The ongoing convergence of telecommunication media presages a "communications" or "information" revolution that is based on collecting, storing, processing and communicating information. At the center of these converging media lies the Internet, which merges the functions associated with the previously distinct media of telephony, TV, publishing and computing.
Nationwide interest in this emerging "information superhighway" has been spurred by Executive and Legislative branch initiatives to remove regulatory barriers between voice, video and data providers, designed to facilitate a nationwide integrated digital network (Information Infrastructure Task Force, 1993; Telecommunication Act of 1996). The Internet has been heralded as everything from a workplace revolution to the greatest advance in human evolution since, perhaps, the move from single to multi-cellular body structures. Preliminary reports indicate that by 1995, 8 million Americans used their computer to telecommute, and over 20 million accessed the Internet each week (Lewis, 1995).
As the information superhighway comes to fruition, we have only a crude understanding of who uses the Internet, why, and for what purposes. To gain a better understanding of this dynamic, the present study profiles Internet users in terms of social locators, media use habits, communication needs and attitudes toward telecommunication technology adoption.
According to diffusion theory, adoption of technological innovations is a function of one's innovativeness, or willingness to try new products. Thus, if we consider Internet service as an "innovation," diffusion theory may offer clues about those who are relatively early to adopt it.
Rogers with Shoemaker (1971, p. 27) define innovativeness as "the degree to which an individual is relatively earlier in adopting an innovation than other members of his social system." Although this is but one of many competing definitions of innovativeness (e.g., Midgley & Dowling, 1978)(1), scholars have yet to account fully for the psychological dynamic driving technology adoption. For that reason, diffusion research may not provide the predictive power associated with other "theories," hence the term "diffusion framework" may be more appropriate.
In building a theory of online service use, the dearth of research on Internet adoption necessitates consideration of a wider literature addressing new media adoption (e.g., Lin & Atkin, in press). For instance, the virtual requirement of computer ownership (and operational skills) for Internet adoption renders that literature relevant to the present study.(2)
Broadly speaking, diffusion research addresses the characteristics of innovations and those who adopt them. Focusing on the former, Rogers with Shoemaker (1971) initially distinguished between continuous innovations -- those representing a variation of existing channels -- and others which are more discontinuous (i.e., more difficult to adopt, perhaps involving the purchase of a separate piece of hardware).(3) Building on that framework, Krugman (1985) proposed a dynamically discontinuous category to reflect innovations (e.g., VCRs) that require a specific purchase and a dedicated set of user skills. Studies of computer adoption (e.g., Lin, 1998), for instance, suggest that computers are perhaps the most discontinuous of media technologies, given the relatively high financial and skill (education) barriers associated with their adoption. Since the same study uncovered a link between computer adoption and intention to use online services, the Internet can also be characterized as a dynamically discontinuous innovation.
Focusing on characteristics of adopters, the diffusion framework offers corollary perceptions of innovations, such as "ease of use" (Rogers, 1995). In fact, the very existence of a literature on "computerphobia" (e.g., Atkin, 1995b; Lin, 1994a) attests to high levels of perceived complexity associated with such information technologies as the Internet. The diffusion tradition also categorizes people in terms of social locators, adoption/uses of other technologies, and attitudes toward adoption, which are addressed in turn.
Since little work has directly addressed Internet adoption, it may be useful to investigate prototypical information services (Garramone, Anderson, & Harris, 1986). Several reviews of the adoption literature indicate that demography is associated with new media adoption and use behaviors (Atkin & LaRose, 1994b; Dutton, Rogers, & Jun, 1987a,b; Krugman, 1985; Rogers, in press), as adopters tend to be upscale, better educated, and younger than nonadopters. This has been found to be true of computer adopters as well (Dickerson & Gentry, 1983; Lin, 1998). Although only about a third of the 35% of U.S. homes with a computer use the Internet, these households tend to be younger and better educated (e.g., Sanberg, 1996). Other market profiles (e.g., O'Reilly, 1995) note the existence of a gender gap in use of online services, as two-thirds of users in one survey were men.
As Rogers' (1995) typology predicts, demographic differences between adopters and nonadopters have been leveling for more "mature" media, including cable, VCRs and other technologies that have reached the "flat" part of their diffusion curve (Atkin & LaRose, 1994b; Lin, 1994b; Sparkes & Kang, 1986). Perhaps because the Internet is still in early stages of diffusion, past work points to an upscale early adopter profile. However, Jeffres and Atkin (1996) found that income and education bear only a weak inverse relationship with interest in adopting specific Internet utilities (e.g., sending or receiving messages, ordering goods). They conclude that those applications may be less expensive substitutes for functions performed by traditional media, such as the telephone, and that communication needs were more explanatory than social categories.
The media substitution hypothesis (Krugman, 1985; Jeffres et al., 1995; Lin, 1994b) suggests that the introduction of a new medium encourages a restructuring in the way consumers view established media. Although displacement effects have been widely reported for the impact of television on radio, and that of new video media on television, the scarce literature on computer adoption provides no clear indication of the impact of online service use on traditional media channels. For instance, James, Wotring and Forrest (1995) found that the use of electronic bulletin boards reduced time spent with television viewing, book reading and telephone use. Similarly, Vitalari et al. (1985) found that computer users spent less time with TV, books, phones and leisure.
However, focusing on functionally similar videotext services, Heikinnen and Reese (1986) discovered that newspaper reading did not discriminate in interest for potential videotext news adoption. Lin (1994a) discovered a similar pattern of "noneffects" for videotext use on other media. Jeffres and Atkin (1996) also found the use of online services was generally unrelated to use of other media, although positive relationships with TV viewing emerged.
New Media Adoption
Research indicates that the adoption of new text services is related to the adoption of other innovations (Ettema, 1984; Lin, 1994a; Jeffres & Atkin, 1996): namely, that as experience with technology encourages adoption of cable, audiotext and computer media (e.g., LaRose & Atkin, 1988, 1992; Reagan, 1989b). Reagan (1987) found that adoption of a given media innovation is most powerfully related to adoption of other technologies; such as videotext, PCs, CDs and cable. Focusing on other dimensions of compatibility, Rice's (1993) study of media adoption in organizations placed media into two categories: traditional (e.g., telephone) and new media (e.g., electronic mail).
LaRose and Atkin (1992) applied Rogers' notion of technology clusters to the adoption of phone-delivered information services, noting that use of audiotext was related to functionally similar information services such as videotext and 1-900 numbers (see also Neuendorf et al., 1998; Yankee Group, 1988). Reagan et al. (1995) expanded that notion to include adoption of functionally similar technology repertoires.
Such a collection might be stimulated by acquisition of a "trigger" innovation (Dozier et al., 1986), such as a computer, which encourages adoption of related technologies. Lin (1998) noted that computer adoption was related to Internet adoption intentions as well as a technology adoption index (comprised of 14 telecommunication media). In their analysis of audience demand for popular online services, Jeffres and Atkin (1996) maintain that scholars should shift their focus away from technological hardware and toward communication needs.
Although some inconsistent findings on adopter profiles may be due to varying methodologies and service packages studied, it is likely that demographics are imperfect surrogates for deeper motivational variables that drive Internet adoption. Research on technology adoption indicates that audience needs are primary determining factors (Neuendorf et al. 1998). For example, in exploring the nature of VCR-use motives, Rubin and Bantz (1987) discovered that perceived utilities are the primary motives behind. VCR-use decisions.(4) As past work (e.g., Lin, 1994b) suggests, these motives all reflect a fundamental psychological element -- the desire to exert control over one's content environment.
Investigating technology adoption in light of user needs, Ettema's (1984) study of agricultural videotext found that adopters were more interested in market updates than in news. Lin (1994a) uncovered interest in a somewhat wider scope of videotext services (ranging from voting to energy management and games)...