This chapter aims at illustrating the application of important scientific principles using a sample design-type research project, which featured the development of a method for online shopping support. Existing schools of thought are described as potentially competing paradigms. A deductive approach is utilized to derive the required features of the artifact based on kernel theories. Falsifiability criterion is met by the development of the concrete form (in terms of structure and behavior) and the proposal of specific testable hypotheses. An example of auxiliary protective hypothesis is given. Ockham’s razor is used in order to refute a more complex version of the method.
TopThe Problem
This section aims to demonstrate the application of the scientific principles outlined in the previous chapter to a particular design-type research initiative. Needless to say, it is difficult to show in a comprehensive fashion how these different principles come to guide research process within the boundaries of a single project. A thorough approach would demand a meta-analysis of a particular design-type research stream targeting a given class of problems. Such a project, however is well beyond the scope of the book. Rather, here the purpose is to illustrate the potential applicability of the guidelines advanced by the philosophers of science using the concrete case of design-type research work. The initial stages of the project have been published earlier in (Vahidov & Ji, 2005).
The case of a design project for online shopping support has been chosen in this chapter to serve as an illustration. This particular choice is an interesting one, as it represents a relatively recent type of problems for design-type research. Thus, in this respect it is somewhat premature to speak about an established (design research) paradigm, and a diversity (anarchy) of various approaches may be proposed by the researchers. On the other hand, the approaches may be based on the application of the previously developed generic methods, or adopted from other problem similar problem contexts. In such a case if an online shopping problem can be shown to be belonging to some known class of problems (problem solving, ranking, etc.), then there would be something akin to a pre-existing paradigm expanded to this particular one.
The problem is also interesting from the point of view of “design-as-a-prediction.” Currently, it is not obvious, whether basic online shopping tools would be complimented in some substantial manner by more advanced capabilities. There are many exciting questions related to such a “predictive” perspective. Would the shoppers be willing to use advanced support tools? Would such a usage help companies in increasing their sales and customer satisfaction levels? Would the benefits of provisioning support tools offset the costs associated with their development and maintenance? Are their specific problem contexts or customer classes to which the tools would be beneficial? In short, do these tools have the right for existence? Are they the true IS meta-artifact forms? Only time could help answering these questions, yet design-type research attempts to illuminate the way the future may progress on one hand, and speed up this progress on the other.
The rapid rise of electronic commerce had opened new opportunities for design researchers as it allowed considering innovative ways in which the computational and communication capabilities of computer networks could be employed to improve performance of individuals, groups, and organizations. The degree of automation entertained by different researchers ranged from the effective and efficient delivery of information, to suggestion generation, and to automated decision making, including bidding and negotiation. For example, various software agent concepts were proposed to tackle different phases of buying, in particular those that could automated negotiations on behalf of buyers and sellers (Guttman, Moukas, & Maes, 1998). Some researchers have even speculated that customers in the digital economy would be equipped with powerful computational tools, which could seriously jeopardize businesses, since the assumption of bounded rationality may no longer be valid (Conway & Koehler, 2000).
The basic mechanisms employed by the majority of online retailers are browsing and searching capabilities bundled with the product catalogs (Detlor, Sproule, & Gupta, 2003). These represent fairly marginal improvements over the traditional paper-based catalogs, including: maintaining up-to-date product information; convenience of using drill-down capabilities via links; and quick access to information about product(s) of interest using provided search criteria. Could there be other ways to enhance the shopping experience, while harnessing the computational and communication powers of computer networks?