The Real Me Online:
Identity Play on the Internet

Dr Andrea Chester
RMIT University, Australia

 

Abstract

This paper examines the widely held belief that people experiment with identity online. Through two studies of the behaviour of people on the internet, the author finds that the intention to present an “idealised self” online is relatively rare and most people have a strong motivation to present themselves honestly online. Identity play, as it had been described in the early literature, is relatively uncommon.

Introduction

Perhaps the best known example of sustained identity play or deception online was described by Lindsy van Gelder (1991), a journalist with Ms Magazine. In her article “The case of the electronic lover”, van Gelder told the story of Alex, an American psychiatrist in his 50s, who in the early 1980s joined a chat room using the screen name “Shrink Inc.”. Although used to dealing with personal issues in his professional practice, van Gelder was reportedly surprised by the vulnerability expressed by women online and the intimacy he was able to achieve. He realized that his gender-neutral handle had been misinterpreted and these women had assumed he was female. Excited by this new potential for helping people as a woman, Alex established a female character, Joan Greene. Joan emerged oi the chat room with a carefully wrought history that provided the perfect excuse for avoiding face-to-face meetings. Her boyfriend had been killed in a car accident in which she herself was paralysed, disfigured, and had lost her ability to speak. Not only did she find it difficult to get around, but also the facial disfigurement caused by the accident left her embarrassed. Face-to-face meetings were therefore physically and emotionally difficult.

Online, however, Joan Greene flourished. Over the course of several years she underwent a gradual transformation, from suicidal recluse to socially confident woman. Her journey was inspiring for others. Over a period of three years she set up a women’s discussion group and encouraged women to work on a range of psychological problems. Joan was bisexual and engaged in netsex with several women.

Finally, according to van Gelder, Alex attempted to kill Joan off. He began this process by hospitalising her with a life-threatening illness. However, the huge outpouring of grief and concern led Alex to organize her recovery.

Although few people engage in deception to the extent played out by Alex, most of us have experienced the liberating potential of the Internet. Online it is not only possible, but also downright tempting to conceal traits we dislike or exaggerate characteristics we desire. After all, who will know if we are a little taller online, a few years younger, or our shoulders a little broader than in real life?

Since the early 199Os, the idea that people play with their identity online has been widely accepted (e.g., Reid, 1994; Turkle, 1995; Waskul & Douglass, 1997). This experimentation can take several forms, including changing physical features. The most obvious example of this is gender swapping, adopting a gender online that is inconsistent with one’s biological sex. Users can also play with identity by taking on desired personality traits and acting out “hoped for selves”. As the now famous New Yorker cartoon illustrates, “On the Internet, nobody knows you’re a dog”. No one needs to present themselves honestly online.  Users are limited only by their imagination. We can present multiple, different selves, recreating identity in different contexts. Unconstrained by our corporeal reality online, we are free to change gender, age, ethnicity and more online.

Therefore, online, it appears relatively easy to conceal, exaggerate, or fabricate traits. What is it about the media that facilitates this play?  There are several characteristics of computer-mediated communication (CMC) that probably contribute to this identity play, but perhaps the most significant is anonymity.

Much CMC, from weblogs through to email to instant messaging, is characterised by some degree of anonymity. At one end of the anonymity continuum is identifiability, where users are identified by the use of their real name, by features of their email address, or even through photographs. At the other end of the continuum is the possibility of being completely untraceable online. Most communication, however, falls somewhere between these extremes, in a state of visual anonymity or physical invisibility, where those who are communicating do not see each other, but have access to some identifying data (Joinson, 2003). This middle ground includes pseudonymity or the use of self-selected screen names, a common practice  in many forms of ODR.

The physical invisibility that still characterises much online communication means that identity is no longer tied to any physical form; the self presented online is not fixed to a physical body or a specific location. This freedom from physical reality gives licence to self-select attributes and present “hoped for selves”. As Joseph Walther (1996) argued, CMC permits communicators to “express themselves in ways more revealing of their … self-ideals than they might otherwise” (p.4). 

Thus, previous research has supported identity play online. Users, we are told, act out idealised selves and experiment with facets of identity that are difficult to change in the offline world. Such “experiments” are facilitated by the anonymity of many forms of online communication. These findings have important implications for online dispute resolution. Are people who they say they are? To what extent do people present themselves honestly online?

This research had two broad aims. First, I wanted to quantify identity play online. What are the major ways in which users play with their presentation online and how common are these strategies? Second, I sought to examine the predictors of such behaviour online. When are people most likely swap gender or present hoped for possible selves online? In particular, to what extent can variables such as previous online experience, personality, sex, age and ethnicity predict identity play?

Study One

The Participants Method

Participants were 75 (54 females and 21 males) undergraduate University students studying in either Australia or Singapore. Two-thirds were born in Australia and one-third in Asia. Nearly all participants (95%) spoke English as a first language. Although all had email experience, in other areas, such as chat and discussion boards, they had low to moderate experience.

Procedure

The Participants described themselves offline, as they perceived themselves to be, using a standard measure of identity known as the Twenty Statement Test. They also described their ideal identity or “hoped for selves”, the selves they wanted to become. Personality was measured using the NEO, a widely used measure of personality.

Online, the participants logged on to the front page of a community called The Place. The Place was described as a community established for people to meet and socialise. After logging on participants were asked to give themselves a screen name for their time in the community, choose a sex, and write a personal description. After completing the data collection tasks, participants logged off. They did not engage in any interaction with other participants. To control for order effects, participants were randomly assigned to complete either the offline or online measure first.

Results

Surprisingly, give the emphasis on gender swapping in the literature, only two participants swapped gender, both of which were female, and both chose a male gender.

The descriptions people wrote of their offline identity were compared with the descriptions written online. When this data was coded using some standard categories several similarities and differences emerged (Figure 1). First, age and sex, ascribed characteristics, which featured infrequently in the offline descriptions of identity, were more common in the online descriptions. That is, online, there was a tendency to state those characteristics that are generally obvious in face-to-face interactions. And these statements were typically honestly presented.

Figure 1. Identity profiles for offline and online descriptions.

Second, physical appearance played a relatively unimportant role in the online descriptions. Physical appearance was mentioned by only 20% of participants. Despite what we might imagine about people constructing a highly attractive appearance, people did not do this in their description. Most of the descriptions that did occur, like these two examples below, appeared to aim for honesty rather than deception. 

I am short, not tall, brown hair, brown/hazel eyes. Not ultra skinny, not fat … (miss , 22-year old, Australian female)

I am medium height, medium to heavy build, ok looking, brown eyes, brown hair… (Tiger, 24-year old, Australian female) 

The third area of interest is the category labelled “character descriptions”. Offline descriptions focused primarily on character traits e.g., “I am an easy going person”, “I am tolerant”, “I am shy”. This same emphasis was noted in the online descriptions. The way people chose to represent themselves to others online was predominantly through these personality traits. Exploring these in more detail revealed several sub-categories of character traits that were presented online. Particularly prominent were descriptions of humour, sociability, and having an easy-going nature, as illustrated in the following quotes.

I have a good sense of humour and am quite witty… (Jimmy, 19-year old, Australian male)

Loves to chat and meet new people as well as having fun. (Noddy, 22-year old, Australian male)

Usually fun, happy, easy-going. (aaa, 19-year old, Asian female)

These are characteristics generally seen as attractive and are likely to be valued in any group whose main objective is socialising.

This finding of “putting one’s best foot forward” was supported by other data in the study. Participants were more likely to present aspects of their identity they identified as positive than aspects they identified as neutral or negative, c² (2, N=65) = 66.22, p = .000. But participants did not typically take the opportunity to present idealised versions of themselves. Offline identity was presented more frequently than “hoped for selves”, Z (65) = -4.54, p = .000, d = 0.85.

Because so much has been made of the potential to present idealised images online, this relationship was examined in more detail. Two variables were related to the presentation of “hoped for selves”. First, those who presented idealised versions of themselves online tended to have less online experience (c² (2, N=65) = 11.41, p = .003) suggesting that the more experience one has online, the less likely one is to present idealised selves. Second, Asian-born participants were less likely to engage in idealised self-presentation than Australian-born participants, c² (1, N=65) = 5.96, p = .021.

Results from the first study therefore suggested that gender play may not be as common as writers in the 1990s implied. Age, sex, and personality variables turned out to be poor predictors of how people presented themselves online. Previous CMC experience and ethnicity, on the other hand, did help to explain some aspects of identity play, suggesting that some demographic variables may be important in understanding the complex ways in which users present themselves online. Although there was little evidence of explicit fabrication or idealised self-presentation, there was evidence that participants edited descriptions to emphasise characteristics particularly relevant to the context and attempted to create favourable impressions.

Study Two

These findings generated further questions, such as how conscious is the process of self-presentation online and how is the process played out over time? These questions were explored in a second study.

Study 2

Method

Participants

The Participants in the second study were 20 students (12 females and 8 males) enrolled in a course I teach online called “The Psychology of Cyberspace”. The students do not meet each other face-to-face over the 12-week period; all our interactions take place in an online MOO community. [1] We all use screen names and select from a choice of 10 genders. The demographic profile of the students was similar to the sample in the first study.

Procedure

Participants kept a journal reflecting on their self-presentation choices throughout the course. This second study involved a content analysis of these journals, a detailed, qualitative analysis of the ways participants understood self-presentation online. 

Results

There are four main findings that I want to report here. First, the intention to present idealised selves was relatively uncommon, replicating the results from the first study. When it did occur, it typically decreased with time, as illustrated in the following quote from a participant who established a character with magical powers.

At the beginning of the semester, I used the character of Oracle to interact with other people… casting a spell or summoning…. However, as the semester passed, I found myself beginning to drop the persona, becoming more of 'myself‘…. Acting in character was a good way to introduce myself to others, but as I found out more about them and their lives, I felt the need to talk more normally, rather than playfully.  

Second, participants articulated a strong and clear motivation to present themselves honestly online. The following statements from two participants were typical.

From the start, my character was supposed to just be a vessel to hold the real me.  I was not trying to be something other than I am in real life… (Andromeda1978) 

I made a decision before the first online conference that I would be myself and tell the truth. (Aquarius)

Third, despite this emphasis on honest self-presentation, there was also an appreciation of the role of selective self-disclosure, of revealing certain aspects of self, while concealing others. This strategy can play an important protective role, however, in this context participants talked about the seductive, attractive nature of concealing some information. Revealing too much too soon, could be seen, as one participant noted, as “boring”.

I wanted to be myself, but with a little mystery. (Tinkerbell)

  … a completely 'open-book' approach can often be perceived as boring. (Oracle)

Fourth and finally, the participants found themselves behaving in ways they had not predicted. Shy participants found greater confidence online. As one participant put it,

You could probably imagine the me online is quite similar to me in real life after a couple of beers… (Andromeda1978).

For most participants this reduced inhibition was a pleasant experience.

In general I enjoyed being Robotics because it was just me with the face to face inhibitions removed, I enjoyed it because I could relax… and I felt no expectations to be something I'm not…

However, for some it resulted in the presentation of aspects of identity they normally kept hidden in everyday interactions.

The environment … caused a few traits to come to the surface that many other people normally do not see.  The MOO was very new and different to me and so [my] character Jogirl often seemed a bit 'ditzy'…. I often felt as if I asked the silliest questions and felt frustrated a lot of the time.  I have many of these qualities in real life, but I try not to let any one see them. This made Jogirl and myself seem a little vulnerable.  

Conclusion

Using two very different methodologies, the two studies produced strikingly similar results. Identity play, as it had been described in the early literature, was relatively uncommon. The most obvious form of identity play, gender swapping, was rare. There was little evidence of deliberate deception, although there was, as is typical of most self-presentational situations, a tendency to shows one’s best qualities and attempts to hide vulnerabilities. The second study in particular highlighted the role CMC plays in facilitating disinhibition and the presentation of “true” selves, i.e., what one really believes oneself to be, even though one might not present oneself that way generally.

Concealment and exaggeration both played a role online. Both existed, however, not so much as a way to present idealised selves, but as a vehicle for the presentation of one’s true self. The same anonymity that was originally hypothesised to enable fabrication and deception, like that played out by Alex in the character of Joan Greene, also appears to facilitate honest self-presentation, permitting users to reveal themselves in ways more truthful than in the offline world. 

The findings suggest that people use similar identity processes online as they do in the other arenas of life. Despite the opportunities for identity play afforded online, these results suggest that users find it difficult to present a self that is not "the real me“. As one of the participants, BritChick, put it, “After using chat for a while I have found that it’s best and easier to tell people the truth.”

References

References

Joinson, A. (2003). Understanding the psychology of internet behaviour: Virtual worlds, real lives. Great Britain: Palgrave Macmillan.

Reid, E. (1994). Cultural formations in text-based virtual realities. Unpublished master’s thesis, University of Melbourne, Australia. Retrieved January 4, 1998, from http://www.ee.mu.oz.au/papers/emr/cult-form.html#Pubs

Turkle, S. (1995). Life on the screen: Identity in the age of the Internet. London: Weidenfeld and Nicolson.

Van Gelder, L. (1991). The strange case of the electronic lover. In C. Dunlop & R. Kling (Eds.), Computerization and controversy: Value conflicts and social choices (pp.364-375). New York: Boston Academic Press.

Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23, 3-43.

Waskul, D. & Douglass, M. (1997). Cyberself: The emergence of self in on-line chat. The Information Society, 13 (4), 375-396.