Abstract. This paper uses critical mass theory to explain how a community network might achieve sustainability. Critical mass theory examines the processes by which interactive communications systems, e.g. Internet-based community networks, become sustainable. The present paper explores the characteristics, motivations, and preferences of "pre-critical mass" adopters by comparing a sample of 1505 individuals who joined the Blacksburg Electronic Village (BEV) community network in the first year of operation with a sample of 1344 individuals who joined in the second year (N=2553). Responses were measured on many variables thought to be of importance in describing how critical mass may be achieved: computer experience, knowledge of the innovation, media use, and a variety of demographic variables. The results of MANOVA analyses are consistent with those posited by critical mass theory about pre-critical mass users of the community network. This research suggests that the presence of a large group of high-resource, motivated early adopters are sufficient for critical mass if they are supported by the designers of the communication initiative. The results are discussed in terms of the implications of critical mass theory for designers of community computer networks.

Introduction

The flowering of the World-Wide-Web and the emergence of Internet-based community networks since the early 1990s has generated much interest about the kinds of persons who adopt these innovative technologies and services, and about the optimal conditions under which these initiatives may become self-sustaining and widely used. By 1996, about 35% of U.S. households used a personal computer, and about 14% were connected to the Internet. During that same year in the rural, university town of Blacksburg in southwest Virginia, 50% of households used personal computers, and 44% of the population were directly connected to the Internet at home, school or work. These dramatic increases resulted from a project called the Blacksburg Electronic Village (BEV), a partnership between the Town of Blacksburg, Virginia Polytechnic Institute & State University (also known as Virginia Tech), and Bell Atlantic.

This paper seeks to discover if community computer networking projects follow the pattern of adoption and growth hypothesized by the literature on interactive communication technology (see especially Markus, 1987, 1990). The results of this study are important for other community networking initiatives. If the Blacksburg Electronic Village model helps to create a sustainable community network, other community networks can build on the successes of the BEV. The essential question this paper poses is who should system designers “target” as the pre-critical mass user of an Internet-based community network?

Theoretical Foundation

Sterling (1995) suggests that a sustained diffusion process begins when a technology merges into many persons' everyday lives. Should a technology become important in the daily lives of enough individuals that technology will continue to be sustained even if current users leave the system and new users join. For example, the telephone became a self-sustaining technological system once it became a wide spread household object, important to the everyday lives of its users.

Yet, other somewhat similar technologies failed in their attempts at sustainability. Consider the failed attempts by telephone companies to create a critical mass of adopters for video-telephones. Those video phone technologies were not sustainable. The importance of sustainability is clear. Should a technology, or technological system, fall short of sustainability, then early adopters of the technology loose the resources they have invested and the technology ceases to become part of their everyday lives.

Geographic communities throughout the world are developing local identities and access points to the Internet. Projects such as the Telluride Info Zone, Seattle Community Network, Capital Free Net and the Blacksburg Electronic Village have developed both significant on-line identities and a means of facilitating Internet access among the members of their communities. Given the substantial amount of collective and individual resources invested, sustainability is essential. Sustainability is especially vital for interactive communication systems where early users of the system are rewarded for their risks only if long-term universal access and use occur.

Community computer networks may be conceptualized as a public good. A public good is some activity, event, or product that benefits not only the people involved in the creation of the innovation, but other members of the community as well (Rafaeli & LaRose, 1993). For example, a concerned group of parents may invest time and money (resources) in the building of a new swing-set in a local park. The parents (and their children) who invest in this swing-set project benefit from their efforts. The public good nature of the swing-set project is realized by people who don't spend resources in the creation of the swing-set but still utilize the innovation after its completion (free-riders - See Marwell & Oliver, 1993).

For a community network, the objective is enhanced local and global multimedia communications. Community computer network organizers expect that the addition of Internet services to the existing communication milieu would increase the quality of life and the level of community involvement (Kavanaugh & Patterson, in press). These changes would be spurred by access to information and individuals, locally and globally and enhanced delivery of education, government, commercial and other information services.

In order for the pro-social effects of a public good to occur, the system must reach critical mass or become sustainable. Critical mass represents a kind of "tipping point or social threshold" in the diffusion process (Rogers, 1995), what Valente (1995, p. 79) calls "a system-level measure of the minimum number of participants needed to sustain a diffusion process." Markus (1990, p 103) defines critical mass in a telecommunications context:

This definition of critical mass has two important implications. First is that critical mass may be considered as a process of building sustainability rather than as a static point in a diffusion process. Second is that the sustainability of interactive communication technologies depends upon persons beyond the initial adopters taking advantage of the public good. Indeed, Markus further speculates that critical mass for interactive communication systems is an all or nothing phenomena:

For Markus, the system doesn't reach critical mass unless there is universal diffusion of the technology through the user community (see also Rice, Grant, Schmitz, and Torobin, 1990). Others argue that the presence of early adopters may be sufficient for the existence of critical mass (Griliches, 1957; Valente, 1995). Whether or not the earliest users represent a pre-critical mass or the entire critical mass necessary, early users provide the necessary foundation from which the critical mass emerges.
Critical mass theory (Markus, 1987, p. 496) posits sustainability hinges on the interactions among two key concepts: 1) the nature of the production function and, 2) the diversity and availability of resources and interests invested in the innovation.

The Production Function Of Interactive Technologies

The key to the diffusion of interactive communication systems like community computer networks are their accelerating production functions (Marwell and Oliver, 1993; Markus, 1987). An interactive technology is of little to no value for the first adopter. What good is a telephone if no one else has a telephone? The telephone owned by the first person becomes valuable only if “desirable” others acquire telephones and connect them to the network. Markus refers to this phenomena as an "accelerating production function."

Technologies with accelerating production functions are characterized by relatively large start-up costs and reciprocal interdependence. Large start-up costs exist because the network must be built and support structures must be in place before users can adopt the technology. Systems with accelerating production functions rely on key players, or builders, to provide the initial resources (time, money, expertise) necessary to build the system. Without these key players, technologies with accelerating production functions can never attract their first adopters. Once the first adopters begin to use the interactive system, reciprocal interdependence occurs.

Reciprocal interdependence characterizes relationships such that early adopters increasingly benefit as new people join the system. The system becomes exponentially more valuable because of the increase in possible connections among users (Rice, 1997). Valente (1995) refers to this as an outcome of the bandwagon effect. The bandwagon effect describes the degree to which individual adoption (or discontinuance) decisions are influenced by their perceptions of others. Thus, a community computer network typically anticipates that early adopters of the system will spread the word about the utility of the system and encourage members of their personal communication networks to join. As these established partners become increasingly accessible on the system, the system becomes more a part of the everyday lives of its users. It is this consequent integration of a new technology into the everyday lives of its users that constitutes a key to sustainability.

Diversity And Availability Of Resources And Interests

The other element in the application of critical mass theory to community computer networks is the assessment of heterogeneity and availability of resources in the community. Resources are conceptualized as being of one of four types: Communication discipline resources, access resources, information resources, and personal information seeking resources. Sustainability is likely in interactive communication systems where the structure of the innovation and its initial adopters conform to expectations about the nature and availability of these four resources (see Markus, 1987 pp. 502-504).

Communication discipline refers to the willingness of individuals to communicate to others via the community network. If the users of a system regularly log-in or actively use the system, they can be said to exhibit high levels of communication discipline. Communication discipline is important for interactive communication systems because if enough people don't answer the phone or reply to e-mails, the system fails. Discipline in this sense is not punitive but rather refers to the ability and desire of people to use a technology for interaction with others.

Access resources can be conceptualized as the general level of accessibility of the system. Hudson (1984) describes four components of system accessibility: physical distance from an access point, cost of access, quality of service, and user knowledge and expectations. In critical mass terms, the less the physical distance from a point of access, the lower the cost of access, the higher the quality of service and the higher the user knowledge and expectation, the greater the probability of achieving critical mass.

Markus's third resource category relates to the kinds of information available via the communication network. The greater the diversity and "value" of the information provided on the network, the greater the probability of the system reaching critical mass. Information on community computer networks is conceptualized both as the materials made available from computers connected to the system as well as the other persons connected to the system.

Markus defines high resource individuals as individuals who are "sought after" by other persons as communication partners. Just what characteristics define a "sought after" person is not elaborated in critical mass theory. However, the conjunction of the opinion leadership construct and diffusion theory provides us with expectations about the attributes of individuals who are early adopters of an innovation. In general, opinion leaders are persons who are "sought after" communication partners (see Katz and Lazarsfeld, 1964). Diffusion theory posits that early adopters of an innovation are characterized by higher levels of education, social status, cosmopolitaness, greater exposure to media channels, and knowledge of the innovation, than later adopters and non-adopters (Rogers 1995). Early adopters are also opinion leaders, who communicate their experiences with an innovation to members of their social networks. This opinion leadership is highly influential in the diffusion process (Valente, 1995).

The final resource category, one that also helps to determine the formation of critical mass, relates to the information seeking characteristics of those individuals who are connected to the system. Markus (1987) suggests that the more users actively seek and use information, the greater the probability that a system will achieve critical mass (see Patterson, 1996 for a detailed discussion).

The Blacksburg Electronic Village Case Study

The Blacksburg Electronic Village (BEV) is an experiment in the development and deployment of an Internet-based community network. This community network began offering Internet access and local information and services in 1993 to Blacksburg and surrounding Montgomery County, Virginia. One of the original purposes of this project was to build critical mass among users through extensive local online content and computer-based communication.

As of 1998, within the town of Blacksburg (population 36,000) an estimated 80 percent of the residents are linked to the Internet at home, work, or school. Kavanaugh, Cohill, and Patterson (2000; see also Kavanaugh & Patterson, in press) present recent data about the diffusion of the computer network and the Internet within the community:

• Most of Blacksburg Internet user population (85%) is affiliated with the local university, Virginia Tech; 58% of users are students, 27% faculty; and 15% staff.
  
• Average age of users is 36, but all age groups are represented from elementary school to the retired.
  
• Most users (60%) in the Blacksburg Electronic Village are male. This proportion is consistent with national averages, where the majority of users are male (Falling Through the Net, 1999). But among users at the public library, there is about a 50-50 split between male and female users.
  
• By 1998, over two-thirds of local Blacksburg businesses were online.

In addition to standard Internet services, BEV services include information about local town and county government, social services, public education, libraries, health care, and community organizations as well as local town and county government web-based forums.

The most desirable outcome for the BEV is the creation of an interactive system that all members of the community can use to facilitate the activities of their daily lives. Detailed histories of the BEV project are available from other sources (see Cohill & Kavanaugh, 2000) and serve to document the cost considerations guiding the initial design and implementation of the initiative. Working from the premise of an accelerating production function, the value of the BEV should increase as more individuals join the system. Because early adopters formed the critical mass of users necessary for the sustainability of the BEV , it is critical to ask: Who were the early adopters of the Blacksburg Electronic Village?

Methods

During the first year of BEV operation, 2294 persons acquired accounts via one of the two BEV public registration sites. Of these persons, 1505 (65.6%) completed and returned the voluntary BEV background questionnaire. In the second year of operation, an additional 1344 persons registered for a BEV account and 1048 (78.0%) returned the background questionnaire.

The sample for this research consists of the 1505 questionnaires returned at the end of year one (November 1, 1993 to October 31, 1994) and the 1048 questionnaires returned during year two (November 1, 1994 to October 31, 1995). This combined sample of 2553 questionnaires returned during the first two years of BEV operation permits detailed comparison of first year and second year registrants.

The survey instrument for year one was a brief paper and pencil inventory consisting of 36 items designed to measure the resource attributes of early adopters of the BEV. The second year instrument was altered to reflect the results of the open-ended questions from the year-one survey and to reflect changes in the services offered by the BEV. Details about the design and implementation of the background questionnaire can be found in Patterson (2000), Patterson & Kavanaugh (1995), and Patterson, Bishop and Kavanaugh (1994). Copies of both instruments and the complete data set are available via the BEV Research Site (http://www.bev.net/research/).

Results

Generally, the results of the analyses indicate support for our expectations about the characteristics of the pre-critical mass users of the Blacksburg Electronic Village. Descriptive statistics for all variables are provided in Table One. In the results presented below, evidence of any differences over time among the variables was obtained via MANOVA conducted with time (1 or 2) as the independent variable. For significant multivariate models, follow-up one-way repeated measures analysis of variance were conducted. Because of the power inherent in the large sample size employed in this research, univariate results were considered statistically significant if they obtained an alpha level of .01 or less. In all cases, particular emphasis was placed on results that represented at least modest absolute magnitudes in addition to statistical significance.

Reciprocal interdependence.

Critical Mass theory suggests the social networks of early adopters of the Blacksburg Electronic Village would spread the word about this innovation if the BEV was to achieve sustainability. The background questionnaire allowed some rough measurement of the role social networks play in the diffusion of the BEV. A great number of first year registrants were affiliated with Virginia Tech, a primary partner driving the growth of the BEV (see Table One). Since BEV operates under an accelerating production function, persons affiliated with one of the main builders of the BEV are more likely to be influenced to join the innovation.

============================================================

============================================================

In the year two instrument, two questions were added to specifically measure the role of social networks in joining the BEV. These data are summarized in Table Two. In year two, registrants were most likely to first hear about the BEV from family and friends or coworkers. Seventy-eight percent of the second year registrants indicated they had heard or read about the BEV in the local media, even if this was not where they first heard about the BEV. Finally, 34% of the second year registrants knew six or more other BEV members, while 86% knew at least one other BEV member at the time of registration.

============================================================

============================================================

The preliminary social network data collected in this paper indicate that interpersonal, group, and organizational communication channels are important variables in understanding the diffusion of the BEV (see Table 2). This "social network diffusion" and the corresponding reciprocal interdependence is an important way to extend this research into the critical mass process of community computer networks.

Heterogeneity of resources and interests

Communication discipline was conceptualized as the sustained willingness to use the technologies of the BEV. In the first year, registrants indicated they were most eager to try electronic mail, followed by accessing information, news groups, and accessing libraries (see Table 1). In year two, electronic mail was still the service most respondents were eager to try, followed by the World-Wide-Web, accessing information, and news groups. Year two registrants were more eager that earlier adopters to try the Blacksburg Electronic Village. Year two registrants were especially eager to try electronic mail and then to access information via the BEV. The desire to try electronic mail is especially important, because the sending and receiving of electronic mail requires a high level of communication discipline. Respondents’ preferences for e-mail also supports the importance of interactivity in building critical mass.

Access resources were quite high for both years of registration. Of the 2553 registered members of the BEV, 96% own a personal computer. The majority of respondents have used a computer for between one and three years and have access to more than one computer. Statistical tests indicate that year one registrants reported significantly higher levels of computer experience than those who joined later. While ownership and experience with computers were important characteristics of the critical mass, it is interesting that experience with computer networking was not significantly different over time. The implication is that lack of direct experience with this innovation (community computer networks) may not have a substantial impact on access.

The analysis of the kinds of people and information available on the network focused on the demographic characteristics of the early adopters and on the kinds of activities for which they hoped to use the BEV. The majority of the registrants were between the ages of 23 and 30 and ranged from the ages of 12 to 83; most respondents were male (67%); most registrants subscribed to cable television (67%) and read a local newspaper (74%). The majority of the registrants had at least some college education and reported a median total household income that ranged from $25,000 to $40,000 (see Table 1). Over time, results indicate that significantly more females registered in year two than in year one. Year one registrants reported higher levels of education and a significantly lower household income than did year two registrants. However, none of these demographic differences were substantively significant (eta-squared < .005). Contrary to the implications of critical mass theory relative to cosmopolitaness and media use, year one registrants were significantly less likely to read both national and local newspapers than were year two registrants.

Registrants were also asked how helpful they hoped the BEV would be for a variety of activities (see Table One). In the first year, registrants indicated they thought the BEV would be most helpful for improving social relations, followed by help with work, knowledge about civic affairs, and learning activities. In year two, the pattern of expectations changed such that year two registrants thought the BEV would be most helpful for learning activities, followed by gaining computer experience, social relations, local news and information, business, and civic affairs. Expectations of the BEV were significantly higher among year two registrants than year one registrants, especially related to the possibility for the BEV to help with consumer affairs (e.g. home shopping and e-commerce) and learning activities (e.g. distance education).

Discussion

In general, critical mass theory appears to be applicable to community computer networking initiatives such as the Blacksburg Electronic Village. The present data support the expectation that diversity and availability of early user resources are essential to network sustainability. Especially important are the critical mass “predictions” that early adopter social networks are essential to develop sustainable community networks.

Returning to our initial question, who should system designers target as the necessary 'pre-critical mass user' of an Internet-based community network? The evidence gathered in this research project indicates that many conditions critical mass theory identifies provide good guides. Specifically, the early adopters of community network tend to be young (under 30), educated, experienced users, and owners of computers. The critical mass of users also share some interpersonal or organizational connection with the designers, builders or key players behind the network. Experience with computer networks seems less vital than experience with computers.

Services and uses that take advantage of the accelerating production function (reciprocal interdependence) appear to be the key services to track in following the emergence of critical mass. Evidence gathered in this research project indicates the primacy of the expectation that the BEV community network will help with social relations. This expectation, coupled with the finding that e-mail is the service users have been most eager to try at the time of registration, leads to the conclusion that other community networks should follow the suggestions of critical mass theory – thus they should emphasize and cultivate the interpersonal and group communication capabilities of their systems.

Are the initial adopters of communication systems sufficient for sustainability? Among the residents of Blacksburg, a remarkably oversized proportion joined the BEV as initial adopters. Because so many people joined early on, the initial adopters had to matter. Markus indicates the benefits of an interactive communication system are only obtained if universal access and use is achieved. Markus posits that a system like the BEV is likely to achieve critical mass if its early adopters have a large and diverse array of resources and interests they bring to the system. These “sought-after” early adopters, if they continue to use the system, will attract new users to the system. In the Markus approach, this process marches inevitably toward universal diffusion if the system is successful.

At the end of the second year of the BEV's existence, fully 44% of the community were connected to the network from their homes and an additional 10% were using from public access locations (Patterson, 1996). However, diffusion in Blacksburg, while still well over national averages, has never reached “universal use.” Internet use in Blacksburg hovers around 85% of the population (Kavanaugh, Cohill, & Patterson, 2000). By all accounts, however, the BEV is a sustainable interactive communication system. Perhaps universal adoption is not the end state of critical mass as posited by Markus. This research suggests that the presence of a large group of high-resource, motivated early adopters are sufficient for critical mass if they are supported by the designers of the communication initiative.

BEV designers and the BEV project relied on inputs from users to shape the communication system from the start. This emphasis on evaluation became an important vehicle for system designers to adapt the BEV to the resources and interests of the people who used it. Markus refers to this phenomena as “interventions from system designers” (p . 504). We believe this user-centered design process goes well beyond an intervention and enters the realm of co-creation. It could be that interactive communication technologies such as community computer networks can achieve sustainability if designers remain involved with users and continue to adapt the system over time. This co-creation seems essential if a community computer network is to become a household object, one important in the daily lives of community members.

*Author's Note: A previous version of this paper was presented at the annual conference of the Association for Education in Journalism and Mass Communication. The authors would like to thank Ron Rice, Joe Schmitz, and several anonymous reviewers for comments on this manuscript.

Cohill, A. & Kavanaugh, A. (Eds.). (2000). Community networks: Lessons from Blacksburg, Virginia (2nd ed.). Norwood, MA: Artech House.

Griliches, Z. (1957). Hybrid corm: An exploration in the economics of technical change. Econometrica, 25, 501-522.

Hudson, H. (1984). When telephones reach the village: The role of telecommunications in rural development, Norwood, MA: Ablex.

Katz, E., and Lazarsfeld, P. (1955). Personal influence. New York: Free Press.

Kavanaugh, A., Cohill, A., and Patterson, S. (2000). The use and impact of the Blacksburg Electronic Village. In A. Cohill & A. Kavanaugh (Eds.) Community networks: Lessons from Blacksburg, Virginia (2nd ed., pp. 77-98). Norwood, MA: Artech House.

Kavanaugh, A. & Patterson, S. (in press). The impact of computer networking on social capital. American Behavioral Scientist.

Markus, M. L. (1990). Critical mass contingencies for telecommunications consumers. In M. Carnevale, M. Lucertini, and S. Nicosia (Eds.) Modeling the innovation: Communications, automation and information systems (pp. 103-112). North-Holland: Elsevier Science Publishers.

Markus, M.L. (1987). Toward a "critical mass" theory of interactive media: Universal access, interdependence and diffusion. Communication Research, 14(5), 491-511.

Marwell, G. and Oliver, P. (1993). The critical mass and collective action: A micro-social theory. Cambridge, England: Cambridge University Press.

Olsen, M. (1965). The logic of collective action. New York: Schocken Books.

Patterson, S. (May, 1994). Gratifications sought from the electronic village. Paper presented to the annual conference of the International Communication Association, Albuquerque, New Mexico.

n Patterson, S. (1996). Interpersonal communication and Internet telephony. Austin, TX: SW Bell Telephone Company.

Patterson, S. & Kavanaugh, A. (1995) Rural users expectations of the information superhighway. Media Information Australia.

Patterson, S., Bishop, A., & Kavanaugh, A. (1994). Preliminary evaluation of the Blacksburg Electronic Village. Washington, DC: Council on Library Resources.

Rafaeli, S. & LaRose, R. (1993). Electronic bulletin boards and "public goods" explanations of collaborative mass media. Communication Research, 20(2), 277-297.

Rice, R. (1997, July 9). Personal communication.

Rice, R. (1990). Computer-mediated communication system network data: Theoretical concerns and empirical examples. International Journal of Man-Machine Studies, 32(6), 627-647.

Rice, R., Grant, A., Schmitz, J., & Torobin, J. (1990). Individual and network influences on the adoption and perceived outcomes of electronic messaging. Social Networks, 12, 1-29.

Rogers, E. (1990). The "critical mass" in the diffusion of technologies. In M. Carnevale, M. Lucertini, & S. Nicosia (eds.), Modeling the innovation: Communications, Automation, and Information Systems (pp. 79-94). North-Holland: Elsevier Science Publishers B. V.

Rogers, E. (1995). Lessons for guidelines from the diffusion of innovations. The Joint Commission Journal on Quality Management, 21(7), 324-328.

Rogers, E. (1995). Diffusion of Innovations (4th ed.). New York: The Free Press.

Sterling, B. (1995). The hacker crackdown: Evolution of the US telephone network. In N. Heap, R. Thomas, G. Einon, R. Mason, and H. Mackay (eds.), Information technology and society: A reader (pp. 33-40). London: Sage.

Valente, Thomas (1995). Network models of the diffusion of innovations, New York: Hampton Press.

---

Copyright 2001 Communication Institute for Online Scholarship, Inc.

This file may not be publicly distributed or reproduced without written permission of the Communication Institute for Online Scholarship, P.O. Box 57, Rotterdam Jct., NY 12150 USA (phone: 518-887-2443).

---

CIOS Support Staff

support@cios.org