About thirty empirical studies on personal home pages can be found in the literature (see Appendix). Most of them refer to the three theoretical perspectives presented above, but only a few of them go beyond description and interpretation to hypotheses-testing. As a systematic literature review has been absent, previous personal home page research is merely cumulative and authors rarely cite each other. To give a coherent overview of the diverse previous findings it was necessary to structure them according to relevant research questions. In doing so, a communication studies framework was used: Personal home pages are regarded as media products with their specific production processes, product characteristics and reception processes.
Production of Personal Home Pages
Six research questions on the process of producing personal home pages can be answered tentatively based on previous research.
What are the characteristics of personal home page owners?
Home page owners, who constitute a minority of 10% within the Internet population (see the “Prevalence of Personal Home Pages” section) tend to be young, male, students and to posess an above-average level of Internet experience (Berker, 1999; Buten, 1996; GVU, 1998). Belonging to certain occupational fields (e.g. politics, Web design) or identifying oneself strongly as a netizen (see Frindte, Koehler & Schubert, 1998; Walker, 2000) increase the probability that the individual has created a personal home page. Home page engagement in one's own social network as well as participation in online-chat communities are also important predictors of home page construction (Karlsson, 1998, 2000). Despite the popular suspicion of narcissism or exhibitionism (see the “Introduction” section), the 96 randomly selected home page authors at the University of Hannover in Germany (96% men, 4% women, average age: 26 years) didn't differ significantly from the statistical norm on the dimensions of “openness,”“inhibitednessm”“social orientation,”“competitiveness,” and “satisfaction with life” of the Freiburg Personality-Inventory, FPI (Albat et al., 1998). (Other FPI scales were not included in the Albat et al. study.)
How intensively are personal home pages maintained?
Ninety-six percent of the 110 home page owners surveyed by Killoran (1998) stated that they have already changed their home page at least once since its inception and also plan further changes. Sixty-eight percent of the respondents in Buten's study (1996) worked on their Web site at least once a month. These indications of intensive home page maintenance are pitted against the findings of Berker (1999) that 26% of the home pages at the University of Fankfurt (Germany) remained completely unchanged over a year, according to the server statistics. Of the 96 randomly selected student home page authors at the University of Hannover, the majority (56%) rarely or never worked on their home page, according to their own accounts, while only a small minority of 7% updated theirs often or very often (Albat et al., 1998). In view of the heterogeneity of the samples and the questionnaires, the findings cannot be aggregated. The wide spectrum of actual home page commitment should be stressed here. Although the frequency of updating did not correlate with the home page's length of existence (Buten, 1996), a positive correlation was discovered between update frequency and Web site volume (Berker, 1999). Surveys regarding the types of updates were not carried out. At best, which changes were made when can be reconstructed anecdotically on many home pages based upon their “update-history” or “what's new”-section.
How are the form and content of the personal home page created?
Both in connection with their design as well as their content, existing personal home pages are used as templates and material resources: ninety-five percent of the 121 subjects in the Buten (1996) study reported adopting elements from other home pages when constructing their own personal home page. A content analysis of 400 randomly selected personal home pages of students at four US universities revealed that 43% of the sites infringed on copyright laws by using protected images (Herbeck & Hunter, 1998). Apart from such legal implications, a collage-like process of publication is to be read as meaningful self-construction in the opinion of postmodern identity theorists (see the “Identity Theories” section). In addition, original creations are also presented in the textual realm, along with copied material both from old and new media (e.g. quotations, sayings, jokes). These textual contributions rest on handed-down bureaucratic, biographical, journalistic and artistic texts and in doing so demonstrate a high degree of identifiablity and authenticity when taken as a whole (see the “Theories of Computer-Mediated Communication” section): social-statistic particulars, tabular and narrative CVs, copies of important certificates and work samples, vacation documentations, portraits, wedding photos and photos of children, online diaries, to-do-lists, in/out-lists, self-interviews, poems, stories, and drawings are among the material to be found (Bates & Lu, 1997; Chandler, 1998; Döring, 2001; Karlsson, 1998; Miller, 1995; Miller & Mather, 1998). In addition, the metaphor of the electronic “home” is quite often taken up (Bates & Lu, 1997, pp. 334, Miller, 1999): My bicycle, my car, my computer, my pet, my electric train, my saxophone - such an inventory is exhibited with detailed descriptions and photo documentation. From a linguistic standpoint, those maintaining personal home pages are momentarily participating in the conventionalization of a new genre, by orientating the function and style of their Web sites to each other (Crowston & Williams, 2000; de Saint-Georges, 1988; Dillon & Gushrowski, 2000).
For which audiences are personal home pages designed?
The survey study by Buten (1996) found, on the basis of six (not completely selective) adressee categories, that home page owners (at least when surveyed) presume a very heterogeneous audience, which extends from their closest personal environment (friends, family) to acquaintances from the Internet, colleagues from work and people with similar interests all the way to unknown random guests (“surfers”). This variety is reflected in direct addresses to the public on the home page (Döring, 2001, p. 228), although in general members of the owner's own social offline and online networks or of particular related groups are more important addressees than the Internet public in the abstract. The “patchwork-audience” that accesses a personal home page is especially difficult to deal with from the self-presentation perspective and marks a departure from face-to-face contexts, in which we have smaller and more segregated audiences and addressees (see the “Theories of Self-Presentation” section). For instance, the 25 Welsh youths interviewed by Chandler and Roberts-Young (1998) reported that some of them were embarassed when fellow students or teachers discovered their home pages, which they had actually prepared for their chat-friends. Haase (1999) analyzed 48 home pages of lesbian women, and reported that the female home page authors each had very specific target groups in mind. In over half of the cases (57%), all those interested were invited regardless of gender or sexual orientation. Seventeen percent of the home pages were expressly intended “only for lesbians,” 8% only for women, regardless of whether they were homosexual or heterosexual, and 8% appeared to be addressed only to gays and lesbians. In 2% of the cases lesbians, gays, bisexuals and transsexuals were explicitly welcomed. Only on 8% of the pages did the target group addressed remain vague.
For what reasons are personal home pages created?
A number of authors consider motives of intrapersonal and interpersonal communication to be the decisive factors in the construction of a personal home page. Killoran (1999) characterizes personal home pages both as personal projects (motives of self-construction, freedom from outside definition) and social projects (motives of sense of belonging, freedom to make contact). Hevern (2000) stresses that in home page construction motivation involving positive self-construction (Self S-Motive) is articulated just as much as motivation involving making contact with others (Other O-Motive). Befring (1997) takes the need to express oneself (expressionist) and to get to know others (communicationist) as fundamental assumptions. The interpersonal orientation of home pages is also pointed out by Erickson (1996, “social hypertext”), Karlsson (1998, 2000, “socio-textual networks”), Dominick (1999, p. 655: “social association” and “linkage” function) and Groth (1998, 1999: “knowledge net”) and is consistent with that noncommercial, humane vision of the WWW that Web inventor Tim Berners-Lee (2000, pp. 226) promotes.
The main reason for maintaining a personal home page is not just the furtherance of intrapersonal and interpersonal communication. Just as often, maintenance of a Web page is maintained by a housekeeping motive of being able to access certain Web sites faster using link collections on one's personal home page, or the autodidactic motive of being able to improve one's own Web competence (Buten, 1996). Bates and Lu (1997, p. 333) suspect that just about a third (29%) of the 114 home pages they analyzed are the result of enjoying “play with system capability.” Finally, there are extrinsic motives, amounting to basic task fulfillment, which must also be accounted for (Befring, 1997). Döring (2001, p. 229) found, among others, the following meta-remark on one of the 363 student home pages she investigated:
Well, I do admit that if I didn't have to, I would never have thought of making a personal home page…. I also don't even want to…especially because I don't know what I should write!!! Oh God, I'm so uncreative again today!! But Prof. Hefele did say that even if you're having a bad day, a landscape architect has to be able to be creative….but that's not helping me any further at the moment! Oh well, to get a first impression of me you can take a look at my wonderful little picture or enjoy my ‘tabular’ CV!! I also still have a few links and email addresses for you!
Do gender effects exist in personal home page production?
Contrary to the thesis of gender masking or gender switching proposed in connection with “virtual identity” (see Turkle, 1995, p. 210), the creators of personal home pages display a strong willingness to supply authentic and reviewable information about their own person. The overwhelming majority of the home page authors surveyed by Buten (1996) assessed both their own home pages (91%) as well as home pages of others (78%) as reliable, unaltered self-presentations. No other self-aspect is so often and so clearly (through name or photos) presented as gender (Döring, 2001, p. 229). When comparing home pages maintained by women and men, as the content analyses of Befring (1997), Miller and Arnold (2000), Miller and Mather (1998) as well as Dominick (1999) indicate, the familiar gender-specific styles of self-presentation reproduce themselves on the net: Men seem more to emphasize their status, to choose motifs related to technology, to utilize the latest net technology and to express themselves briefly in writing, while women tend to integrate more addresses to the audience, to fall back on floral designs and pastel colors, to opt for a technologically simpler execution and to offer longer biographical narrations.
We should be careful, though, not to make any rash affirmative statements on the basis of exploratory studies. The quantitative content analysis of Dubi, Lauper, Schlapbach and Witschi (1998) found more similarities than differences between men's and women's personal home pages. The qualitative content analyses of Karlsson (1998) and Stern (1999) suggest, on the basis of home pages of female pupils from Sweden and the USA, how subversive gender constructions can be inferred alongside traditional self-presentations of a “nice and kind girl”. Kibby (1997) argues that the presence of corporeality on women's personal Web sites does not just make them susceptible to unwanted sexual attention, but can also be a self-determined contribution to sexual identity construction (Stern, 2000). The typical self-presentation dilemma of highly qualified women, namely, the collision of femininity with competence, power and autonomy, is also an issue in home page production processes. Twenty-seven female home page creators working as professors at English and North-American universities interviewed by Miller and Arnold (2001) reported being concerned that their professional status, perceived as structurally threatened, could further be undermined by private self-presentation on the Web. They did, however, describe the opportunities for self-presentation on the Web as a positive contribution to emancipation, as did the seventeen female home page authors identified as feminists who were surveyed by email by Kennedy (2000). The fifteen women working in the sex industry who were personally interviewed by Podlas (2000) also gave the same account. Their work conditions had clearly improved thanks to their own home pages, not least of all through greater independence from men. Whether and to what extent the online boom of (semi-)professional sex-sites will reinforce or reduce gender-specific victimization in general is still a hot topic of feminist debate (Döring, 2000).
Classification of Personal Home Pages
According to the heterogeneity of personal home page production processes described, above the resulting media products are also fairly diverse. A basic classification of all personal home pages is proposed which can be refined by a subclassification of self-expressive personal home pages.
Basic classification of personal home pages
Before we classify personal home pages according to their identity, self-presentational or communicative functions, we must first take their form of existence (home page not available: “nominal home page” versus home page available: “actual home page”) and their construction status (advance notice of a home page: “projected home page” versus home page with substantial content: “realized home page”) into account. Also the content focus (treatment of one topic without explicit reference to one's own person, an “instrumental home page” versus making one's own person the topic, an “expressive home page”) is from the standpoint of identity construction and self-presentation an important classification characteristic.
In a content analysis of 279 student home pages selected from university directories, Döring (2001) found that only 42% of the personal home pages listed in the university directories corresponded to the image of the typical self-presentation page (see illustration 1), that is, a) were accessible, b) offered substantial content and in doing so c) placed one's own person at the center of focus. In a random selection of 500 home pages from the Yahoo directory, Dominick (1999) found that about 30% (149) were not accessible, and therefore fell in the category of nominal home pages.
Only a minority of 42% of home page owners have well-maintained Web sites which are dedicated to expressive self-presentation. Still the the well-maintained personal home page which is dedicated to expressive self-presentation is repeatedly postulated as the “typical personal home page” in several home page studies (e.g. in de Saint-Georges, 1998; Erickson, 1996; Walker, 2000; Wallace, 1998, p. 33; Wynn & Katz, 1997). Therefore undermaintained personal home pages as well as instrumental personal home pages represent an understudied majority of personal home pages.
Subclassification of self-expressive personal home pages
Expressive personal home pages are not the most common but the most interesting personal home page type when it comes to questions of identity, self-presentation and computer-mediated communication. They need to be subclassified based on several dimensions of home page use:
- •Reason for the construction of the personal home page: intrinsic versus extrinsic home pages. Home pages which arise from intrinsic motivation can be distinguished from extrinsically motivated home pages. Befring (1997) discovered, in a sample of 25 home pages of Norwegian youths, that intrinsically motivated “stray pages” (with sophisticated and individual design) clearly differed from extrinsically motivated “school project pages” (with minimalist and stereotypical construction). Similar differences can also be observed among student home pages, which either come into being of one's own initiative or have to be laid out within the framework of courses (see the “For what reasons are personal home pages created?” section).
- •Addressees of the personal home page: intrapersonal versus interpersonal home pages. Expressive home pages can serve intrapersonal communication in the sense of identity-affirmation as well as be orientated to the influence of or contact with others in the sense of self-presentation. This dual function is not realized to the same extent on all expressive home pages. Home pages can be partially classified as more personal or more social projects (Killoran, 1999), as more expressionist or more communicationist (Befring, 1997) (see the section “For what reasons are personal home pages created?”). Intrapersonal home pages can be completely hermetic by not providing any means of contacting the owner, in the extreme case. On interpersonal home pages, however, interactive components are represented especially strongly: e.g. appeals to make contact via email, guest books, newsboards, chat rooms. Walker (2000, p. 106 ff.) distinguishes between two subtypes of interpersonal home pages according to whether previously known persons from the real environment are the primary addressees (e.g. family, friends, colleagues, former pupils) or whether the page addresses an unknown and wide Internet public. If the addressed home page audience has been primarily recruited from one's face-to-face circle of acquaintances, the virtual self-presentation is only a supplement and refinement (if appropriate) of previously established personal impressions. If, however, an unknown Internet public is addressed, the home page is a place of “first contact” and becomes in terms of self-presentation more meaningful and because of the want of information about the audience at the same time more complicated (see the “Self-Presentation Theories” section).
- •Themes of the personal home page: Expressive personal home pages can be differentiated according to which and how many self-aspects they select as central themes. Miller (1995) names five thematic focal points: 1. the own person, 2. the own person as as an organization member, 3. the own family, 4. the own interests, 5. the own competencies. A different selection of self-aspects is provided by de Saint-Georges (1998, p. 76) in her definition: “Personal Home page: presentation of the self in digital (hypertextual) form, authored by one individual, and which (i) emphasizes a person (minimally, by a picture or name); and/or (ii) a person's current activities, and/or (iii) professional experience; and/or (iv) displays a person's interests (in the body of the text and/or through hyperlinks to other sites).” It should be noted that identity theories and theories of self (see the “Identity Theories” section) do emphasize the multiplicity of the self, but up to now no conclusive classification of self-aspects has been agreed upon. Corresponding listings suffer from a lack of mutual exclusiveness and often aren't exhaustive, either. Neither Miller (1995) nor de Saint-Georges (1998) name self-aspects which refer to ethnic, religious or sexual identity, although these have been have been found on personal home pages (e.g., Hevern, 2000).
- •Presentational styles of the personal home page: Categorical, relational and narrative forms.Walker (2000) distinguishes three stylistic forms of self-presentation which can be found both alone as well as in hybrids. Categorical home pages supply self-related information through key-word-like self-categorization based upon age, gender, occupation, place of origin, and so on. Relational home pages characterize a person through hyperlinks to Web sites which represent the individual interests, attitudes or hobbies of the author. And finally, narrative home pages group the self-related information into an autobiographical story, which can be, for example, decorated with childhood photos.
- •Technological characteristics of the personal home page: technologically simple versus sophisticated home pages. As regards the technological expenditure, personal home pages can be fixed on a continuum, on the one side of which a page of pure text is located and on the other side, a complex Web site with multimedia and interactive elements. Technologically sophisticated pages indicate particular personal home page commitment that can be accompanied by playful and autodidactic use of new Internet technologies (see the “For what reasons are personal home pages created?” section) or also may be part of a competence-orientated or gender-specific self-presentation strategy (see Befring, 1997; Dubi et al., 1998).Designing a useful subclassification system for expressive personal home pages demands further research on transforming the different dimensions into well-defined, mutually exclusive and exhaustive categories.
Reception of Personal Home Pages
Four research questions on reception processes of personal home pages can be answered tentatively based on previous research.
How intensively are personal home pages received?
According to a study of the Benchmark Group (1999), among the 17 most popular French Web sites, which generate more than 10 million page impressions a month, there are a total of four community Web sites, where users can, among other things, anchor their personal home pages. Berker (1999) assessed the protocol data (20.1.-4.2.1998) of the proxy servers of the University of Frankfurt and found that personal home pages had already occupied second place among all Web page downloads of university members with 13% (sex-sites took first place). Personal home pages can become favorite surfing goals if they offer certain services (e.g. archives with erotic stories, software collections, factual and expert information, online diaries etc.). In such cases the construction of a personal impression of the home page owner is no longer a point of focus from the recipient's side, but instead much more a more concrete content-related use. But also for the purpose of social networking and cooperation, there are ocassions in both the private and professional realms for seeking out personal home pages, be it to get to know chat acquaintances better (Karlsson, 2000) or to find colleagues working in the same field (see Bly, Cook, Bickmore, Churchill & Sullivan, 1998; Groth, 1998, 1999).
Tauscher and Greenberg (1997) observed the surfing behaviour of n=23 experienced Web users for six weeks and discovered that users visit very few Web pages frequently. Consequently, many Web pages are only visited once (60%) or twice (19%). Of the 23 survey subjects, 18 had their own home page. These home pages were also typically among the Web sites most often visited by the person maintaining them. Only with 2 of the 18 home page creators did the download of one's own home page not make it onto the top 15-list. For 9 personal home page owners, their own site was the page most often downloaded at all, because it was used as a springboard for Web searches over its external links. This finding confirms the statement already made concerning the motives for home page construction, that personal home pages also fulfill other tasks (for instance housekeeping or autodidactic) than the function of self-presentation (see the “For what reasons are personal home pages created?” section).
When, how often and from which computer addresses documents from one's own home page are downloaded can be registered with Web statistics-programs. Making visible the popularity of one's own page in terms of page visits might spur on further home page commitment. Basic statistical information about the home page audience can also be evaluated as part of self-presentation: For instance, if one notices that one's non-English personal home page is often being downloaded from international computer addresses, one might want to think about putting together English-language content. One drawback of many free Web statistics is that one's own page call-ups can't be filtered out and the impression of the popularity of one's own home page can be greatly distorted.
What audience expectations are associated with the genre “personal home page”?
Dillon and Gushrowski (2000) analyzed more than 100 personal home pages with respect to individual elements (e.g. page title, email address, update dates, table of contents, photographs, biographical notes, electronic guest books ecc.). Afterwards they presented the list of elements to a sample of n=57 students, who should decide which of the elements on the list a good home page cannot do without. It was revealed that their expectations coincided very well with the frequency distribution found on the actual home pages. The five most important elements are presented in Table 2. Personal home page elements which can only rarely be found on existing home pages and whose availability is often not expected by the home page recipients (e.g. frames, sound-files) are to be differentiated from these genre-defining characteristics.
Table 2. Core elements of the genre “personal home page” (Dillon & Gushrowski, 200, p. 203)
In one experiment Dillon and Gushrowski (2000) checked the validity of the genre-typical characteristics. They presented subjects eight home pages as stimulus material. These home pages could be ordered according to the number of genre-typical (or untypical) elements realized on the page. This objective ordering could be replicated by the subjective judgments of the recipients, who were asked how well they liked the home pages. Where personal home pages play a role in professional life, the exchange of work- and project-related information has become the norm (see Groth, 1998, 1999). Personal home page owners and home page visitors also partially initiate discourses on page design by creators' asking their audience for design suggestions in their welcome message or by home page visitors' pointing out deficiencies or reporting wishes via email or guest book.
Do impressions based on personal home pages differ from face-to-face impressions?
Sherman et al. (1999) had thirty subjects assess subsets of 86 authentic personal home pages along four dimensions: 1. general impression of the persona of the home page creator, 2. sympathy, 3. perceived similarity and 4. completeness of impression. For purposes of comparison, the same subjects then assessed someone they knew well and were very familiar with, and someone they knew as a casual acquaintance, along the same four scales. Results indicated that impressions based on the personal home page visit didn't in general differ significantly from impressions based on casual face-to-face contacts, but did lag behind the positive impressions of familiar persons. Sherman et al. (1999) interpreted these findings as undercutting both the channel-reduction model and the model of hyperpersonal interaction. In fact, the results can be interpreted as a validation of the model of social information processing (see the “Theories of Computer-Mediated Communication” section).
Do impressions based on personal home pages differ from the expectations of the home page owners?
Independent of whether people present themselves face-to-face or on a personal home page, they overestimate the favorability of the impression they make on others. In the home page situation, however, this discrepancy is clearly larger: Home page creators particularly overestimate how congenial and how similar they are perceived to be by home page visitors (Sherman et al., 1999; Sherman et al., 2001). The lack of direct (and possibly negative) feedback may lead, in the personal home page scenario, to a stronger positivity-bias serving to protect self-esteem. In addition, the one-sided self-disclosure typical of personal home page communication could play a role (see Wynn & Katz, 1997): When home page owners reveal something private about themselves, they imagine attentive and well-disposed addressees. And even this is not completely illusory, for the WWW is a pull-medium. As a consequence, one can overwhelmingly count on interested home page visitors who express themselves in electronic guest books and emails supportingly and with respect, thereby reinforcing the intended identity construction (see Hevern, 2000, Kennedy, 2000). This form of self-selection of the audience disappears when in an experimental study personal home pages are presented which the subjects otherwise wouldn't have visited. Dominick (1999) showed through a content analysis that just about two thirds of the home page authors wanted to have a congenial effect (ingratiation), while almost a third were primarily concerned with conveying a competent impression (self-promotion). Other strategies of self-presentation (see the “Self-Presentation Theories” section), such as striving to be a role model (exemplification), were observed in less than 10% of the home page sample. These personal impressions composed by the external coders could be contrasted with the home page creators' goals of self-presentation.