Adopter focused development and diffusion theory

The following is a first draft of the next section in Chapter 5 of the thesis. It’s the first section which starts describing the various different changes that were made to Webfuse and how it was supported from 2000 and on. IMHO, this particular change is something that continues to be missing from almost all university attempts to support e-learning. If not missing, they simply haven’t recognised the need for greater levels of skills. It is particularly sad that my current institution, the institution at which these changes/lesson were first implemented and written about 10+years ago, still hasn’t learnt the lesson – or perhaps the lesson just isn’t important enough.

Adopter focused development and diffusion theory

By 1999, the Webfuse experience led Jones and Lynch (1999) to identify three problems facing the development of web-based learning – appropriation, adoption and evolution – and propose a model for the design of web-based system, particularly for Web-based learning. The three problems were described as follows:

  1. Appropriation;
    As described in Chapter 4, only a small number of staff were heavily using web-based learning. These innovative staff members were developing a number of interesting and useful applications of web-based learning. However, few of these applications were being appropriated for use by the remaining teaching staff. This mirrored experience reported elsewhere (Geoghegan 1994; Taylor 1998; Mendes and Hall 1999).
  2. Adoption, and;
    Beyond appropriation of innovations, there was a broader problem with limited adoption of web-based learning and instructional technology in general (Surry and Farquhar 1997). Jones and Lynch (1999) identified the distinction between developer-based and adopter-based development methodologies as an explanation for this limited adoption. The pre-dominant developer-based approaches assume that a demonstrably better artifact will automatically replace existing products or practices (Surry and Farquhar 1997). By not paying attention to the users and the context, developer-based approaches create systems that are difficult to use and provide little benefit for the user.
  3. Evolution.
    Traditional development approaches aim to develop an “ideal” or all-encompassing system that meets all needs. Jones and Lynch (1999) argue that the time and resources necessary to achieve this goal generates little pay off due to the ever-changing requirements of web-learning. It is suggested that as the context changes, these ideal systems become a burden preventing adaptability (Jones and Lynch 1999).

To address these issues Jones and Lynch drew on insights from diffusion theory (Rogers 1995), related insights from adopter-based development (Surry and Farquhar 1997), the design patterns community (Alexander, Ishikawa et al. 1977; Gamma, Helm et al. 1995), and established links between design patterns and hypermedia templates (Nanard, Nanard et al. 1998) to propose a development model to address these problems. The model can be described as a combination of the following principles or assumptions:

  1. There exists a development team that actively seeks to understand the social context within which web-based learning is occurring. The inter-relationships between the developers of the system, the developed system, the potential adopters of the system and the contexts in which they system is developed and used is of significant importance.
  2. The development team develops a set of constructive templates (Nanard, Nanard et al. 1998) that teaching staff can use to create course websites for use by students.
  3. To encourage adoption and use by teaching staff the design of these templates is informed by insights from diffusion theory.
  4. Template design is informed by design patterns that encapsulate knowledge around learning, teaching and web-based services.
  5. There is recognition that innovative staff members are likely to do unexpected things with the constructive templates, even ignore them all together and use other means. This is allowed, and where possible enabled.
  6. The development team continue to observe and support staff throughout the use of the course sites to identify what is working and what is not.
  7. Based on this observation the development team abstracts new design patterns, retires those no longer appropriate and does the same for constructive templates.

Jones and Lynch (1999) believed that this development model would provide three major benefits:

  1. Develop systems which are more likely to be adopted. This is achieved by a major emphasis on context, adopter led development approaches and theory from the diffusion of innovations.
  2. Enable the appropriation and reuse of prior experience. Gained by a continual process of evaluating the work of innovators for potential abstraction and storage in a pattern repository and implementation as a constructive template.
  3. Enable the continued evolution of the system to meet changing needs. Evolution is provided by the continued application of patterns in a form of piecemeal growth and emphasising design for repair rather than replacement.

By 1999 it was well understood that a significant majority of academic staff, for whatever reason, were not going to construct elaborate course websites. Jones and Lynch (1999) describe the plan to develop a Webfuse Wizard. Built on top of the Webfuse template library, the Wizard would provide a “wizard” interface that would guide teaching staff through the creation of a course site. The working prototype due in July 1999 was not completed due to various factors, including the need for the author to continue teaching. The eventual replacement for the wizard, in terms of providing a easier method for creating course websites, was the development of the default or minimum course site idea described in Section 5.3.5. The principles and assumptions of the development model described by Jones and Lynch were to form the basis for how the Infocom web team operated from 2000 through 2004.

The use of diffusion theory continued and evolved throughout this period. Jones, Jamieson and Clark (2003) propose a model to “aid educators increase their awareness of potential implementation issues, estimate the likelihood of reinvention, and predict the amount and type of effort required to achieve successful implementation of specific WBE innovations”. This model is based on one aspect of diffusion theory and was based on the practices of the Infocom web team during the period 2000-2004. Rather than focus on the potential objective benefits of an innovation (e.g. online assignment submission will result in faster turnaround times) the model recommends an evaluation of the innovation in terms of the likely rate of adoption, perceived innovation attributes, the type of innovation decision, the available communication channels, the social system in which it will be implemented, and the nature of the change agents and their efforts. Jones et al (2003) argue that this type of evaluation is highly context specific to the exent that even with the same innovation and the same institution, as time passes the evaluation should be repeated.

Diffusion theory is not without its limitations or problems. Bigum and Rowan (2004) argue that “the problem for a theory of change that relies on pre-established categories is that it is limited in its capacity to account for new and unanticipated arrangements or orderings”. McMaster and Wastell (2005), in arguing that diffusionism is a myth with a potency based not on empiricial validity but on a synergy with a colonialistic mid-set, offer a description of the many flaws that have been identified in diffusion theory. Amongst these is that diffusion theory is deterministic and positivistic in philosophical orientation which leads its proponents to predict outcomes based on the measurement of a small number of variables (McMaster and Wastell 2005). The “factor approach”, where key features/factors are correlated with outcome measures, is common to diffusionist research in the IS field (McMaster and Wastell 2005). The approach described by Jones et al (2003) is an example of one such approach. McMaster and Wastell (2005) cite numerous authors to support the argument that while such “factor approaches” can “highlight impotant influences, they necessarily fail to capture the dynamic, processual character of social-technical innovation”.

The development model developed by the Infocom web team over the period from 2000 through 2004 addressed the “dynamic, processual character of social innovation” with its particular focus on evolution. Have been guided by “diffusion theory” in its selection of a particular set of changes, the web team maintained a close eye on how these innovations were being used by individual staff. This was enabled by the fact that the web team were not only the developers of Webfuse, they also provided helpdesk support and training for users of Webfuse. The web team was not employed by the central IT division, they were employed by the faculty. This meant the web team had offices in the same building, went to the same team room and attended faculty retreats. These social interactions provided a deeper insight into the experiences of the academic users of Webfuse. This meant that the Webfuse development model move towards what McMaster and Wastell (2005) describe

Here we would argue that the innovation succeeded due to internal development, through a participative process involving strong local leadership, engaged staff and the fortuitous occurrence of a series of local crises that aligned all stakeholders around the need for change.

Bigum and Rowan (2004) argue that a serious problem with diffusion theory is that the innovation is understood to pass through the adoption process largely unchanged meaning that the social is seen either to conform or not to conform with the requirements of the innovation. As an adopted focused development process, the Webfuse development model recognised that not only will the innovations change during the adoption process, great benefit can arise by being able to recognise that change and respond to it in ways appropriate to the social setting. The development model understood and sought to enable the users and consequence of the innovations to emerge from the complex interactions between the social and technical. As described by Markus and Robey (1988) this type of perspective replaces prediction with a detailed understanding of dynamic organisational processes, the intentions of actors and features of information technology.

References

Alexander, C., S. Ishikawa, et al. (1977). A Pattern Language: Towns, Buildings, Construction, Oxford University Press.

Bigum, C. and L. Rowan (2004). "Flexible learning in teacher education: myths, muddles and models." Asia-Pacific Journal of Teacher Education 32(3): 213-226.

Gamma, E., R. Helm, et al. (1995). Design Patterns: Elements of Reusable Object-Oriented Software. Reading, Massachusetts, Addison-Wesley.

Geoghegan, W. (1994). Whatever happened to instructional technology? 22nd Annual Conferences of the International Business Schools Computing Association, Baltimore, MD, IBM.

Jones, D., K. Jamieson, et al. (2003). A model for evaluating potential Web-based education innovations. 36th Annual Hawaii International Conference on System Sciences, Hawaii, IEEE.

Jones, D. and T. Lynch (1999). A Model for the Design of Web-based Systems that supports Adoption, Appropriation and Evolution. First ICSE Workshop on Web Engineering, Los Angeles.

Markus, M. L. and D. Robey (1988). "Information technology and organizational change: causal structure in theory and research." Management Science 34(5): 583-598.

McMaster, T. and D. Wastell (2005). "Diffusion – or delusion? Challenging an IS research tradition." Information Technology & People 18(4): 383-404.

Mendes, M. and W. Hall (1999). "Hyper-Authoring for Education: A Qualitative Evaluation." Computers and Education 32: 51-64.

Nanard, M., J. Nanard, et al. (1998). Pushing Reuse in Hypermedia Design: Golden Rules, Design Patterns and Constructive Templates. Proceedings of the 9th ACM Conference on Hypertext and Hypermedia, ACM.

Rogers, E. (1995). Diffusion of Innovations. New York, The Free Press.

Surry, D. and J. Farquhar (1997). "Diffusion Theory and Instructional Technology." e-Journal of Instructional Science and Technology 2(1): 269-278.

Taylor, P. (1998). "Institutional Change in Uncertain Times: Lone Ranging is Not Enough." Studies in Higher Education 23(3): 269-278.

4 thoughts on “Adopter focused development and diffusion theory

  1. Pingback: The road not taken « The Weblog of (a) David Jones

  2. Pingback: Emergent and agile development « The Weblog of (a) David Jones

  3. Pingback: Course websites and “libertarian paternalism” « The Weblog of (a) David Jones

  4. Pingback: Workarounds « The Weblog of (a) David Jones

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s