven follow-up questions.  In the act of interviewing online, I could see the story unfolding and the response developing textually as the participant sent message segments.  This meant I could attend to the message more than once.  I could re-read what the participant had just sent, and while she was composing her next message, I could discern appropriate follow-up questions. 

This is both a tactical and strategic improvement to interview techniques.  The process becomes more granular in nature, observable in time steps.  Not only do I have more time to consider the direction of the discussion, or scroll back to previous comments, I can adjust the form of the question mid-utterance.  For example, with Sheol I could ask a closed-ended follow-up question, “Are you an addict?” or, after seeing it in text, consider carefully what I really wanted to know, and edit my own text to ask an open-ended question that allowed Sheol to set the parameters: “What would define ‘addict’?”  In another instance, I started with the question, “Why did you start using the Internet?” and changed it to, “What drew you to the Net?” because even as I was writing the question, what I was reading in Sheol’s answer scrolling up on my screen led me to ask the question in what I perceived would be a more provocative manner.  These actual questions vary only slightly, but as any interviewer knows, the form of the question is vital for facilitating a participant’s response. 

Having response and interaction lag time is beneficial for another reason as well:  It provides valuable processing time for consciously formulating overall interview direction, depth, and flow.  Within the synchronous time frame of an interview, I could pretend I had been interrupted by the phone or other people (which I learned by actually being interrupted by the phone or people who didn’t realize I was conducting an interview while sitting at my keyboard).  Contrived interruptions gave me time to ponder and reapproach issues that were emerging.

Considering authenticity:  How much do one’s texts represent the real person?

At a recent conference, a journal editor indicated that he believed I should interview participants offline as well as online to get a “more holistic picture of who they really are.”  This is not the first time I have delved into the debate that surrounds the issue of authenticity and whether or not one must devise methods that will establish the real identity of the participant, a process which, for many researchers, requires offline interviewing.  The debate raises fundamental questions about knowing, truth, and our beliefs about reality.  The question often asked about participants in online contexts is “Who are they, really?”  By this, one often means, who are they, as I can see, verify, and know them in a body?  Asked from a slightly different perspective, one might ask: How much do we rely on our bodies and the bodies of participants to establish presence and know other?  Is this reliance warranted or desirable?  Will our picture of other, in person, make our understanding of them more whole?  More directly:  Does the embodiment of a participant gauge their authenticity?

The answer is, as with most of life’s important questions, “it depends.”  It depends not only on the question one is seeking to address but also on the researcher’s underlying epistemological assumptions.  If one is simply using the Internet to expand one’s reach to participants and interviewing them online is merely a convenience, one may want to seriously consider the extent to which people can and do express themselves well, truly, or fully in text.  But if one is studying Internet contexts as cultural formations or social interaction in computer mediated communication contexts, the inclusion of embodied ways of knowing may be unwarranted and even counterproductive. 

Let’s take my research participant named Sheol.  Online, Sheol’s texts were eloquent, albeit horribly misspelled, full of poetic expression and a genuine love of life and Shakespearean language.  Conversing with Sheol was a delicate balance of interviewing and flirtatious play with language.  I have no way of knowing if he would have engaged in these same behaviors offline.  At the time, I insisted that I should not interview any of my participants offline; my goal in that project was not to compare but simply interpret the richness of experience.  Even if my goal were comparison of online and offline contexts of performance and interaction, I am unwilling to conclude that interviewing both on and offline would satisfactorily accomplish the goal. 

Christine Hine argues we should question seriously our desire to seek authenticity in these contexts, as any degree of authenticity is negotiated and situated: “A search for truly authentic knowledge about people or phenomena is doomed to be ultimately irresolvable” (p. 49).  Still, researchers are plagued by questions of authenticity; we desire trust in our interviews with participants, often believing that if we know each other truly, in some authentic sense, we will know who they really are.

As qualitative researchers, we have neither fully explained nor adequately examined the function of text, practically and aesthetically, in the performance of self, perception of self and other, and sustenance of relationship.  Does the form of the text matter?  To what extent does an online persona’s text represent the embodied being?  Is the text merely a tool for interaction?

My overall goal as a researcher has been to utilize symbolic interactionist principles and interpretive ethnography practices to analyze and build knowledge of how people interact and negotiate identity and culture via the Internet.  Of paramount concern during the research project is the design, exchange, and interpretation of text messages, for this is where the building blocks of culture reside.  Online, culture is literally constructed discursively.  Sensemaking is wrapped up in the text more obviously than in physical spaces because other mediating factors are perceived as absent.

Yet, even as I emphasize text and truly believe in the totality of intertextuality as the means by which we construct, maintain, and resist social structures, my training as a cultural researcher is embedded in embodied theory and practice.  I am socialized to rely on and privilege the five senses, as most social scientists are.  Having grown up in the wilds of Idaho, knowledge through the body is sometimes a matter of survival.  Natural Biologist Diane Ackerman describes vividly this reliance on traditional bodied senses as the means for knowledge:

We live on the leash of our senses.  Although they enlarge us, they also limit and restrain   us, but how beautifully. . . .  To understand, we have to use our heads, meaning our minds. . . .  Most people think of the mind as being located in the head, but findings suggest that the mind doesn't really dwell in the brain but travels the whole body on caravans of hormone and enzyme, busily making sense of the compound wonders we catalogue as touch, taste, smell, hearing, vision. (2-5, passim)

Should one interview Internet users in person as well as in text so one can know what the other senses say they are?  Would this make them more authentic or make my study more valid?  These questions are both instinctive and pragmatic.  To dismiss them as mere logistical issues, however, is to overlook the ethical consequences of this fundamental research design decision.  Although there are no easy answers, the process demands ongoing reflexive consideration of the issues within the context of each individual research project.

When  I  am  online,

I restrict some of the senses that would typically help me make sense

of   this   place,   Other,     and   the   context.

I   cannot   see   the   body   of   others, 

but   I   can   get           

a   verisimilitude   of   them  

as   they   describe   themselves   to   me.               

Of   course,   all   I   have  

is    their   description   of   themselves.                       

I   must,   therefore,   trust  their   vision    of   themselves,

rather   than   using   my   own   sensibilities              

and stereotypes

to   interpret   what   they   look   like                     

....to me.                        

...but I'm still using my stereotypes and presuppositions and experiences to filter them into something I can recognize.....                                          

This   is   only   fair.                            

After   all,   they chose this   context

to   mingle,   and   chose   word   play

as   a   way   for   Other   to   know   the   Self.                 

Yet....                                                                  

....I  also   cannot   hear   their   voices  

or   the   sounds   of   the   worlds   where  they  sit.              

It's   a   lonely   place   for   my   ears.....                                    

I   only   hear   the   sound   of

my   own   fingers                        

tapping                

the     

keyboard....           

...or  the  humming  of  the  air  conditioner  in  the  background.                 

I  can't smell or taste the air they must be breathing

as   we   interact.                                              

Does   this   really   make   a   difference?                       

             ....Maybe.                                           

This is not a new problem, of course.  But in Internet ethnography, the problem is salient because the existence of both the researcher and the researched within the research context is solely text; a fact that emphasizes the already complicated features of "being with," not to mention the problem of being compelled, by our disciplines, to verify, validate, or otherwise authenticate one’s sensemaking of the participants, of the perceived place, and of the shared context.  One might reply that this is impossible in any study, and I would agree.  This doesn’t completely remove the trained desire to distinguish between the authentic and the apparent and to find consistency and reality using one’s eyes and ears as well as one’s mind. 

Considering textuality:  How much does the form of one’s texts matter?

To give this abstraction more shape, let me offer two examples.  One is the use of poetic form by academic authors.  In a chapter of a handbook on qualitative research methods, for example, how would a typical reader make sense of an author who used poetry to make claims about matters of social and scientific importance?  How would the use of poetic forms or fonts or unconventional grammar and punctuation influence your perception of the author, the supposed scholar?  If the text is primarily comprised of traditional scholarly texts, the poetic chapter would seem out of context, perhaps even dismissible at an initial glance.  If the reader is trying to move beyond the content of the chapter to understand something about the author, paralinguistic cues provide vital data for the reader.  One could glean a great deal of information from the length of line, use of tense, word clusters, degree of formality, active versus passive voice, and flow of thought created through particular punctuation choices. 

In this framework, the complexity of understanding other is in some ways inseparable from the texts presented.  On one hand, we might consider the exercise one of futility as the reader may not be able to discern the degree to which language use is deliberate or unintentional, and therefore to what extent it represents the author’s intended, true, or perceived sense of self.  On the other hand, whether intentional or accidental, communicative behaviors reveal important aspects of a person’s identity, which makes the interpretive exercise quite fruitful.  If there could be a third hand, I would mention the underpinning, consistent fact:  The interpretation will always be the reader’s.  Moreover, the analysis may represent the reader as much as or more than the intended object of the reader’s gaze.

A second example illustrates the elusiveness of the subject as a concrete knowable entity and highlights key complications inherent in any interpretive activity, online or offline. 

Sheol is a self described “heavy user” of the Internet and a “budding hacker” interviewed in an anonymous text based synchronous chat room.  Sheol’s interview is marked by frequent and intense inclusion of emoticons and punctuation to accent the content, such as LOL, exclamation marks!!!, and smiley faces :-) 

<Sheol>     *LOL* This is way cool!! I have never  been asked for an interview before:)

<Sheol>     I am intrested in talking to:)  Could you be more spesific about what questions you will ask?  Just let me know when you want to talk, and I will try to accomidate! :)

<Sheol>  On the net you can be who or what ever you want to be.  That is the trap! when you want your cyberlife to be your real life.  That’s what hapened to me. 

<Sheol>  I became a very popular (I know that sounds conseeded) figuar on the line I called home.  I am ruled by the right side of my brain so I liked the diea of being that personality.

<Sheol>     My cyberfriends and I liked to roleplay ... we went on fantastic adventurs over the net.  The only limit was our imagineations.  Not anything like in the real world!!  I am shy by nature...I am also a big fan of Shahspear langue. I can use that style of speaking, and not be shy about on the net:)

Of immediate interest and concern to me as a qualitative researcher conducting close textual analysis is the form that envelops the content.  I could elect to bracket or set aside the form and focus only on the content.  However, the goal of my research in this case is to explore how people experience the Internet and how their identities are presented and negotiated.  To ignore the form in this interview could be seen as a poor choice, given the well founded premise that nonverbal behaviors function discursively in the presentation of self, negotiation of identity, and eventual symbolic construction of culture.  Yet, multiple dilemmas can plague the Internet researcher:  How much does text represent the reality of the person?  Put more personally, how much would I want to be bound by what I wrote at any particular time?  How might the findings shift if I focus on form versus content?  How much are my own preconceptions and stereotypes influencing how I make sense of this data?

One of the first problems impeding the interpretation of this interview is that I do not know Sheol in culture.  I am interviewing from the outside, therefore it remains difficult to assess the intention of Sheol’s use of graphic accents.   It is also difficult to assess the meaning in any accurate sense, because the interaction is abstracted from the typical context of Sheol’s online existence.  More directly, Sheol is not participating in online culture, Sheol is participating in an online interview. 

Should I base any of my interpretation on Sheol’s grammar?

As the researcher, I have numerous choices.  I have many more choices than Sheol does, in terms of creating cultural knowledge about how people interact in cyberspace.   Reflection on each research decision as well as the premises undergirding the choices I make is crucial if I am to preserve Sheol’s dignity as a human being and his autonomy as a human subject in my research.  The interview yields fruitful insight about Sheol’s discursive practices and sensemaking practices.  Throughout our interactions, Sheol appeared unconcerned with how the writing appeared and unaware of how the construction of text might mediate identity for others.  Although Sheol mentions spelling once, Sheol never tried to change it or correct errors. 

I, on the other hand, could not ignore Sheol’s presentation of self through the text, both content and form.  I am a creature of my upbringing; the number of social labels I mentally attached to Sheol during our interviews probably came close to the number of spelling errors I found—a number that was considerable.  This is not a tangential point.  It illustrates a potentially irresistible tendency to leap to conclusions and make hasty judgments about people.  The interpretive lens is not separable from the researcher’s frame of reference and history, but researchers often deemphasize or totally ignore this limitation under the protective guise of scientific tradition. 

In order to preserve the integrity of the online interview project, it is essential that the researcher address these issues.  For example, for two hours of the interview, Sheol was female (stereotypical gendered language style was very evident in tags, qualifiers, expressions of emotion, and heavy use of graphic accents).  Sheol was young (spelling was phonetic, attention to language misuse was not at all evident).  Sheol was perhaps not very intelligent (multiple spelling errors, unreadable messages, apparent lack of ability to be a real hacker).  Sheol was, of course, Caucasian (default characteristic because of mainstream cultural assumptions about use of the Internet as well as the tendency to make the online other look more like the self).  Additionally and solely based on the interpreter’s frame of reference, Sheol was heterosexual, middle class, and American.

It is essential to expose these assumptive interpretations to detailed examination.  In this case, each interpretation I was making about the subject based on the text had to be reviewed.  Conclusions could not be driven by my own social and academic conditioning.  Even so, one must acknowledge that caution may improve and justify interpretive decisions in some cases but not others.  As much as one understands the qualitative researcher’s right and responsibility to build knowledge through interpretation, this example highlights the extent to which the very existence of the online persona being studied is comprised solely by the pixels on a computer screen.  Therefore, the choices we make to attend to, ignore, or edit these pixels has real consequences for the persons whose manifestations are being altered beyond and outside their control.  If a subject types solely in lowercase and uses peculiar spelling, the researcher’s correction of grammar may inappropriately ignore and thus misrepresent a participant’s deliberate presentation of self.  If someone spells atroshiously or uniQueLY and the researcher corrects it in the research report for readability, alteration of a person’s desired online identity may be the price of smooth reading.

On the other hand, a participant’s exclusive use of lowercase may be simply a time saving device.  A new keyboard, carpel tunnel syndrome, broken finger or a project due tomorrow for the boss may prompt typographic errors that the participant ordinarily would avoid.  Our interpretation of certain data as meaningful or dismissal of other data as meaningless may be well founded or absolutely unwarranted depending on any number of underlying factors, only some of which are comprehensible.  The methodological dilemma is to be sensitive to the context, to figure out what the most suitable interpretive path is, and to remain epistemologically consistent. Of course, in my own experience, this is easier said than done, as the following segment illustrates.

Considering anonymity and identity:  How reliable is a shifting subject?

Complicating the issue of authenticity, the online persona may be much more fluid and changeable than we imagine.  Anonymity in text based environments gives one more choices and control in the presentation of self, whether or not the presentation is perceived as intended. 

The best example of this developed quite unexpectedly in an online course I was teaching.  We had met online for six weeks, never meeting face to face, as the participants were both local and distant.  We had met in various online environments to assess the impact of each technology on our participation in class as well as the development of individual identity and overall sense of community.  One night my students and I met in Internet Relay Chat (IRC), a synchronous anonymous chatting environment.  At the request of the students leading class discussion, we adopted colors as our names.  I thought I would be satisfied with "Forest Green," but I got bored, and switched it.  As I changed my “nick,” this message appeared on everyone’s screens:

*** Forest green is now known as   "GhostlyGreen"                

For me, GhostlyGreen was satisfactory for awhile.  After all, it was very close to Halloween.  But I was feeling playful--finally, I could experience a classroom environment in which I was not immediately identified and characterized as Annette or Dr. Markham.

*** GhostlyGreen is now known as babypuke

Much better.  I acted out my "color identity"--- made rude comments, interrupted other participants, and such.  Still, I thought, rather gross.  It wasn’t really “me.”  I continued my spectrum of development.

*** babypuke is now known as "RottenJackOrange"   

This still did not quite feel right, and I was in an obnoxious student mood, so I shifted my nickname again.

*** RottenJackOrange is now known as oatmeal”

I oozed and squelched while the rest of the class attempted to carry on a scholarly conversation.  Occasionally they would get into the playful mood and walk around me, get their shoes stuck in my porridge-ness.  One student threatened another, using me as the potential weapon.  We all had a good laugh about that, which disrupted the class even more.  Finally and wisely, the students running class discussion decided it was time to reveal the actual identities behind the colors. 

As I watched various students reveal themselves, I saw IndigoBlu turn to AMarkham.

*** IndigoBlu is now known as AMarkham

I had never chosen IndigoBlu as my color identity.  I thought to myself, someone’s playing a good game.  So I went along with it and after all the other students had presumably revealed their actual names, I unmasked as if I were the only unnamed student remaining.

 ***oatmeal is now known as DennisL. 

For the remainder of the class, almost two hours, the rest of the students believed he was the professor and responded to him as if he were me.  I played the role of student.  They believed I was a student.

What does this example tell us?  Importantly, this example reiterates the symbolic interactionist concept that identity is negotiated (Blumer, 1969).  It seems to support the idea that we all wear masks and adopt roles that eventually come to represent our authentic self (Goffman, 1959).  This example can also illustrate Turkle’s notion that we have fragmented selves in Internet contexts (1995).  Basically, it tells me that titles can mean everything or nothing in the moment of the interaction, implying a negotiation of reality based on discourse as much as perception of one’s title.  Anonymous Internet based interactions facilitate knowledge of self and other that is interwoven with naming, perception, and yet fundamentally grounded in the exchange of texts. 

Authenticity, in this case, is found as much in the perception of participants as in the body/title attached to the name.  Richard MacKinnon (1995, p. 119) aptly points out that in cyberspace, the phrase, “I think, therefore I am” is woefully inadequate.  It is not even enough to say “I type, therefore I am.”  In cyberspace, the more appropriate phrase is “I am perceived, therefore I am.”  MacKinnon means this in a literal sense, where one’s texts must be both seen and acknowledged for one’s existence to be meaningful in context.  Read in a different sense, perception is the defining point for reality, making authenticity either a label defined by other or at most, negotiated by the participants in context.  This notion has significant methodological impact.

I return to the question I hear in so many classes and conferences, when students want to study the Internet as social space but feel uncomfortable about the issue of authenticity, embodiment, and reality:  “Does it matter that people may be different than who they appear to be?”  Given the example above and the multiplicity of experience in anonymous Internet facilitated environments, this is not the most useful question.  The flexible negotiation of identity in a text-based social space seems to require a different set of questions for the ethically sensitive researcher:

•   “As researchers and members of various communities and cultures, what do we use to construct a sense of who the Other really is?"

•    “In what ways do our methods of comprehending online others either disavow or validate multiplicitious, polyvocal, ever shifting constructions of identity?”

•   “To what extent to we acknowledge our own participation in the construction of the subject of inquiry?”

As pioneers on the research frontier of qualitative Internet studies, we must continually address what I call the embodiment dysfunction.  When we rely on our embodied sensibilities of knowing, we are not necessarily getting a better or more “accurate” picture of the subjects of our studies; we may be simply reflecting our own comfort zones of research.  Critical reflection on the product of our gaze can reveal some of these comfort zones for introspection and interrogation.

Considering the shape I give them:  Why do they all look like me?

To what extent do we transform the subjects of our research into images of ourselves?  With what effect?  Every time I study online social interactions and conduct interviews online, I visualize the participants.  I always give them a body.  Before I became very reflexive about my practices and premises, my imagination wasn’t that great; they all ended up looking like friends of mine. 

In my first study using exclusively online interviews, Beth Ann and Sheol were very young, based on their typing and spelling ability.  Melissa was in her 40s because she spoke clearly, seemed comfortable with herself, and was so sensible in her attitude toward the Internet.  Matthew was old, or in any case, older, because he told me he had children.  Teri was short, dark haired, wore lots of either black or pink and had a New York accent because she typed fast and I knew she was attending school there.  In my mind’s eye, Beth Ann had straight brown hair that extended a little past her shoulders.  She had translucent creamy white skin and downcast eyes.  Sheol never combed his hair.  They were all some variation of Caucasian.  They were all middle-class.

When I teach in distance learning environments, my students seem white to me.  I think they speak middle-America English with very little accent if they have an anonymous name.  If they use the name that corresponds with their enrolled identity, I adjust my image based on what I think the name represents.  I didn't realize I was making these judgments until my editor pointed out the absence of discussion about absence of race or socioeconomic markers for self identification.  It was not the focus of my study so I had not asked about race.  Participants never mentioned it.  I had been so preoccupied with my research focus and their presentation of self through their texts, yet I had been equally unreflexive of my own filters of interpretation.  This is typical of the way many people make sense of another who is known only via the text.  This fact illustrates how much we rely on and use our own parameters to categories others into something we can comfortably address.  Scholarly discussion of race and the Internet, particularly of how the Internet has been created and perceived naively as a raceless space is growing (Kendall, 1998; Kolko, Nakamura, and Rodman, 2000; Poster, 1998).  These discussions will help researchers better reflect on the spaces studied as well as the assumptions made during the collection and interpretation phases of the project.

Ironically, although I have always been aware of and critical of others' perceptions of me and my physical embodiment, I rarely reflected on this when I began studying Internet culture.  I assumed others would perceive me as I constructed myself via the text.  At times, I used sentence structures and terminology that I thought would mark me a scholar.  Other times, I modified my voice to give the impression that I was young, street wise, hacker-friendly.   I did this glibly and without guile.  I knew that I was engaging in impression management, but isn't that what we do all the time anyway?  If it loosens up the participant and helps them tell good stories, then I believed I had been a productive researcher.

Mediation of the self via the computer becomes a problematic proposition. As a middle class white female, I am privileged to have relative invisibility in most physical contexts.  Still, internet technology beguiles one into thinking that one has a high degree of control over the presentation of one’s own self.  More succinctly, in my early forays in online research, I assumed that everyone I met would interpret me as I chose to convey myself.  A naïve and somewhat ironic misperception since one finding of my study was that one’s identity for others is not a user controlled variable, but a negotiation.

It didn’t occur to me I would appear to them differently than I appear to myself when I look in the mirror every morning.  If I had stopped to consider the image of me looking through the screen and seeing someone else, I might have wondered at their possible perceptions of me.  But to many users, the Internet as a medium for being with others is under the user’s control.  Users believe they can see and know others, but remain comfortable in the perception that they can control the way others see them.  Another way to look at this is to consider that for the most part, users believe they see others truly and also believe they control what others see/know of themselves, whereas actually, both parties see only the words of the other, superimposed by their own reflections on the screen.  Hence, while early discussions of Internet contexts lauded the capacity of ultimate freedom in the presentation of self (Benedikt, 1991), more recent thought acknowledges that all presentation is a much more complicated negotiation between people whose interpretive frameworks are fraught with their own preconceptions.

The Internet provides a unique space for the construction of identity in that it offers anonymity and an exclusively discursive environment.  The difficulty of interviewing in this space is that our expectations remain rooted in embodied ways of collecting, analyzing and interpreting information.  Our methods are still more suitable for research in physically proximal contexts.  Moreover, although the technology of the internet has afforded us greater reach to participants and provided a space for researchers to interact with participants in creative ways, our epistemological frameworks don’t yet fit.  It is necessary not only to accommodate the features of computer-mediated communication into our basic assumptions, but also to rework the very premises we use to make sense of the world.

I hope to have complicated the simple and elegant project of studying identity formation or knowing and presenting the other in Internet contexts.  I have many questions but few answers.  In the end, however, our goal as qualitative researchers remains; we strive to understand other in context, analyze some of what it means, and, when we think we know something, present this knowledge to colleagues.  Rommetveit (1980) reminds us that what we know is more often a shared faith rather than any actual understanding of what is meant by what is said.

In whatever ways we utilize the potential of Internet-mediated communication to facilitate our social inquiry; as a tool, a place, or a way of being, ethically sensitive approaches are complicated, even impeded, by methods.  Depending on the academic discipline we find ourselves working within, we will be encouraged to varying degrees to oversimplify the complexity of human experience, transforming the mysteries of life into discrete variables that are easily measured.  This is done for admirable reason and by no means am I recommending a complete dismissal of traditional means of collecting and analyzing data.  At the same time, Internet contexts prompt us to reconsider the foundations of our methods and compel us to assess the extent to which our methods are measuring what we think they are, or getting at what we have always assumed they did. This is not an inconsequential point:  Through the Internet, identities, relationships, and social structures can be constituted solely through the exchange of texts.  This is unique in that we have the opportunity to observe how written discourse functions to construct meaning and how textual dialogue can form the basis of cultural understanding.  The taken for granted methods we use to make sense of participants in our research projects may need to be thoroughly reexamined in light of our growing comprehension of how intertextuality happens, literally. 

How can this be accomplished?  Give careful reflection to the outcome of interpretation and critical examination of the extent to which the interpretation reflect one’s own biases versus the experiences of the participants.  Even within a contemporary framework of sociological inquiry, whereby the distinction between the researcher and researched is problematized, the researcher’s role is acknowledged, and bias is accepted as a fundamental fact of interpretation, our obligation to the participant remains.  Here, I don’t offer a set of prescriptives.  Rather, I raise essential questions and reflect on my own practices, which may help researchers see the utility of such questioning in their own works.

Works Cited

Ackerman, D. (1991). The natural history of the senses. New York: Random House.

Association of Internet Researchers (2002).  Ethical decision making and internet research.  Retrieved December 1, 2002 from http://www.aoir.org/reports/ethics.pdf

Baym, N. (1998).  The emergence of on-line community. In S. Jones (Ed.).  Cybersociety 2.0 (pp. 138-163).  Thousand Oaks, CA: Sage Publications.

Benedikt, M. (1991).  Cyberspace:  First steps. Cambridge: MIT Press.

Blumer, H. (1969).  Symbolic interactionism.  Engelwood Cliffs, NJ: Prentice-Hall, Inc.

Bromseth, J. (2002).  Public Places…Private Activities?  In Morrison, A. (Ed.) Researching ICTs in Context (pp. 44-72).  Oslo:  Intermedia Report 3/2002. Oslo: Unipub forlag.  Retrieved December 1, 2002, from http://www.intermedia.uio.no/publikasjoner/rapport_3/

Buber, M. (1958).  I and Thou.  2nd ed. Translated by R. G. Smith.  New York:  Scribner.

Gajjala, R (July 2002) An Interrupted Postcolonial/Feminist Cyberethnography: Complicity and Resistance in the "Cyberfield" - Feminist Media Studies, 2(2), 177-193.

Goffman, E. (1959).  The presentation of self in everyday life.  New York:  Anchor Press.

Hine, C. (2000). Virtual ethnography.  London: Sage Publications.

Hudson, J. M. & Bruckman, A. (2002).  IRC Francais:  The creation of an Internet based SLA community.  Computer Assisted Language Learning, 15(2), 109-134.

Jones, S. G. (1995). Understanding community in the information age. In S. G. Jones (Ed.), Cybersociety:  Computer-mediated communication and community (pp. 10-35). Thousand Oaks, CA: Sage.

Kendall, L. (1998). Meaning and Identity in 'Cyberspace': The Performance of Gender, Class and Race Online. Symbolic Interaction, 21(2), 129-153.

Kolko, B.E. (2000). Erasing @race: Going White in the (Inter)Face. In B.E. Kolko, L. Nakamura, & G.B. Rodman (Eds.) Race in Cyberspace. (pp. 117-131). New York: Routledge.

Kolko, B. E., Nakamura, L. & Rodman, G. B., Eds. (2000.) Race in Cyberspace.  New York: Routledge.

MacKinnon, R. C.  (1995). Searching for the Leviathan in usenet.  In S. G. Jones (Ed.), Cybersociety:  Computer-mediated communication and community (pp. 112-137). Thousand Oaks, CA: Sage.

Mann, C. & Stewart, F. (2000). Internet communication and qualitative research:  A handbook for researching online.  London:  Sage.

Mann, C. (June 1, 2002).  Generating data online: Ethical concerns and challenges for the C21 researcher.  [Keynote address].  Nordic conference on internet research ethics.  Trondheim, Norway.

Markham, A. (1998).  Life online: Researching real experience in virtual space. Walnut Creek, CA: Altamira Press.

Miller, D. & Slater, D. (2000).  The Internet:  An ethnographic approach.  New York:  New York University Press.

Nakamura, L. (1995). Race In/For Cyberspace: Identity Tourism and Racial Passing on the Internet. Works and Days, 13(1-2), 181-193.

Novak, M. (1991).  Liquid architectures in Cyberspace.  In M. Benedikt (Ed.), Cyberspace:  First steps (pp. 225-254).  Cambridge, MA:  MIT Press.

Poster, M. (1998). Virtual Ethnicity: Tribal Identity in an Age of Global Communications. In S. G. Jones (Ed.), Cybersociety 2.0 (pp.184-211). Thousand Oaks, CA: Sage.

Rommetveit, R. (1980).  On ‘meanings’ of acts and what is meant and made known by what is said in a pluralistic social world.  In M. Brenner (Ed.), The structure of action  (pp. 108-149).  Oxford:  Basil Blackwell.

Turkle, S. (1995). Life on the screen:  identity in the age of the Internet. New York: Simon and Schuster.

Witmer, D. F., & Katzman, S. L. (1998).  Smile when you say that:  Graphic accents as gender markers in computer-mediated communication.  In S. Rafaeli, F. Sudweeks, & M. L. McLaughlin (Eds.), Network and Netplay:  Virtual Groups on the Internet.  Boston:  AAAI/MIT Press.