International Conference: 25-27 March 1998, Bristol, UK IRISS '98: Conference Papers

Proceedings

Title: The Presentation of Self in WWW Home Pages

Authors: Hugh Miller and Russell Mather

Introduction

Identity is socially mediated (Gilligan, 1982), and much of that mediation is through language (Harre, 1989). It follows that as new social processes and new ways of using language emerge, it may be possible to develop new aspects of identity. It has been suggested, for instance by Gergen (1991, 1992), that the developing communication technologies of the last twenty years have had profound implication for our sense of self.

The World Wide Web is one such new technology, which allows what has been, up till now, an unusual form of communication. Rather than one-to-one or one-to-many, the Web is semi-interactive one-to-the-World communication. It has long been possible for a few individuals to publish advertisements or manifestoes or autobiographies which might be read by large numbers of people, but now anyone with a few megabytes of server space can publish material which might be read by anyone in the world with a browser. This publication is usually mainly one-way, but if the author chooses to put an email address on the page (generally considered good Web practice) most readers can contact the author directly if they wish.

There has been considerable research on the social aspects of electronic communication, including the implications for identity, but most of that research has focussed on interactive communication in newsgroups, chat rooms, MUDs and MOOs, where the communication is text based and intended to be interactive (Danet,1996; Turkle 1995). Less attention has been paid to the WWW as a social phenomenon, though Erickson (1996) has identified the development of a 'social hypertext', and Chandler (1997) and one of us (Miller, 1995) have considered the implications of, to use Chandler's phrase, 'writing oneself in hyperspace'. Donath (1998) points out that the Web page can provide a more solid backup for the problematic claims of identity, gender or expertise that are made in interactive electronic communication: 'In a forum where a link to one's Web page is the norm, the opportunity to explore multiple personas may be greatly curtailed', and Kelly (1995) asks ' Why are we emphatic about who we are on the web but then go on to MUDs and IRC to be someone else?'

On a Web site, it is customary to have a home page which provides an introduction, and begins to establish the identity of the site. When that site is a personal one, it seems that an aspect of identity, someone's Web self, will be established and communicated through the home page. In an earlier, impressionistic, survey (Miller 1995) one of us came to the conclusion that most personal home pages drew on established print ways of presenting the self (such as the high school year book entry, CV, pen pal letter, or lonely hearts ads, and advertisements for the self, either egocentric or commercial), and there was not much sign of new variants of identity emerging through the WWW. Chandler (1997) adds several more categories - autobiographies, scrapbooks, fanzines, family slide-shows and others - but still mainly drawing on models of presentation in print. Of course, we print-reared commentators would say that: we may not have developed the language or the conceptual framework to deal with alternative forms of self-presentation which are out there on the Web.

On further consideration, perhaps combined with the maturing of the medium, we can see examples of a new way of presenting the self - the hypertext self. The Web has the ability to present lots of pieces of information linked together in complex ways with no necessary order or hierarchy. By using this, people can present many aspects of themselves simultaneously (or at least non-hierarchically), or their extended selves, or themselves as nodes within an extended community (though Rubio (1996) points out that this may 'only be a simulation of a community, not even as communal as a mailing list'). So pages may provide views of 'X the chemist, X the family man, or X the singer', family pages can present a family identity comprising members past and present, or people can present a self built round some structure , sometimes allowing others to add to and modify the self-structure. The site that used to be 'Bianca Troll' became 'Bianca's Smut Shack', and is now 'Bianca.com, the alternative online community' (http://www.bianca.com/).

It's interesting that these last kinds of pages often use buildings as metaphors for the structure being presented. The use of spatial metaphors for data is very common - it is cyberspace after all - ever since William Gibson's Neuromancer (1984). Perhaps pure non-spatial hypertext is too difficult to navigate - or perhaps the authors themselves need a structure within which to conceptualise their extended selves. A house or home does seem to be the ideal metaphor for the structure of the data of the self. One of Chandler's (1997) home page categories is 'entire living spaces and homes, with their furnishings, posters, bookshelves, music collections, photos and so on'. Bauerbach (1997) suggests that the real-house home is a useful analogy to help people evolve an acceptable etiquette in the undefined social setting of the Web.

There is another reason for studying social processes in electronic communication: as well as looking for new manifestations, we can check out old theories. We can assess the robustness of psychological explanations developed in other areas by seeing if they apply to this new situation.

Gender Differences in Home Pages

We have chosen to look at gender differences in personal home pages on the WWW. Are there differences between the kinds of identity presented by men and women, or in the ways that identity is presented? Gender has been a significant topic in writings about electronic communication (Herring, 1994; Stone, 1991; Turkle, 1995; Donath, 1998), but this has mainly been about the negotiation of gender or gendered styles of communication in interactive communication.

We are aware that there are some basic problems with our approach. Looking for differences between the genders has long been criticised as being sexist in itself. Why establish differences, unless it's for the sake of validating discrimination (Wilkinson, 1997)? On the other hand, in a psychology where masculine is 'normal', it seems important to give equal attention to other ways of being (Burman, 1996) - especially in male-dominated areas like electronic communication.

It also seems inappropriate to have two men looking at an aspect of women's self-construction. What would we know about what's going on and why - especially as we haven't bothered to ask the people concerned what they think they're doing? I don't think we can counter this objection. If we extend this research, it would be essential for us to work with women collaborators, and essential to start talking to home page authors about their understandings and intentions. Working as detached outsiders, we may be missing the point.

All the same, we gave in to the temptation to make a start at looking at gender differences on the Web. Women are establishing a presence on the Web, there is (as far as we know, and Chandler (1997) supports this) little gender play or ambiguity in Web pages compared with other forms of electronic communication, and the pages themselves are relatively stable and provide plenty of opportunity for analysis.

The analysis we used was partly theory-driven, but mainly resulted from simple-minded pragmatism, which we think is appropriate for an initial approach. We counted those things which were easy to count, and which we thought might be affected by gender difference.

We looked at 35 women's and 35 men's personal home pages. There is a problem in establishing a sampling strategy for research like this. No-one really knows just what's out there on the WWW, how many of the pages are personal, who the people posting them are, and so on, though we have some indications (Buten, 1996; Hoffman, Kalsbeek and Novak, 1996). Even if we did know, there doesn't seem to be an established systematic way of sampling sites. We took an easily available source of home pages, Yahoo's White Pages (http://www.yahoo.com/society_and_culture/people/Personal_home_pages/white_pages/), and sampled fairly systematically from that. From each initial letter group, we picked roughly proportional numbers of sites which seemed gender-identifiable, and recorded the first ones we found which were personal, rather than commercial (about 10% were discarded as non-personal), and where the gender of the author was clearly identified. We carried on until we had equal numbers of men's and women's pages. This required a longer search for women's sites; the overall ratio of pages in the list was about 15% female, 75% male and 10% other (companies, universities or unidentifiable by name alone).

The page that first loaded on each site, whatever the structure of the overall site, was recorded as the home page, unless it was an almost content-free 'welcome' page, and was printed out in black and white for later analysis. Any distinctive features of background or colours, which wouldn't appear on the printout, were noted.

Analysis

The length of each page was measured, in half pages of A4 printout (roughly equivalent to screens). Most pages were one, two or three screens (means of 3.1 for women, 2.5 for men). The differences in means mainly results from a subgroup of longer womens' pages (six screens or more).

Style: two main types of page were found: 'low-content' pages mainly made up of links to other pages/other sites, and 'high-content' pages, with more information, and often a narrative of self, up front. Pages were classified as low-content, high-content or unclassifiable. There was no gender difference at all here, with 22 low-content, 11 high-content, and 2 unclassifiable pages in each group.

A traditionally-identified gender difference has been between 'expressive' and 'instrumental' orientation (Bem, 1981). We examined this by looking at what was mentioned and linked to on the page. A more expressive style would focus on feelings, people, and relationships, while the instrumental style might show itself in reference to abilities and achievements, material goods, and organisations and products rather than people. Mary Gergen has pointed out similar biases in popular autobiographies (Gergen, 1994), and Csikszentmihalyi (1991) and Belk (1987) both found male-female differences along these lines in people's reactions to material objects and accounts of how those objects related to their idea of self.

Various measures might relate to this dimension. We counted links to other people, compared with links to non-personal sites. Women did put up more links to other people (mean of 1.8 compared with 1.2 for men), but they also had more links to non-personal sites (12.0 vs 9.4). Women also show more awareness of, and engagement with, the visitor to the site. Women's pages had a mean of 4.5 references to the reader (using words like 'you', 'yours', or expressions of awareness of the reader), whereas men's pages had 2.6. Guestbooks were more common on women's pages (10 to 6) as were counters (21 to 13).

Many home pages show pictures of the author. This is so common that it has become a norm, that can be satirised, opposed, or apologised for ('sorry no picture - I'll get one up just as soon as I can get my scanner working'). We thought there might be gender differences here. Aspects of objectification and male gaze (Berger, 1972), the way the dominant culture problematises self-portraits for women (Edholm, 1992), and abuse by men (as in the 'Babes on the Web' page, on which women who do put their photo on their Web pages are rated by a distant observer for their attractiveness), all make the posting of a photo more problematic for women than for men.

We identified four categories for self-image on the page:

• straight: an image which purports to be a straightforward likeness
• joke: a distorted or caricatured or unrepresentative image: cartoon, baby photo, author just after falling off bike into mudhole, author caricatured as frog, etc
• symbolic: an image which represents a human being, but not the actual person who posted the page. This is often a piece of clip art, like a cherub or a generic silhouette.
• none: no images of humans

We counted blurred or pixellated photos which might be of the author, but were so unclear that they didn't really represent an individual, in the 'symbolic' category.

We were a bit surprised to find that there were several (15 out of 35 for both groups) pages with no images at all. Men's pages had more 'real' images (10 compared with 6), as we expected. The big difference was in the other two categories. Joke images only featured on men's pages (on 4), and symbolic images only on women's (on 10 pages; the most common form of image on women's pages).

Conclusions

So, in the small sample we looked at, there are some interesting gender differences in personal home pages. Women's pages are sometimes longer, contain more links, and show more awareness and responsiveness to the reader, in various ways. When women and men represent themselves visually in their pages, only men used joke images, only women used symbolic representations.

This is interesting, but there is a lot more to do. We would like to do more qualitative analysis; the kind of language used (Herring 1996 has found clear gender differences in language used in interactive electronic communication), the way in which people use their names (given name, nickname, name as title for the page or not), the topics discussed, and usage of 'I' and 'me'. All these results need to be related back to theory.

Most importantly, though, we would like to involve the subjects in the research. If people are 'constructing themselves in cyberspace' (Chandler, 1997), we may learn more by talking to people about what selves they think they are constructing and why, than by analysing the constructions on the screen.

References

Bauerbach, S., Meditations on the Metaphysics of Home Pages. Available at http://www.amherst.edu/~sbauerba/philo.html (1997).

Belk, R., Identity and the Relevance of Market, Personal, and Community Objects. In D. Umiker-Sebeok, Marketing and Semiotics, pp151-164. Mouton de Gruyter (1987).

Bem, S.L., Gender Schema Theory: a Cognitive Account of Sex Typing. Psychological Review, 88, 354-364 (1981).

Berger, J., Ways of Seeing. Penguin (1972).

Burman, E., Challenging Women. OUP (1996).

Buten, S., The First World Wide Web Personal Home Page Survey available at http://www.asc.upenn.edu/usr/sbuten/phpi.htm

Chandler, D., Writing Oneself in Cyberspace. Available at http://www.aber.ac.uk/~dgc/homepgid.html (1997).

Czikszentmihalyi, M., Design and Order in Everyday Life. Design Issues, 8, 26-34 (1991).

Danet,B., Text as Mask: Gender and Identity on the Internet. Available at http://atar.mscc.huji.ac.il/~msdanet/mask.html (1996).

Donath, J. S., Identity and Deception in the Virtual Community. In P. Kollock and M.Smith (in press), Communities in Cyberspace. University of California. Available at http://judith.www.media.mit.edu/Judith/Identity/IdentityDeception.html

Edholm, F., Beyond the Mirror: Women's Self Portraits. In Bonner, Goodman, Allen, Janes and King, Imagining Women, pp 154-172. Polity Press (1992).

Erickson, T., The World Wide Web as Social Hypertext. Available at http://www.pliant.org/personal/Tom_Erickson/SocialHypertext.html

Gergen, K. J., The Saturated Self. Basic Books (1991).

Gergen, K. J., The Decline and Fall of Personality: How the Fax, the Phone, and the VCR are taking us beyond ourselvelves. Psychology Today Nov/Dec, 59-63 (1992).

Gergen, M., The Social Construction of Personal Histories: Gendered Lives in Popular Autobiographies. In Sarbin and Kitsuse, Constructing the Social, pp19-44. Sage. (1994).

Gibson, W., Neuromancer. Gollancz (1984).

Gilligan, C., In a Different Voice: Psychological Theory and Women's Development. Harvard University Press (1982).

Harre, R., Language Games and the Texts of Identity. In J. Shotter and K. Gergen Texts of Identity. Sage (1989).

Herring, S.C., Gender Differences in Computer-Mediated Communication: Bringing Familiar Baggage to the New Frontier. Available at http://www.cpsr.org/cpsr/gender/herring.txt (1994).

Herring, S. C., Gender and Democracy in Computer Mediated Communication. In R. Kling, Computerization and Controversy. Academic Press (1996).

Hoffman, D. L, Kalsbeek,W.D., Novak,T.P., Internet and Web Use in the U.S. Communications of the ACM, 39 (13), 36-46 (1996).

Kelly, J., Human Identity Part 1: Who Are You? Available at http://www-home.calumet.yorku.ca/pkelly/www/id1.htm (1995).

Miller, H. W., Goffman on the Internet: the Presentation of Self in Electronic Life. Paper presented at the Embodied Knowledge and Virtual Space Conference. Available at http://www.ntu.ac.uk/soc/psych/miller/goffman.htm (1995).

Rubio, S., Home Page. Bad Subjects, issue 24. Available at http://english-www.hss.cmu.edu/bs/24/rubio.html (1996).

Stone, A. R., Will the Real Body Please Stand up? Boundary Stories about Virtual Cultures. In M.Benedikt, Cyberspace: First Steps. MIT (1991).

Turkle, S., Life on the Screen. Simon and Schuster (1995).

Wilkinson, S., Feminist Psychology. In D.Fox and I. Prilleltensky, Critical Psychology. Sage (1997).

The Authors

Hugh Miller is Principal Lecturer in Psychology in the Department of Social Sciences, The Nottingham Trent University. He splits his teaching time between psychology and design, and is particularly interested in the interactions between designed objects, technology, and social processes.

Russell Mather is a third year student on the BSc Psychology degree at the Nottingham Trent University.

Themes	Home	Proceedings