Diffusion of InnovationsExcellent Teaching and Early AdoptionThe InnovationRogersí Diffusion Theory: Related Contexts
The Innovation-Decision Process
Adopter CategoriesInnovators (INs)
Early Adopters (EAs)
Newbies and Enthusiastic Beginners
Early Adopters of Instructional Technology
Addressing a Gap in Rogersí (1995) Theoretical FrameworkDeveloping Long-Term Plans for Campus-wide DiffusionGlobal Characteristics Versus Personal StoriesLinks Between Diffusion Research and the Present Investigation
Case Studies of Expert Teachers
Motivation To Become Expert at Integrating Technology
Critical Mass and the Chasm Between Early Adopters and MainstreamPresent Investigation
Institutions as a Change Agencies
An Application of Rogersí Innovation-Decision Process(1) Knowledge of an Innovation and (2) Persuasion to AdoptAlternatives to Campus-Wide Plans That Build From Pioneers
(3) Making a Decision to Adopt or Reject, and (4) Implementation
(5) Confirmation of Decision to AdoptRely on Natural Diffusion Patterns
Rely on Cross-Disciplinary Research
The academic literature related to using technology for teaching and learning is large and diverse. This section will review the literature on the adoption patterns and characteristics of faculty who integrate technology. The first topic to be examined is the framework provided by Rogersí (1995) theory of the diffusion of innovations. Focus will be on how adopter categories and the innovation-decision process are conceptualized, and how these concepts can be applied in an investigation of faculty adoption patterns. The second major topic is an examination of the development of teaching expertise, and the potential relationship between early adoption and excellent teaching. Rogersí (1995) theory and a consideration of the development of expertise provide a rationale for the chosen research methodology. The third major topic embodies a review of the literature on the implications of developing long-term plans for campus-wide adoption of technology that are based on the characteristics of early adopters. There is growing recognition of the need to provide a different support infrastructure for mainstream faculty than sufficed for early adopters of technology for teaching and learning. A number of system-wide initiatives have been implemented at various higher education institutions which provide models for encouraging wider diffusion of technology for teaching and learning, and bridging the gap between early adopter success and more mainstream adoption.
Descriptions in the literature suggest that faculty who are innovators or early adopters of instructional technology for teaching and learning are intrinsically motivated, self-taught, "lone-wolves" and experimenters (Wertheimer & Zinga, 1997), who are confident and efficacious (Rogers, 1995), comfortable with constant change, attracted to the technology, risk takers (Gilbert, 1995), and excellent teachers whose use of technology appears to be a natural extension of their area of expertise (Hadley & Sheingold, 1993). Individuals in this group have often used technology to "reengineer" (Hammer & Champy, 1993), or transform the teaching-learning transaction, thus changing teacher and student roles (Roblyer, Edwards, & Havriluk, 1997). What differentiates the early adopter of instructional technology from other faculty members?
Diffusion of innovations theory (Rogers, 1995) provides an approach to discussing the differences between early adopters and others. The importance of a theoretical framework is rooted in the cycle of knowledge development: observations lead to theory to classify, explain, and predict observations (Davis & Parker, 1997). The theory leads to questions about the behavior or actions being observed. Theory-based research defines expected outcomes and the variables associated with them, and provides a reason for expecting to find certain results. Research provides evidence for or against the theory-based expectations.
A theoretical framework for analyzing of the characteristics of adopters is provided by Everett Rogersí (1995) theory of the diffusion of innovations. A majority of Rogersí (1995) studies have investigated discontinuous innovations, and technological innovations fall into this category. "A technological innovation usually has at least some degree of benefit for its potential adopters. This advantage is not always very clear-cut, at least not to the intended adopters. They are seldom certain that an innovation represents a superior alternative to the previous practice that it might replace" (Rogers, 1995, p. 13). Therefore, there is a good match between Rogersí (1995) theory of the diffusion of innovations and the specific innovation of interest in the present investigation. Rogersí (1995) defines diffusion as "the process by which an innovation is communicated through certain channels over time among the members of a social system" (p. 5). The four main elements are the innovation, communication channels, time, and the social system. The following sections elaborate upon these four elements with a focus on the innovation, the innovation-decision process, and adopter categories.
Rogers (1995) defines an innovation as an idea, practice or object that is perceived as new by the individual, and diffusion as the process by which an innovation makes its way through a social system. The characteristics of innovations, as perceived by individuals, tend to influence their rate of adoption and are associated with the persuasion stage of the innovation-decision process. Rogers (1995) lists five characteristics of innovations. Relative advantage describes the degree to which an innovation is perceived as better than that which it supersedes. Compatibility is the degree to which an innovation is consistent with the existing values, past experience, and needs of the potential adopter. Complexity is the degree to which an innovation is perceived as difficult to understand and use. Trialability is whether an innovation may be experimented with on a limited basis. Observability is the degree to which the results of an innovation are visible to others.
The innovation in the present investigation is instructional technology, and diffusion is the extent to which all faculty on campus have adopted this innovation for teaching and learning. An important conceptual and methodological issue is to determine the boundaries that define a technological innovation (Rogers, 1995). Therefore, instructional technology in this investigation includes 44 types of computer-based software and tools used for teaching and learning, and only implies the hardware on which these run, the peripherals needed for CD-ROM, laserdisk, an so on, and the network infrastructure for communications technology. Thus, presenting information available on the World Wide Web in the classroom using a projector would be considered using instructional technology, as would displaying PowerPoint slides using a projector.
An investigation into the adoption of computer technology must also consider the concepts of interrelatedness and re-invention. A technology cluster consists of one or more distinguishable elements of technology that are perceived as being closely interrelated. Rogersí (1995) asserts that an adopterís experience with one innovation influences that individualís perception of the next innovation in a technology cluster to diffuse through the individualís system. Thus, if an adopter has a negative first experience with one computer application, they may regard all computer applications through this perspective. Rogersí (1995) also describes how diffusion research has evolved from regarding an innovation as an invariant quality to the concept of re-invention, defined as "the degree to which an innovation is changed or modified by a user in the process of its adoption and implementation (p. 17)". Ram and Jung (1994) discuss the concept of "use innovativeness" in the context of adopting computers. A great deal of computer technology can be described as "model-less"; there are some inherent use characteristics, but there is also a wide margin for "invented" uses (Ram & Jung, 1994). Quantitative research methods are useful for gathering information about current diffusion patterns and rate of adoption of certain technologies on campus, while qualitative methods, such as open-ended questions and interviews are useful for exploring the concepts of interrelatedness and re-invention.
When an innovation has been adopted by most or all of the members in a social system, diffusion has reached the saturation point. Geoghegan (1994) suggests that this saturation point has been reached with early adopters of instructional technology. When new ideas are invented, diffused, and are adopted or rejected, leading to various consequences, social change occurs (Rogers, 1995). This social change can be planned or spontaneous, intended or unintended; for example, a physics department invents a new network interface and protocol for exchanging leading edge information among physicists versus the spontaneous and exponential demand for access to the Internet with the advent of the World Wide Web.
Rogersí (1995) asserts that an individualís decision
to adopt an innovation is not an instantaneous act. Rather, it is a process
that occurs over time, consisting of a series of actions and decisions.
Rogersí model of the innovation-decision process, conceptualized as consisting
of five stages, is depicted in Figure 1. The innovation-decision process
is "the process through which an individual (or other 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" (Rogers, 1995, p. 163).
Figure 1. A Model of Stages in the Innovation-Decision Process (Rogers, 1995)
The innovation-decision process is essentially an information-seeking and information processing activity in which the individual is motivated to reduce uncertainly about the relative advantages and disadvantages of an innovation (Rogers, 1995). Knowledge occurs when an individual (or other decision-making unit, hereafter referred to as "an individual") is exposed to an innovationís existence and gains some understanding of how it functions. Types of knowledge range from awareness about the innovation, how-to use an innovation properly, and principles-knowledge dealing with the functioning principles underlying how the innovation works. Predispositions such as selective exposure and selective perception may influence an individualís behavior toward communication messages about an innovation and the effects that such messages are likely to have. Hassinger (1959), cited in Rogers (1995), argues that even if individuals are exposed to innovation messages, such exposure will have little effect unless the innovation is perceived as relevant to the individualís needs and as consistent with the individualís attitudes and beliefs.
Persuasion occurs when an individual forms a favorable or unfavorable attitude toward or opinion of the innovation based upon perceived characteristics of the innovation, such as relative advantage, complexity, and so on. Persuasion is also influenced by information sought from near-peers whose subjective opinion of the innovation is most convincing (Rogers, 1995). When someone who is like us shares a positive evaluation of the innovation, we are more motivated to adopt it. Decision occurs when an individual engages in activities that lead to a choice to adopt or reject the innovation. Adoption is a decision to make use of an innovation as the best course of action available. Active rejection means considering and trying the innovation out on a limited basis, and deciding not to adopt. Passive rejection, also called non-adoption, consists of never really considering the use of the innovation.
Implementation occurs when an individual puts the innovation into use. Until this stage, the process has been a mental exercise. Implementation involves an overt behavior change as the new idea is actually put into practice. This stage may continue for a lengthy period of time until the innovation finally loses its distinctive and noticeable quality as a new idea. Re-invention, the degree to which an innovation is changed or modified by the user, can also occur in this stage. Confirmation occurs when an individual seeks reinforcement of an innovation-decision already made, or reverses a previous decision to adopt or reject the innovation if exposed to conflicting messages about the innovation. Each stage in the innovation-decision process is a potential rejection point. One can gain awareness of an innovation in the knowledge stage, and then simply forget about it. Rejection can occur even after a prior decision to adopt, which is called discontinuance.
The time element of the diffusion process allows us to generate diffusion curves and to classify adopters into categories. Because individuals in a social system do not adopt an innovation at the same time, innovativeness is the degree to which an individual is relatively earlier or later in adopting new ideas than other members of a social system (Rogers, 1995). For example, word processing is becoming a ubiquitous technology on campuses; faculty who used text editors twenty years ago have a higher degree of innovativeness than faculty who started using word processing yesterday. According to Rogersí (1995) theory, the diffusion of an innovation usually follows a normal, bell-shaped curve when adoption is plotted over time on a frequency basis (Figure 2). If the cumulative number of adopters is plotted, the result is an S-shaped curve. A diffusion curve allows us to compare the innovativeness of an individual or other unit of adoption with other members of a system, usually measured as the number of members in the system to adopt the innovation in a given time period. Many human traits are normally distributed; physical traits such as height or weight; behavioral traits such as intelligence or learning of information. Hence, Rogers (1995) reasons, a variable such as the degree of innovativeness is also expected to be normally distributed; decades of diffusion research on innovativeness from across disciplines has supported this expectation.
Mahajan, Muller, & Srivastava (1990) suggest that Rogersí classification model based on innovativeness offers several advantages for describing the adoption patterns of individuals in a group: (1) it is easy to use, (2) it offers mutually exclusive and exhaustive standardized categories, by which results can be compared, replicated, and generalized across studies, and (3) because the underlying distribution is assumed to be normal, continued acceptance of an innovation can be predicted and linked to the adopter categories.
Rogers (1995) suggests that the adoption of a new idea results from information exchange through interpersonal networks. The first adopter of an innovation discusses it with other members of the system, and each of these adopters pass the new idea along to other peers. The diffusion curve begins to level off after half of the individuals in a social system have adopted, because each new adopter finds it increasingly difficult to tell the new idea to a peer who has not yet adopted, for such non-knowers become increasingly scarce. The segment of the diffusion curve between 10 to 20 percent adoption is "critical mass" or the "heart of the diffusion process" (Rogers, 1995) and represents the transition from the "early adopter" level of innovativeness to the "early majority".
Figure 2. Adopter Categorization on the Basis of Innovativeness (Rogers, 1995)
Having been abstracted from empirical investigations and market research, the five adopter categories Rogers (1995) describes along the continuum of innovativeness (i.e., innovators, early adopters, early majority, late majority, and laggards) are "ideal types" designed to make comparisons possible based on characteristics of the normal distribution and partitioned by the mean and standard deviation. Although the values of the mean and standard deviation differ for each sample of observations, Crocker and Algina (1986) describe three characteristics that all normal distributions share: 1) the mean, mode, and median are always the same value, 2) every normal distribution is symmetric; that is, the right and left halves of the curve are exact mirror images, and 3) approximately 68% of the scores lie in the interval between µ and µ +/- 1s x. Ideal types are not simply the average of all observations about an adopter category, that is, exceptions can be found, and pronounced breaks do not occur between each of the five categories.
A different diffusion curve can be generated for each type of computer application to compare percentage of diffusion. For example, personal use of word processing is described as almost completely diffused because it has been adopted by a majority (> 90%) of faculty (Geoghegan, 1994), whereas presentation software and email use in classrooms is just beyond "critical mass" at 20 percent adoption (Green, 1996). Levels of college student (33%) and home (40%) computer ownership have passed the early adopter stage, with faculty ownership (50%) diffusing into the late majority (Green, 1996). Research comparing both the adoption of various technologies and the extent to which they are used effectively by university faculty for teaching and learning would generate different diffusion curves. The following summary descriptions provide a useful starting point to differentiate between adopters using Rogersí (1995) categories.
Rogers (1995) describes the EA as the "heart of the diffusion process" because they decrease uncertainty about a new idea by adopting it, and then convey a subjective (i.e., hunch or gut feeling) and or an objective evaluation of the innovation (i.e., empirical investigation of effectiveness) to peers through interpersonal networks. EAs have been found to differ from later adopters across a number of personality variables. EAs have more empathy, less dogmatism, a greater ability to deal with abstractions, greater rationality, greater intelligence, a more favorable attitude toward change, a better ability to cope with uncertainty and risk, a more favorable attitude toward science, less fatalism, and higher aspirations for formal education and occupations than do later adopters.
The LM is a skeptical one-third of a social system, and adopts new ideas after the median (i.e., 50th percentile) member of a system. Adoption may be both an economic necessity and as a result of increasing network pressure from peers. Innovations are approached cautiously, the LM do not adopt until most others have done so, and system norms must definitely favor an innovation before they are convinced. Their relatively scarce resources mean that most of the uncertainty about a new idea must be removed before the LM feels that it is safe to adopt.
Information has been collected by researchers in an attempt to characterize individuals at the tail end of the distribution as a specific personality type. Rosen and Maguire (1990) conducted a meta-analysis to examine the personality characteristics of computerphobics, and found that none of the common beliefs characterizing the computerphobic (i.e., they are female rather than male, older rather than younger, and possess other types of anxiety) represent reality. However, with regard to the current discussion of EAs and innovation adoption patterns, it seems likely that computerphobics represent the tail-end of the distribution and are small in number. Rosen and Maguire (1990) state that the computerphobic group is actually quite small (<10%). We might extrapolate this finding to the faculty population to predict that the number of faculty who resist technology because they are computerphobic is probably small.
Rogersí Diffusion Theory: Related Contexts
Researchers have provided evidence for Rogersí theory by examining the diffusion of various innovations. Dickerson and Gentry (1983) found that EAs of home computers displayed similar characteristics to adopters of other innovations: middle-aged, higher income, more education, an opinion leader and information seeker. They found that EAs of home computers have had more experience with a variety of technical products and services than non-adopters. Consistent with Rogersí proposition that the more compatible the innovation is with the adopterís background, the more likely it is to be adopted, the two experiences which best predicted adoption of the home computer are those related to functions (i.e., games, programming) superseded by the home computer. In a study of school counselors, Casey (1995) describes innovators as advanced, self-taught "power users" who are authoring programs using programming languages, and the laggards as "technophobes" who avoid computer technology at all costs. He believes most counselors fall somewhere between these two extremes. Caseyís (1995) EAs were more mainstream than the innovators, effective at amplifying promising developments engineered by innovators, and eager leaders who provided workshops and publications for peers while struggling with the slow pace of mainstream acceptance.
Ram and Jung (1994) looked beyond diffusion patterns to investigate adopter characteristics with regards to use innovativeness with personal computers. Use innovativeness is the degree to which an adopter uses a previously adopted product to solve a novel consumption problem. EAs are found to have higher usage variety than do later adopters, which may be a result of their higher involvement with the innovation. In other words, EAs are likely to be more use innovative and capitalize on the wide variety of uses to which a computer can be put, be more aware of its various features and capabilities, and seek different uses for their computers than do later adopters. EAs, like expert computer users, use more options, features and software on their computers, whereas the early and late majority, like novice users, use fewer options to start with. Ram and Jung (1994) suggest that later adopters are more intimidated with new technology and need different kinds of support than EAs, such as additional training and user-friendly manuals. Another appropriate strategy may be product differentiation through simplification: create a no-frills computer for the later adopters, rather than trying to make them as diversely accomplished as the EAs are with the fully loaded model (Ram & Jung, 1994).
There is a growing number of computer-using faculty who are not necessarily highly skilled, or computer literate in the traditional sense, but are very enthusiastic about adopting technology because they see the potential of newer tools, such as e-mail and the World Wide Web, for their students. Many observers agree that communication technologies may be what entices mainstream faculty to adopt other technologies for teaching and learning (Foa, 1993). Once they are intrigued by e-mail and the Web they may start asking questions about other technologies (Gilbert, 1996). These enthusiastic beginners see technology as a methodology for doing neat and exciting things with their students rather than being fascinated with the technology itself.
With the development of graphical interfaces, technology has become somewhat more transparent and user friendly. However, there are still barriers that may constrain use by enthusiastic beginners, and a fairly steep learning curve to climb before integration becomes effortless. User friendliness is a seductive term that does not accurately represent current technology reality. Computers are still not well-designed, fault-free, and easy to use. In fact, software manufacturers seem to be swinging from a "user friendly" simple design with few features but great functionality, to a more complicated, feature-rich design. Donald Norman (1993), a cognitive scientist who researches human-computer design, must be having a field day examining the thousands of new, and often poorly mapped, features and capabilities of current software! For example, Microsoft Word 6.0, a current market share leader, is a powerhouse 16 MB word processor with thousands of features that will probably never be used by the average user. Because of their use innovativeness, EAs might maximize their investment in such a program by utilizing many of its capabilities. However, later adopters may not need a feature-rich program to start with, and may be intimidated by all of the bells and whistles.
Results from a faculty survey conducted 10 years ago (Jacobson & Weller, 1988) indicate that early adopters, with self-reported good-excellent computer skills, had different perceptions about obstacles than did later adopting, mainstream faculty with poor-fair computer skills. While a majority of faculty agreed that lack of funds for hardware and the lack of technical support were obstacles, a larger percentage of mainstream faculty viewed the lack of technical support as more problematic than early adopters. EAs were more self-sufficient with regards to support and wanted more access to hardware resources for experimentation. Although the EAs reported acquiring computer use mainly through self-training and assistance from colleagues, both EAs and mainstream faculty felt that a lack of training was an obstacle to widespread use of computers. Jacobsen and Weller (1988) found that although the reported use of some computer applications was quite low, enthusiasm for adopting additional innovations was quite high across both groups.
These findings suggests three trends: (1) that the use of computers for one purpose may encourage enthusiasm for further computer use, (2) that mainstream faculty may be limited adopters because of the lack of technical support and training, and (3) that colleague supported training is a viable way to encourage diffusion of computer applications and use. There appears to be an opportunity to capitalize on the early adopterís knowledge and skill base, and somehow share this with mainstream faculty who have concerns about support and training.
Hamilton and Thompson (1992) provide a good summary of certain personality characteristics displayed by EAs in their study of the adoption of an electronic network for educators. A communications network was established to create an electronic link between student and practicing teachers and the education faculty at a college to: decrease the isolation often experienced by student and practicing teachers, to make faculty expertise readily available, and to increase faculty awareness of any problems in the field. EAs in this study shared similar levels of education, social status, and social participation, had a cosmopolitan outlook, accessed information from mass media, belonged to wide interpersonal communication networks, displayed a high degree of innovation information seeking, possessed positive attitudes toward change and risk, and had similar aspirations and neutral attitudes toward fatalism. EAs played an important role in this diffusion process because their adoption was visible to the early majority and influenced their subsequent adoption. Hamilton and Thompson (1992) suggested that network developers should seek out EAs who will enhance the diffusion process.
The following study examined early adopters with respect to teaching methods. Philipp, Flores, and Sowder (1994) studied Kindergarten-to-Grade 12 (K-12) mathematics teachers who were identified as early adopters of innovative teaching methods, and found their characteristics to be similar to those summarized by Rogers (1995). EAs focused on problem solving, conceptual relationships and understanding, and communication in mathematics. These characteristics are similar to the "discrete and successful use of new ideas" described by Rogers (1995). Teachers had a comprehensive knowledge of the mathematics they were teaching, which is consistent with the alleged "higher rationality, higher intelligence" of EAs. These teachers participated in their own professional growth by attending conferences and inservice programs, completing graduate studies, and seeking encouragement and support for their reform from peers and administration, suggesting that these EAs were involved in and contributed to a rich interpersonal network.
Often, the individuals who have integrated technology for teaching and learning have done so in a university climate that has provided little or no external or explicit recognition or incentive for either excellent teaching or technology implementation (Sammons, 1993). There is no professional training requirement for university teachers as far as their teaching is concerned (Laurillard, 1993), faculty members receive little or no formal training on using computers for teaching and learning, and the annual review process often fails to recognize innovative teaching as part of the merit system (Sammons, 1993). Instead, faculty rely on colleague support and self-teaching.
A faculty member may combine teaching and research with technology. However, development time for computer-based teaching materials may extend over years, with little reward for the final product. In fact, many universities have a policy which requires the developer to share or give copyright of software products to the institution (Reeves, 1991). It appears that system-wide changes will be needed in the reward system and training for faculty members in order to encourage broader diffusion of instructional technology within the mainstream.
Excellent Teaching and Early Adoption
Characteristics of exemplary teaching apply directly to the effective use of technology in undergraduate teaching. The integration of technology implies more than adopting computers for teaching and learning tasks. Integration implies a transformative or re-invention process where instructional strategies and outcomes are redefined by technology; the innovative capabilities and possibilities of technology are used to fundamentally change teaching and learning. Therefore, any attempt to understand the integration of technology using Rogers (1995) theoretical framework has to also include a consideration of the individual case, the teacher who has adopted technology and is figuring out ways to transform their classroom environments. Not all professors who adopt technology integrate this innovation in meaningful ways to transform teaching and learning.
One of the potential limitations of considering the integration of technology using only Rogersí (1995) adopter categories lies in their very nature as summaries of global characteristics and time of adoption. While the innovation-decision process and adopter categories are useful for simplifying the complexity of adoption patterns in a social system by describing the central exemplar or summarization of the early adopter and other categories, these "defining characteristics" also understate the uniqueness of the individual member. It is worth remembering that early adopters are, at the same time, unique and variable individuals who may resemble each other much less than they resemble the general subgroup characteristics. For example, one can imagine that EAs possess various and different: levels of ability and skill, beliefs and visions about the value of technology, specific personality traits, levels of risk-taking behavior, motivations to learn about technology (internal, external, environmental, opportunity), development patterns (self-taught, peer teaching, courses), and have implemented computers in different environments, under different conditions (i.e., vendor, department and self support) and with different expectations. Indeed, an interesting question worth further investigation is whether early adoption depends on personality or environment. Although not a goal of the present investigation, it appears that there is a need to develop a model, similar in nature to Sternberg and Horvathís (1995) "Prototypical Model of the Expert Teacher" which allows for variability among experts, against which one can compare EAs of instructional technology to better understand their commonalities and differences.
Although early adopter categories are useful to describe general group characteristics and trends, there is a need for more focused and careful description of individuals within this category. Donald Norman (1993) eloquently describes this problem in his distinction between logical facts and the power of personal stories in decision making processes. The problem is that in an attempt to abstract the relevant from the irrelevant, logical analysis oversimplifies to the extreme and only applies to information that can be readily measured; however, what can be measured and what is important are not necessarily related (Norman, 1993). Subjective concepts, like value or beauty, moral good or evil, cannot be measured like number of classroom events or details identified, or types of planning and teaching strategies. These are all subjective concepts, and even though all may agree that morality and values are important, there is no simple way to translate these into a language of logic; no way without badly distorting their content (Norman, 1993).
Personal stories, or profiles, of experts are well suited to capturing exactly those elements or details that formal models, such as theories of diffusion, may leave out. A theoretical framework is an attempt to generalize and summarize the characteristics of adopters, to remove from the analysis subjective emotions and thought. Personal stories, such as the profiles of expert teachers (Sternberg, 1997), capture the subjective emotions, thoughts, and beliefs of the category members. Logic generalizes, stories particularize (Norman, 1993). Logic allows one to form a detached, global judgment; storytelling allows one to take the personal point of view, to understand the particular aspects of expertise and experience embodied by the individual. Stories are not better than logic; logic isnít better than stories (Norman, 1993). It is appropriate to use both in the attempt to characterize early adopters of technology on campus.
In their review of knowledge elicitation techniques for expert system design, Ford and Adams-Webber (1992) present a convincing argument for gathering data from multiple experts both case by case and within a model: "From a constructivist perspective we would expect experts to agree about the vast majority of their knowledge (i.e., widely shared consensual beliefs) and yet have major differences in their largely self-constructed expertise" (p. 131). They suggest that it is often preferable to build separate knowledge bases for each expert rather than attempting to incorporate their expertise into a single data base to avoid a homogenized, "averaged" opinion about an area of expertise. This view is consistent with the individual case studies presented in Sternbergís (1997) book about expert psychology teachers; although there are similarities, each expert is unique.
Rogersí (1995) theory of the diffusion of innovations and adopter categories are appropriate for capturing the global characteristics of early adopters, but a case by case description captures the self-constructed nature of the individualís expertise gained from years of personal experience "consisting of functional but fallible anticipations held with high confidence and uncertain validity" (Ford & Adams-Webber, 1992). Normanís (1993) summary on the value of both logic and storytelling for decision making captures best the process by which the present investigation of adoption patterns and characteristics of faculty who integrate technology in teaching and learning will follow in an attempt to characterize EAs: "First the data and the logical analysis, then the stories in order to let the personal, emotional side (of early adopters) have the last word" (p. 130).
Let us consider the contribution of case study research to understanding the characteristics of expert teachers in higher education. Scardamalia and Bereiter (1993) write rich and detailed accounts of experts from various domains, including elementary teaching, and a number of books have been written that profile the individual experiences of expert higher education teachers (Cahn, 1978; Ellis, 1993; Jones, 1995; Sternberg, 1997), with many of the essays written by the expert teachers themselves. Captured in these accounts are the views and stories of individual faculty who have been identified (by various means) as expert teachers. Individual profiles fill in some of the gaps left by definitional models by describing the expertís approach to excellent teaching. For example, Sternbergís (1997) book profiles the views of expert psychology professors who have written textbooks for their course. Many of the expert professors concur that their mission is to "give psychology away", and describe a multitude of complex pedagogic means and strategies to encourage, teach, promote, and require critical thinking. Although each story conveys a rich and unique account of the expertís approach to teaching, there is also remarkable consistency across profiles in what the individual experts regard as important and valued instructional goals. Berliner (1992) has also found this consistency across experts, which suggests commonality, and both he and Sternberg (1997) suggest that expert teachers are the best mentors for beginning teachers.
Perhaps the greatest value of case studies which profile expert post-secondary teachers is to provide role models for junior faculty. The number of experts is small, and the case study may be the most efficient and practically feasible way to share and distribute their expertise with other faculty. Another potential contribution of the personal case study is its value as a textual protocol of the expertís reflection on what it means to be expert. Taken together, these case studies offer a variety of perspectives that give insight into the uniqueness and variability of the individual earlier adopters, as well as the commonalities between them.
Once a faculty member has adopted technology for teaching, what motivates them to move beyond competent use to further develop their technological and pedagogical expertise? Rogersí (1995) theory is very useful for understanding and prediction the diffusion of an innovation in a social system over time, and his adopter categories provide a logical way to summarize and characterize early adopters and others. Clearly, EAs share characteristics that differentiate them from the majority of mainstream faculty. A consideration of EAs who readily integrate technology for teaching and learning, leads quite naturally to questions about the difference between integrating technology for teaching and learning, and merely using technology as an add-on to instructional strategies.
Rogers (1995) calls for increased understanding of the motivation to adopt an innovation (p. 109), and diffusion research does not adequately address the motivational aspects of becoming expert at integrating an innovation. Bereiter & Scardamalia (1993) provide more insight into the motivation to develop expertise by highlighting three ideas from cognitive psychology: (1) flow, which suggests that individuals put effort into the process of expertise because it feels good, (2) second-order environments, which, unlike other social environments, provide support for the process of expertise, and (3) the heroic element of expertise, which acknowledges that the other explanations do not quite complete the motivational picture.
Flow requires a nice balance between ability and challenge; if challenge exceeds ability, the result is frustration and anxiety, and if ability exceeds challenge the result is boredom. Combined with the effects of learning, repetition of the same activity will eventually cease to produce the pleasurable flow experience, and something must be done to increase the level of challenge to bring it in harmony with the increased level of ability. Thus, there is a progressive element to maintaining flow.
Experts seldom exist in isolation, and instead are linked together by associations or informal networks. The expert subculture embodies ideals and goals which help direct the expertís development, and provide support, cooperation and recognition of success. In a second-order environment, or expert subculture, one of the requirements of adaptation is to participate in the pursuit of ideal goals of the group, and this necessitates progressive problem solving (Bereiter & Scardamalia, 1993). Conditions to which an individual must adapt change progressively as a result of successes of other people in the environment. Each expertís advance in technology, strategy, or contribution to knowledge, sets a new standard which others try to surpass. Individual experts do not merely adapt to constant change. Instead, one adapts to changes that keep raising the ante, by setting a higher standard of performance, by reformulating problems at more complex levels, or by increasing the knowledge that is presupposed (Bereiter & Scardamalia, 1993).
The heroic element of the process of expertise captures the development of individual experts who exist in first-order environments that do not necessarily support or reward the development of expertise. Heroic experts are found delivering mail, at home caring for children, or even teaching in a solitary university or college classroom. These experts reinvest mental resources in their work, elevating it or expanding its scope to take in a broader set of concerns-such as the concerns of their students. These experts exhibit professionalism in its most favorable sense, but often without the benefit of professional identification or a subculture to support them (Bereiter & Scardamalia, 1995). The intrinsic nature of flow may motivate them, but the nature of their work suggests that the benefits of flow may be few and far between. In such cases, pursuing high standards and continuing to advance requires an element of heroism in the sense that arduous efforts that benefit others are disproportionate to what others provide in the way of rewards and supports. The image of the heroic mail carrier braving storm and flood to deliver the mail reinforce the fact that the hero must go it alone; there are few social forces lending support. Athletes and performing artists also convey the heroic image of the expert by virtue of the arduous drill and training that they sustain to bring their performance up to that moment which we ignorantly applaud as a display of natural talent (Bereiter & Scardamalia, 1993). Even with experts who seem to achieve their status without arduous effort, there is an element of heroism. For if expertise involves progressive problem solving, and progressive problem solving entails working at the edge of oneís competence, then at least a bit of daring is required; working at the edge risks failure and loss of esteem, but it also provides a certain excitement, which is probably addictive (Bereiter & Scardamalia, 1993).
It may not be the case that "early adoption of instructional technology" and "excellent teaching" are qualities that often exist in the same faculty member. Rob Chandhok, from Carnegie Mellon University, reminds us that "there are plenty of innovators in education that make no use of technology at all" (Gilbert, 1995, p. 33). Universities have to design technology integration plans that focus both on excellent teaching and integrating various technologies to support teaching and learning. Early adopters of technology who are also excellent teachers have much to contribute to this planning process. Kearsley (1996) suggests that excellent teaching should be our first priority, because adopting technology will not improve poor teaching, except temporarily. He argues that in the absence of knowledge about and enthusiasm for the discipline, student participation, explicit expectations, well-defined course structure, and an enjoyable learning environment, technology will not enhance learning to any appreciable degree. If cases are found where early adoption and excellent teaching exist in the same individual, then it is worth profiling this rare expertise for the benefit of other faculty members who wish to develop both their technology and teaching knowledge and skills. The present investigation begins this task by interviewing and profiling the characteristics of a small number of earlier adopters who are also excellent teachers.
Rogers (1995) identifies the pro-innovation bias as a potential shortcoming of diffusion research. The pro-innovation bias is the implication in diffusion research that an innovation should be diffused and adopted by all members of a social system, that it should be diffused more rapidly, and that the innovation should be neither re-invented nor rejected. The topic of interest in the present investigation is the integration of technology for teaching and learning. As such, a survey has been designed that measures time of adoption of technology for teaching and learning in order to plot the diffusion curves for this innovation. Aspects of the research methodology have also been designed to address a potential pro-innovation bias. For example, both early, later, and non-adopters are participants in this investigation in order to explore the different characteristics of early adopters and mainstream faculty. Also, different subscales have been chosen to gather information about changes to classrooms, incentives, and barriers to integrating technology. This information will help to increase understanding of both the motivators and impediments to adoption, as well as to begin to understand reasons for rejection and discontinuance. The recognition that the integration of technology implies more than just whether or not a faculty member uses computers, leads to a consideration of how and why this innovation is adopted. Motivations for adoption are a difficult issue to investigate (Rogers, 1995). Seldom are direct questions in a survey adequate for uncovering an adopterís reasons for using an innovation. However, diffusion research that attempts to see an innovation through the eyes of the adopters and non-adopters may result in a better understanding of why the innovation was adopted or rejected, and yield descriptions of what is good and bad about a technology. Hence, in addition to collecting quantitative data about time of adoption and experiences with the innovation, the present investigation includes both restricted-response and open-ended questions that gather qualitative data about the incentives and barriers to integrating technology for teaching and learning, and methods for using and evaluating technological applications. Additionally, interview methods have been chosen in order to gain insight into the innovation-decision processes and motivation of EAs who have also been identified as excellent teachers.
Universities are in a situation where there is widespread adoption of instructional technology by innovators and early adopters, but limited adoption by mainstream faculty. It is apparent from descriptions of EAs and the "early-late majority" mainstream, that these two groups have different characteristics, motivations, and needs. Therefore, campus-wide integration plans cannot be developed on the assumption that mainstream faculty will naturally use computers as readily and easily as the early adopter. In the relatively short period of time that instructional technology has been used on campuses, many hard lessons have been learned and it is up to each and every "learner" to share those lessons (Lessons Learned Home, 1998; Reeves, 1991). This knowledge sharing process can be made more efficient and widespread through institution level commitment and support of IT.
Critical Mass and the Chasm Between Early Adopters and Mainstream
According to Greenís (1996) annual Campus Computing Survey, adoption of technology for classroom use rose between 1994 and 1995. E-mail use doubled to 20 percent, use of presentation software was over 25 percent, and the use of multimedia resources and CD-ROM-based materials has risen to just under 10 percent. Green (1996) suggests that the use of information technology is approaching the "critical mass" level, described by Rogers (1995) as the point at which enough individuals have adopted an innovation so that the innovationís further rate of adoption becomes self-sustaining. However, Green (1996) also indicates that of all the issues surrounding the adoption of technology for teaching and learning, individual faculty rated "user support and training" as the most important. Unfortunately, the investment in instructional development (that is, providing assistance to faculty eager to use technology in their classrooms) has remained flat on some campuses over the last six years. Although infrastructure supports innovation, and many campuses have taken steps to replace obsolete equipment and provide access to multimedia capable computers, technical assistance and user support are still the more critical catalysts for adoption and integration of instructional technology (Green, 1996).
Geoghegan (1994) describes what he refers to as a "chasm" between early adopters and the early majority, such that the innovation is never adopted by the mainstream. He contrasts early adopters, who are risk takers, more willing to experiment, generally self-sufficient, and interested in the technology itself, with early majority faculty who are more concerned about the teaching content and learning problems being addressed than the technology used to address it, view ease of use as critical, and want proven applications with low risk of failure. Geoghegan (1994) suggests that critical mass is insufficient by itself to support continued diffusion because of the lack of institutional support for: (1) developing instructional software, (2) plans for further integration of computers into the curriculum, (3) shortages of equipment and facilities, and (4) unrealistic expectations by administration based on innovatorsí and early adopterís successes.
Early adopters make an innovation visible to the mainstream and decrease uncertainty about the innovation. EAs are more experienced with technology and have higher use innovativeness, thus capitalizing on technologyís many features and options. They seek different uses of technology to solve novel problems and contribute to new and better uses of technology. However, by making adoption look relatively easy, they disguise the extensive knowledge and skills that mainstream faculty will need in order to adopt. Geoghegan (1994) believes that without wide-spread institutional support, the successes of EAs will not effectively and efficiently diffuse into the mainstream.
A survey conducted by Spotts and Bowman (1993) at Western Michigan University supports Geogheganís (1994) view that mainstream faculty have different needs. Factors identified by more than half of the faculty as important in influencing their use of instructional technology were: availability of equipment, promise of improved student learning, funds to purchase materials, compatibility with subject matter, advantages over traditional (existing) methods, increased student interest, ease of use, information on materials in their discipline, compatibility with existing course materials, university training in technology use, time to learn the technology and comfort level with the technology.
An additional factor identified by Ehrmann (1995), "The medium is not the message", may also contribute to the mainstreamís hesitance to adopt. Communications media and other technologies are so flexible that they do not dictate methods of teaching and learning. Ram and Jung (1994) also referred to the "model-less" nature of many computer applications. The mainstream needs direction on where to start with flexible technologies that can be integrated in any number of ways. However, administrators often assume that once faculty get access to technology they all will easily, automatically, and quickly change their teaching methods and course materials to take advantage of IT. The chief culprit for this belief is the varied and extensive use by EAs and basing expectations for mainstream faculty adoption on this use innovativeness.
The literature suggests one clear message: administration has to be convinced to let go of the infrastructure-driven, "if you build it, they will come" approach to technology integration on campus if they want to address the chasm between early adopters and mainstream faculty. Faculty and administration have a deep mutual dependency. The top-down program advocate needs convincing exemplars to justify large investments in technology at a moment when funds are scarce, and the bottom-up project advocate and enthusiastic beginner needs a well-conceived and reliable working environment for successful implementation of innovative concepts (Noblitt, 1997). Change agents in the administration (i.e., the president, deans, and directors of service units), opinion leaders (i.e., early adopters), and mainstream faculty, need to find ways to discuss implementation strategies and develop technology integration plans for campus-wide adoption.
Universities traditionally have flatter organizational structures with loosely coupled organizational units to provide the primary services of higher education (Bull, et al., 1994) compared to the private sector. Initiatives for the innovative use of instructional technology (IT) in teaching and learning tend to come from early adopter individuals and research units. With the reduction in size and price of computing resources and the required investment, decision-making for IT investment more easily fits the traditional organizational structures of higher education with decentralization and local responsibility for decisions.
However, these individual initiatives and efforts, as well as decentralized investments in IT, scattered all over an institution, or scattered all over the institutions within one province or country, are insufficient by themselves to fully develop the potential of instructional technology for teaching and learning (Bull, et al., 1994). Critical mass is just not enough. Early adopters might be committed and enthusiastic in developing new technology-based teaching methods and computer assisted instructional software. However, to make these efforts more widespread and their results used more comprehensively, incentives, training, support and reward structures "from above" are needed to build a strong human infrastructure (Daigle & Jarmon, 1993), as well as providing the technological infrastructure (i.e., networks, hardware and software) to drive integration. IT investments for teaching have to be similar to the state of the art in the world of work, as higher education prepares graduates for the future. These ever-new investments cannot be left to uncoordinated departmental or individual initiatives, as they often exceed respective budgets (Bull, et al., 1994).
Administration needs to recognize that to cause change they will have to address the reward system and commit to system-wide investment in IT in order to address the needs of mainstream faculty; the key to diffusion will be training and support. Without investment in the human infrastructure, nothing of sustainable value will be achieved (Foa, 1993).
Brace and Roberts (1996) describe a campus-wide approach to technology integration based on Rogersí (1995) innovation-decision process that targets mainstream facultyís needs. Innovations are likely to gain more rapid acceptance if they are perceived as having high relative advantage, or as being better than the idea they supersede (Rogers, 1995). Innovations with a high compatibility with existing values, past experiences and needs of potential adopters also have an advantage.
The campus-wide strategies described by Brace and Roberts (1996) are: (1) to build awareness of the possibilities and advantages of technology, EAs from various disciplines demonstrated how they developed multimedia applications and used them in their courses, and the university sponsored yearly technology conferences and symposia, (2) ready access was provided to up-to-date, stable and reliable technology, as well as providing each faculty member with a personal desktop computer, (3) training was made available through developmental workshops, orientations, and one-on-one sessions, (4) technical support for both hardware and software was provided by service units for acquisition, installation, information and implementation, and (5) funding was provided for release time and summer grants, and recognition was provided through incentives and encouragement. Although no data were provided to evaluate the outcomes and the success of this integration plan, the implications seem clear: instead of relying on "critical mass" and serendipitous diffusion to bridge the "chasm" between EAs and mainstream faculty, those who propose wide-scale adoption of a technology-based curriculum must find a way to combine innovation with a responsible, campus-wide plan for implementation (Noblitt, 1997). Rogersí (1995) innovation-decision process and stages of adoption will be used to frame the following discussion of the contribution that EAs can make to a campus-wide technology integration plan.
Visionaries who believe in the value of information and instructional technology are needed on campus. They need to be leaders who can effect real change by somehow increasing how-to and principles-knowledge among the mainstream. How-to knowledge consists of information and skills necessary to use an innovation properly (Rogers, 1995). In the case of complex innovations, such as computer technology, the amount of how-to knowledge needed for proper adoption is much greater than for less complex innovations. An inadequate level of how-to knowledge may lead to rejection and discontinuance because of the frustration likely to be encountered. Principles-knowledge consists of information dealing with the functioning principles underlying how the innovation works (i.e., research on the effects and outcomes of using certain technologies in teaching and learning). It is possible to adopt an innovation without principles-knowledge, but the danger of misusing the new ideas is greater, and discontinuance may result.
A campus-wide culture that promotes adoption of technology can be developed by leaders at each level of the organizational structure. Those at the executive levels are the hardest to convince to take the lead in using technology, perhaps because many belong to the pre-computer generation (Foa, 1993). Characteristics that are beneficial to long-term planning are capturing the vision and enthusiasm for innovation displayed by EAs, and channeling this into system wide initiatives that benefit all faculty. The biggest challenge is cultural: in computing organizations and cliques, the "techies" are at the top of the pecking order and like to tinker with technology, while the "teachies" regard technology as a possible solution to a teaching and learning problem (Gilbert, 1995). What is needed is some way to get the "top-down" folks, the "techies", and the "teachies" to talk to one another. Starting with the president, and including vice-presidents, deans, and directors of each division, a technology-rich culture can start from changes to communication channels. For example, to promote e-mail use (and take advantage of the campus network) ensure that every faculty member, including the president, has a computer, network access, and thorough training in how to use the email system. Then, instead of using the paper-based, internal "snail-mail" system to distribute news and information, ensure that the president, deans, department heads, and directors put news or information on the system and nowhere else (Foa, 1993). This commitment will require management and administration to abandon the "real (wo)men donít type" approach to communication. When new ideas are adopted, leading to various consequences, social change occurs (Rogers, 1995). E-mail and the Internet are already attractive to mainstream faculty, and are fully diffused among EAs. If campus leaders demonstrate their commitment to information technology by adopting changed communication channels, they will start a ripple effect throughout the institution, and indeed, maybe within themselves. And, the use of computers for one purpose encourages future computer use and questions about other technologies (Broholm, 1993).
A role for EAs in the knowledge and persuasion stages of adoption is to share what they have learned about instructional technology with the mainstream through in-house and across discipline demonstrations, campus conferences and symposia. Rogers (1995) posits that mass media channels, as knowledge creators, are often most important for informing people about an innovation, while interpersonal channels are more important in persuading someone to adopt a new idea. EAs play an important role in further diffusion because of their role as opinion leaders in communication channels and social systems. The transfer of ideas in a social system is most effective when participants belong to the same groups or are drawn together by the same interests (Rogers, 1995). Shared meanings and mutual language mean communication is likely to result in greater knowledge gain, attitude formation and change, and overt behavior change. Generally, faculty who are homophilous (degree to which a pair of individuals who communicate are similar) develop stronger communication relationships with each other than those who are heterophilous (not alike on the categorical variable of interest) (Valente, 1996). The similarity may be in certain attributes, such as being in the same faculty or department, type of computer used, and the like. When two individuals share common meanings, beliefs, and mutual understandings, communication between them is more likely to be effective (Valente, 1996).
Change agents and later adopters may have difficulty developing trust and finding common ground if their beliefs about adoption are dissimilar. EAs share characteristics and attributes that make communication between EAs of instructional technology effective (i.e., informal networks composed of Mac users, web-course developers, interface designers, and so on). Interpersonal diffusion networks are mostly homophilous. However, in order for instructional technology to diffuse into the mainstream, interdisciplinary EAs and mainstreamers have to exchange knowledge. Heterophilous network links often connect two cliques, thus spanning two sets of socially dissimilar individuals in a system (Broholm, 1993). These heterophilous links are especially important for exchanging information about innovations, as is implied in Valenteís (1996) description of the strength of "weak ties"; there is a higher information exchange potential in communication channels when the communicators are heterophilous (Valente, 1996). Homophilous diffusion patterns cause new ideas to spread horizontally, rather than vertically, within a system. For example, a computer science professor uses web-based publishing as a communication network in a senior class, or a computer engineer discovers a new algorithm to compress video images to a fraction of their current size. It is more likely that the computer science professor will tell other computer science professors, and the programmer will share knowledge of the new algorithm with other programmers who speak his/her language (i.e., horizontal), than either of these innovators immediately sharing their findings with an educator who is intrigued by using video segments and on-line journals on a class web page (i.e., vertical). Homophily therefore can act to slow down the rate of diffusion in a system, thus requiring the work of change agents with various opinion leaders in a system. New ways must be found to encourage more heterophilous communication in the current university structure of disciplines and specializations that encourage homophilous exchanges.
Gilbert (1996) promotes the development of institution wide, collaborative communication networks encourage and promote the diffusion of information technology. He provides guidelines for forming a local Teaching, Learning, and Technology Roundtable (TLTR) that would include two categories of faculty (both early adopters and mainstream), representatives from service organizations (such as library, computing centers, faculty teaching development office, student affairs, facilities management), the Chief Academic Officer and or President, student representation, and a TLT Roundtable Coordinator. The TLTR would be responsible for developing integration plans that address the needs of current, mainstream adopters, by capitalizing on the knowledge and skills of EAs, and the support structures of various campus organizations. No individual faculty member can find or know all teaching options using information technology that may be used for a particular course, much less across campus. Thus, mechanisms for sharing valuable information among faculty and others must be provided (Gilbert, 1996). Mainstream faculty have to contribute their point of view, different motivations, and needs so that a common ground can be reached between early adopter fluency and skill and campus-wide requirements. By organizing a TLTR, heterophilous communication would become part of the universityís culture and technology implementation and integration strategy.
There is valuable information to be gained from the early adopterís knowledge and skill as a technology user and integrator and the mainstreamís reaction to being new users. For example, from a human-computer interface (HCI) perspective, we can determine from EAsí experiences what obstacles or incentives within the computer systems themselves encouraged the development of their competence (Bannon, 1991; Weber, 1990). Bannon (1991) discuss the need for both novice and expert user input when designing computer systems. Valuable information can also be obtained from first time learnerís experiences with computer systems or applications (Howard, 1994). Bannon (1991) also suggests there is much to be learned from an examination of how expert users became competent, skilled users of a system. They can provide information on the obstacles and incentives there are within a system to encourage the growth of competence. In the same way that designers should include both novice and expert usersí perspectives and feedback when developing systems or applications, administration should include both early adopter and mainstream faculty in the development of technology integration plans and strategies.
As noted above, the main reasons that mainstream faculty hesitate to adopt are the lack of effective training and support. A number of different approaches to maximize the communication impact of early adopter knowledge and skill on training come from the literature. Brace and Roberts (1996) suggested developmental workshops, orientations, and one-on-one training sessions. However, integration plans have to take into account that EAs are faculty members with teaching, research, and service workloads much like other faculty. Thus, without release time for the EAs to instruct the mainstream, much of the training and daily support will have to come from other service units on campus. Most institutions did reasonably well in the past 10 years at developing support services appropriate to the character and needs of EAs. However, proportionally more support will be required because the mainstream is more numerous, and those providing it will need better and more varied interpersonal skills and sensitivity to deal with the easily bruised egos of faculty who lack the "special propensity for technology" (Gilbert, 1996) that characterizes the early adopter.
One way service units can capitalize on the knowledge of EAs is by including them in the development of training modules that can be used by service units for workshops (Foa, 1993). This approach must address release time and the merit system for EAs, and the increased financial and human resource needs of service units. Gilbert (1996) suggests involving undergraduate students in the mainstream faculty development plan. Many undergraduates have better skills and more current knowledge about information technology than most faculty and staff members (Gilbert, 1996). Student assistants can help increase the use of information technology for teaching and learning, and alleviate some of the financial and human resource costs of support units, resulting in a win-win situation for the institution, faculty, and students. Students benefit by developing both instructional and technological skills that increase their employment marketability. Another option for increasing the quality and availability of support services while holding down costs is to engage early adopter faculty as peer mentors (Gilbert, 1996) and thus increase the impact of their opinion leadership. Stipends, release time, and professional recognition through the merit system can be used to provide incentives for this type of knowledge sharing and interpersonal communication between heterophilous groups.
Roundtable discussions between different representatives and stakeholders on campus must recognize the importance of on-going support and recognition of integration efforts by mainstream faculty. Integration takes time, there are a number of barriers and pitfalls, and progress often seems painfully slow. Faculty members and educational institutions are more likely to participate in gradual change rather than making a sudden, diametrically opposite choices (Gilbert, 1996). Smith (1996) summarizes an iterative technology integration process, which includes awareness and interest, planning and design, support and development, refinement and delivery, assessment, and research. Faculty will want to assess whether their uses of technology for teaching and learning are having any effect. Roundtable discussions have to focus on the successes and failures in order to make relevant changes to the process. It will take time to move through the iterative integration cycle, to implement and then assess the results of innovative efforts, and conduct research on the relative benefits. During the confirmation stage, the individual wants supportive messages that will prevent dissonance from occurring (Rogers, 1995). Recognition for faculty efforts must be provided at each step through incentives and encouragement.
Rely on Natural Diffusion Patterns
It seems apparent that there is much we can learn from EAs about possible uses of technology. As opinion leaders, EAs can persuade other faculty to adopt. An alternative to learning from the experiences and characteristics of EAs is to maintain the status quo and rely on natural diffusion patterns of adoption based on critical mass. Individual efforts will continue to be scattered throughout an institution, and eventually these may be adopted by the mainstream. This is not a completely negative scenario for EAs. A collective administrative effort that is developed "top-down" may stifle creativity and initiative by imposing arbitrary and bureaucratic organizational constraints, such as defining policies about the "right-way" to integrate technology for teaching and learning. EAs will continue to flourish in a status quo model because of their interpersonal networks. Few instructional technology theories, laws, and principles have stood the test of time and rigorous validation. The field is still new and constantly evolving because of technological advancements and developments that present new challenges to researchers and educators. EAs will continue to exchange information and develop their knowledge and skills as they wrestle with these challenges. However, a status quo approach ignores the different needs and characteristics of the mainstream. Enthusiastic beginners may be discouraged by unexpected technological barriers, the lack of training and on-going support, and decide to actively reject technology. By not studying and learning from EAs we can continue with "business as usual" and attempt to maintain campuses as they exist now. Administration can focus mainly on technological infrastructure, and individual departments can continue to ignore faculty training and support. However, this strategy greatly impedes participation by the mainstream.
An alternative, which deserves additional research and more commentary than it will be given here, is to rely on the increased interrelatedness of various disciplines as they investigate common (but complex) questions to do with technology. Faculty, who are experts in their diverse fields, are often self-constrained by their homophilous, horizontal communication social systems. However, technology seems to be a catalyst for bringing the basic and applied research findings of different disciplines to bear on common questions that require contributions from each part of science in order to better understand the whole. Communication technology facilitates this interdisciplinary exchange. In a recent issue of the American Psychological Association Monitor, Beth Azar (1998) describes how interdisciplinary data pooling and sharing is becoming more common and easier in the social sciences using databases that are accessible using the Internet.
The technology itself also seems to demand and require multidisciplinary collaboration for research to progress. For example, computer scientists were pioneers in investigating the nature of artificial intelligence. Investigations into programming a machine to think, however, requires an understanding of the nature of thinking. Computer science has not traditionally focused its research efforts on teaching, learning, and human development, but an investigation of artificial intelligence demands a better understanding of the human mind, and results in the growth of such disciplines as cognitive science.
The multimedia design and development process is also collective effort requiring diverse individuals with graphics skills, video and audio design skills, content expertise, programming knowledge, and instructional design skills for software development (Liu, Jones, and Hemstreet, 1998). Faculty members who are interested in developing multimedia applications often need to draw upon the skills and capabilities of individuals outside their discipline.
Disciplines, although still distinct, are becoming more interrelated as they investigate common (but complex) questions related to technology. The increase in cross-disciplinary basic and applied research and development may play a role in increasing heterophilous communication about the applications and integration of technology, which may lead to more widespread diffusion of technological innovations on campus.
In three main sections, this literature review discussed
the characteristics that differentiate EAs from others, the implications
of developing a long-term campus-wide plan based on the characteristics
of EAs, and summarized some alternatives to building from such pioneers.
Rogersí (1995) theory of the diffusion of innovations provides a theoretical
framework for the present investigation. Survey and interview methods will
be used to build and extend upon Rogersí (1995) theory with respect to
current adoption patterns and characteristics of EAs and mainstream faculty.
The guidelines for conducting on-line research will be discussed with respect
to how they were applied in the present study in Chapter Three. Literature
on the nature of teaching excellence and the development of teaching expertise
was reviewed, and the relationship between early adoption and expert teaching
will be explored in this study using qualitative methods. Links between
diffusion research and the design of the present investigation were identified.
Finally, this review discussed some recommendations from the literature
for developing long-term plans for campus-wide diffusion of technology
for teaching and learning using Rogersí (1995) innovation-decision process,
and alternatives to building from EAs. The results of the present investigation
will used to discuss current adoption patterns and characteristics of faculty
who integrate technology for teaching and learning, explore web-based research
methods, and to develop recommendations for bridging the gap between EAs
and mainstream faculty.
© Dawn Michele Jacobsen 1998