The Boulder Valley Internet Project: Lessons Learned

Thirty-five years ago, Everett M. Rogers developed a theoretical framework, based on research evidence, that described the adoption and the diffusion of innovations throughout organizations and social systems. To Rogers, there were five steps in the adoption process. An individual, a department, an organization, or any other type of decision-making unit passes "from first knowledge of an innovation to forming an attitude toward the innovation, to a decision to adopt or reject, to implementation of the new idea, and to confirmation of this decision."[1]

Rogers, like his colleagues in the realm of diffusion scholarship, primarily envisioned an organization as a structured social entity in which power and control in the system was concentrated in the hands of relatively few individuals. In such a system, innovations originate from a centralized source and then diffuse to users.

However, this model is neither appropriate for, nor d'es it capture the complexity of, relatively decentralized systems in which innovations originate from the ground up. In a decentralized system characterized by a site-based decision-making structure, such as the Boulder Valley School District, innovations spread by horizontal communication networks among peer teachers in a relatively spontaneous fashion. A high degree of modification occurs as an innovation is "re-invented" by users to fit their particular conditions.


What sort of innovation are we dealing with here? Throughout the projectís five years, Boulder Valley Internet Project leaders introduced the use of telecommunications in the classroom to the Boulder Valley School District by training. First an initial cohort of 26 teachers were taught how to use the Internet, the local area network, and the World Wide Web for research and professional communication. The initial cohort then returned to their schools and proceeded to share their newfound knowledge and skills with their colleagues, in a "trainer of trainers" process similar to that used successfully in Canada.[2]

Another aspect of the project was development of an online foundation for curriculum-related resources to be made available to all teachers in the district. Though the training program was successful, the impact on curriculum and instruction was less so. To understand the reasons for this, we must delve more deeply into the Rogers' model of diffusion of innovations.

Diffusion of Innovations

In a decentralized system, innovations tend to fit more closely with individual usersí needs and problems. Users seek information through personal networks of colleagues, participate in making decisions about what sort of training and support they would like to see as they learn more about the innovation, and then tailor it to their own specific needs as they begin to develop the expertise, knowledge, and skills to use it effectively. As a result, a decentralized diffusion system is closely geared to local needs. A solution that works for one particular school may not be suitable for another, even within the same school district.

Though horizontal diffusion networks are effective within an individual school, they are far less effective among dispersed schools, even at the same grade level. One of the initial objectives of the project -- to have three networked middle schools collaborate on several shared curriculum units -- was not achieved, partly due to lack of full connectivity, and partly due to the site-based nature of the district.

Another important aspect of the project was that the attributes of the innovation, as perceived by the end users, changed radically during the five-year implementation phase of the project. In the Rogers model, there are five attributes of an innovation. These are perceived by members of the social system in the process of adopting it, and determine its rate of adoption:

  • Five Attributes to an Innovation
    1. Relative advantage:  Is the innovation seen as better than what it replaces?
    2. Observability: Can others see how the innovation works and observe its consequences?
    3. Compatibility: How consistent is the innovation with the values, past experience, and needs of potential adopters?
    4. Complexity: Is the innovation easy to understand, use, and maintain?
    5. Trialability: Can the innovation be tried out on a limited basis?

Unlike the type of innovation conceptualized by Rogers and the diffusion scholars, the Internet was a rapidly evolving technology. At the inception of the project, the system was very complex. Innovators and early adopters had to learn a plethora of Internet tools including gopher, archie, veronica, telnet, ftp, various types of e-mail, and the like. One of the early adopters reflected on her experience with the project:

  • "The main thing with the project is when it was first started, where the technology was then, and where it is now. If I had to start getting back with all those, you know, ftpís and so on! I mean, I love the Web. That is what I live on!"
  • In contrast, by the time the second wave of adopters was trained by their near-peers, the complexity of the system had been vastly reduced through the introduction of graphical browsers such as Netscape Navigator. An innovation that required several training sessions and many hours of hands-on individual exploration by new users could now be mastered by young students in less than an hour.

    Toward an Integrated Adoption & Diffusion Model

    Our study of the adoption and diffusion of Internet- and Web-based telecommunications throughout the Boulder Valley School District, over the past five years, took the form of a full-blown qualitative research study. It included surveys, in-depth interviews, focus groups, a work group, examination of system logs and artifacts, and an embedded case study of a cutting-edge elementary school. Based on Rogersí concepts, we developed a four-module Integrated Technology Adoption and Diffusion Model comprising technological, individual, organizational, and instructional factors. (See Table 1.)

    Throughout the entire duration of our study, we found ourselves addressing many of Rogersí concerns and reflecting on many of his findings. This article is a short summary of our results. A full-scale research paper, The Boulder Valley Internet Project: A Case Study, is available online at

    We were also able to investigate some new aspects of the adoption and diffusion process, such as its spread within a decentralized system and the re-invention process as the innovation itself evolved.

    Critical Elements of the Diffusion Process

    The "trainer of trainers" model met with excellent success, primarily because it incorporated two of Rogersí fundamental concepts: homophily and observability. It also capitalized on the relative advantage of telecommunications, especially in isolated, mountain schools.

    Homophily is "the degree to which two or more individuals who interact are similar in certain attributes."[1] Rather than perceiving the Internet as an innovation introduced from outside, teacher-trainees learned their necessary telecommunications skills in the classroom from fellow teachers or the in-building technical resource person.

    Observability is "the degree to which the results of an innovation are observable to others"[1] -- especially near-peers and colleagues within the same school. Open labs and demonstrations enabled new users to explore the system on their own, observe its effects directly, and evaluate its effectiveness within their own, non-threatening environment. New users often ask questions like "what are the innovationís consequences?" and "what will its advantages and disadvantages be in my particular situation?" In the focus group, one participating teacher reflected:

  • "Well, with the exception of things like Netscape, I pretty much understood the impact it would have on kids. But this whole concept of the World Wide Web with Netscape and the graphics now, moving beyond Lynx and gopher, that is what has been the really, really big change. It is very possible to have full motion videoconferences, to confer with other classrooms anywhere in the world. This is going to be the tough part -- how to get teachers ready for major, major change."
  • Relative advantage is "the degree to which an innovation is perceived as better than the idea that it supersedes."[1] Part of the relative advantage of the Internet was its growth in popularity within the district and its acceptance within the American culture as a whole. Public awareness of the Internet in the last five years has multiplied exponentially. For the past two years, the Boulder Valley Community Network has electronically linked the community with local weather reports, theaters, restaurants, the bus terminal, and other community facilities such as hospitals and social services. One project leader commented,

  • "When we started, the goal was really to see if this technology was going to be worth our time and effort or not. Then as time evolved, more and more people got interested. It almost got to the point where there was no question about whether it would be useful, just because society had already decided it was going to be useful, if nothing else, than as a research tool. There is such a hype about research. It is harder for people than to think about it as a communication tool. My vision is that we can figure out the most creative meaningful uses of technology and how it can support the changes that we are trying to make in schools."
  • While information about a new innovation is usually available from outside experts and scientific evaluations, teachers usually seek it from trusted friends and colleagues whose subjective opinions of a new innovation are most convincing. Since the initial cadre of trainers were chosen from peer teachers, they were able to empathize with their trainees and see things from the new usersí perspective. This resulted in a positive attitude among the potential adopters (teachers), and crucial to the project's success.

    Iterative Design of the Training Program

    A natural outcome of the "trainer of trainers" model is iterative or participatory design. Though not actually part of the Rogers model, it has been an important aspect of performance support systems for over a decade.[3] When a staff development or training program is sensitive to the needs of its typical end users, and when it takes their feedback into account when modifying tutorials, demonstrations and hands-on exploratory sessions to meet stated needs, it tends to meet with a higher degree of success than a fixed, linear-design approach.

    The staff development program of the Boulder Valley Internet Project changed rapidly as the technology evolved and became more user-friendly. Old tools were dropped as the new graphical interfaces became more widespread.

    Creation of the first district Web pages and a revitalized set of training sessions quickly followed the introduction of graphical browsers like Netscape Navigator. Some of the structured classes were replaced by hands-on, open-lab sessions where new users could bring specific questions to the facilitators. These sessions emphasized the simplicity and usability of Netscape for accessing relevant information on the Web that could then be used to enrich instruction. User perceptions became more positive as the project evolved.

    Expansion of Communication Networks

    Originally, the Internet was considered to be a tool for research; its use for communication was only recognized later on. As a result, administrators and teachers were on different networks. Administrators used a secure, local area network, while teachers used the Internet, its e-mail and the usual set of Internet-based tools.

    As newly trained teachers became adept at using the Internet, their peer network slowly expanded beyond their building and their district to include new colleagues. These came from listservs and mailing lists, Usenet newsgroups, experts such as Ask Dr. Math and Ask A Scientist on Web sites, fellow teachers from as far away as Antarctica and Sweden, and even their own students.

    One interviewed teacher mentioned that whenever she ran into a problem with the network, she would get online, find out which of her students were also online, and then send a query to whichever one she felt might be able to answer her questions. Thus, students became not only co-explorers with their teachers, as one might expect in collaborative ventures; they became a vital part of their teachers' support network as well.

    Experimentation and Re-invention

    Since the optimal uses of the Internet in education were undetermined at the beginning of the project, its leaders were flexible and allowed the system to define and be re-invented by the teachers who used it. "Re-invention is the degree to which an innovation is changed or modified by a user in the process of its adoption and implementation."[1]

    Since Web-based browsing was originally perceived as a new tool for research, it soon became an important facet of student research projects. Teachers noted that the new projects presented by their students in fulfillment of class requirements took on an air of originality that was lacking in traditional library-based research. One teacher commented:

  • "In the very beginning it took me three months to figure out e-mail. I mean literally, how do I do this and that? Now it seems so natural for the kids. They can do whatever they do on Netscape. It has opened up so many doors."
  • Trialability, or "the degree to which an innovation may be experimented with on a limited basis"[1], contributed to adoption of the innovation by teachers and students alike.

  • "there is no control, no predictability. If you donít find what you are looking for, the skill is to go back. Can I refine my keyword search? There are some pre-sorting and management skills for bookmarking so that once the kids find a site, they donít have to spend an hour tomorrow finding that site again."
  • However, the project was not without its unintended side effects. Students could easily download sections of Web pages verbatim, cite them in more or less cavalier fashion, and hand them in as their own original work.

  • "We still donít have good ways for kids to document their resources, so now the district is saying, ëmaybe we need to get together.í"
  • Teachers now had to contend with the problem of tracking down the original sources and verifying their authenticity -- often a time-consuming task. They came up with some rather novel solutions that required student performance as well as products. Could the student retrace his/her steps and find the original source? Could the student prove that he/she could actually carry out the procedures documented in their paper, such as calculating elapsed time for various dog teams in the Iditarod race? As student meta-skills increased with their use of the Internet, their teachersí assessment procedures had to change as well. It was a learning procedure for all parties concerned.

    One of the most innovative members of the school at which we conducted the embedded case study enlisted about a dozen multi-age students to join a new Web-based science course geared toward student-generated questions and inquiry learning. Students used the Web not only for accessing information; they also contacted astronauts, Jason Project oceanographic researchers, explorers, and other experts. Building knowledge with their distant colleagues and their own classmates, they created their own Jason Project Web Page and designed a set of projects that were used by the Denver Natural History Museum.

    Collaboration to Gather Resources

    Rogers found that, though the initiation of innovations such as The Jason Project Web Page within a centralized organization is less frequent than within a decentralized organization such as the Boulder Valley School District, centralization may encourage the implementation of innovations, once the innovation decision is made.[1, p380] This is due in part to the administrative vision that is clearly communicated to all stakeholders in a centralized organization, and to the resources that a centralized organization is able to gather in support of the innovation. Clearly, this was not the case in Boulder Valley. There was, and continues to be, little organizational slack (uncommitted resources available to the organization).

    Original funding for the project came through the National Science Foundation. Project directors collaborated successfully with other projects such as the Annenberg/CPB Math and Science Project to provide release time, training and other incentives to newly trained teachers, as well as with the University of Colorado for technical support. As funding continues to be cut within the district, and as in-building technical resource people are reassigned to the classroom as a result, dependence on outside funding and support will continue to be an important factor in the success of the project. An elementary school principal commented on this situation:

  • "When we moved into the building, we all moved into a brand new Mac lab, and there were few people in the building who had ever worked on a Mac before. ëOh my goodness, what is a mouse?í We went through that communal loose bond the first couple of years, managing the computer lab on our own. Then, with planning time, money, the staff running special training, and the Annenberg Project, Internet training was really infused with teacher training. In the past two years, we had planning time like crazy, and we just bailed out and turned to a specialist. That part has come back, and we are all in there doing it, but now we have on-site available training and resources."
  • The school district, seeing the necessity of supporting the cadre of newly trained teachers eager to use telecommunications in the classroom, recently hired both a technical specialist and a computer repair person, at the district level, to respond to the needs of all 53 schools in the Boulder Valley. Though this may be an important step by the district in supporting the project, the lack of empathy between a perceived "district specialist" and the typical teacher g'es against the very core of Rogersí concept of homophily (likeness and empathy) between change agent and client. It remains to be seen how successful this change from reliance upon in-building, peer support to off-site, district support will be, and what its implications will be for the project as a whole.

    Compatibility Issues

    In this case study, we examined the recent changes in the school district that appeared to influence the outcomes of the project. The effectiveness of the adoption of the Internet was impacted by its alignment with the districtís direction and its relationship to major changes that occurred during the projectís implementation.

    In the ecological formulation of the classroom as a concentric arrangement of four nested, interacting systems, i.e., classroom, school, community, and policy making institutions, these systems "are interrelated by ëa common cultural blueprintí that sets the pattern for the structures and processes that occur within and across the systems."[4] Cultural/ecological change processes result from combinations of acting and interacting factors within and across these connected systems. In the Peled at al. model, information technology interventions must aim at not only the individual participants such as teachers and technology resource people, but also at the school as a whole, the administration, the school board and associated policy-making institutions, and the entire expanded ecological environment that characterizes the district as a whole.

    The lack of compatibility between the beliefs, needs and values of the potential adopters and the school board may possibly become a major inhibiting factor to the future use of telecommunications in Boulder Valley classrooms.

    Recently, the political atmosphere within the district has become more conservative, and there is now a general movement away from innovative teaching methods toward a "back to the basics" emphasis on content. This new trend has caused a difference of opinion among members of the school board, resulting in a recall petition for the board's president. A number of central administrators have resigned within the past few months (from January 1997 back), including the superintendent. This turmoil within the district, and especially within the leadership, continues to have important implications for the success of the project. Although both sides of the political constituency currently express support for technology, each sees the role of technology in a different light. The role of the Internet within the district will be affected by their decisions in many ways including usage, financing, staff development, and allocation of equipment.

    Teachers are concerned.

    As less money is available for supporting the staff development so essential for introducing telecommunications tools and resources to the adopting teachers, it will become increasingly difficult to maintain the current level of expertise during the current, rapid evolution of the technology. The cuts will also affect curriculum development, the very area that the project leaders plan to address in the near future.

    Concluding Comments

    Over the span of the five year project, teachersí skills and comfort levels have improved, the Internet technology itself rapidly evolved, and the district gone through a series of challenges related to limited resources and a shift in educational philosophy. Not only is the Boulder Valley School District a complex system; it is also an elaborate, dynamic system, with the adoption of the Internet comprising but one variable among a large number of complex factors that continue to influence educational outcomes.

    Ideally, the adoption of a technological innovation should be part of a larger reform effort, a concerted effort toward common goals. Even if a new vision of instruction is not promoted with the use of telecommunications, the nature and structure of the Internet itself is encouraging teachers to learn and teach in a very different manner.

    A larger, qualitative research report is online at:
    Lorraine Sherry is a research assistant for RMC Research Corp., where she helped evaluate the Boulder Valley Internet Project and the CPB/Annenberg Math and Science Project. She is currently a doctoral student in the Educational Leadership Innovation Program at the University of Colorado in Denver. Primary interest area of research is computer-supported collaborative learning (CSCL). Sherry also serves as Web mistress of the School of Education's Home Page and on the university's Internet Task Force as well. E-mail: [email protected]


    1. Rogers, E.M. (1995), Diffusion of Innovations, Fourth Edition, NY, NY: The Free Press. p. 36
    2. Main, T. (1996), and Teaching Teachers to Use Telecommunications: The British Columbia Experience. [Online]: Path:
    3. Sherry, L., & Wilson, B. (1996), "Supporting Human Performance Across Disciplines: A Converging of Roles and Tools," Performance Improvement Quarterly, 9(4), pp. 19-36.
    4. Peled, A., Peled, E., & Alexander, G. (1994), "An Ecological Approach for Information Technology Intervention, Evaluation, and Software Adoption Policies," in E.L. Baker & H.F. OíNeil Jr., (Eds.), Technology Assessment in Education and Training, 35-61. Hillsdale, NJ: Erlbaum & Associates, p.37.
    5. Sherry, L., & Lawyer-Brook, D. (1997), The Boulder Valley Internet Project: A Case Study. [Online.] Path:

    This article originally appeared in the 09/01/1997 issue of THE Journal.