Article Review #5 – Mendes et al

Chris Wills

Mendes, E., Mosley, N., & Counsell, S. (2001). The cognitive flexibility theory: an approach for teaching hypermedia engineering. SIGCSE Bulletin , 33 (3), 21-24.

What was/were the research question(s) in the article?

In this article, the authors presented the results of an attempt to teach web engineering principles, a segment of a larger hypermedia engineering class, using cognitive flexibility theory (CFT) as an instructional theory.  This article is different from others we have read, in that the authors did not state a research question, but instead appeared to be giving CFT a “test drive” as an instructional method.  The unstated research question, therefore, is “using student exam results as the measure, will concepts taught using CFT as the instructional method be learned better than concepts taught using more conventional instructional methods?”

Was the literature review relevant to the research question(s)?

The literature review, although minimal and citing only a handful of sources, was relevant to the research question; the authors note that “to date [CFT] has not been applied as a teaching method in a classroom environment,” which helps to explain the brevity of the literature review.  The primary reference material the authors used to introduce CFT was an article by Spiro et al, published in 1991 and which we’ve covered in class, that the authors used to outline the principles of CFT and to justify its use in “ill-structured domains.”  This source is probably one of the best that could be used, as Rand Spiro is the “father” of CFT.  Other sources cited provide a research-based confirmation of the effectiveness of using CFT principles in the design of hypermedia applications to teach poorly structured domains.

What was the methodology used and was it sound?

The sample for this study consisted of 41 fourth-year computer science students who were enrolled in a twelve-week Hypermedia and Multimedia Systems course at the University of Auckland in 2000.  The coursework covered simple and advanced hypermedia systems, multimedia production, and web engineering; the hypermedia and multimedia class sessions were taught using normal instructional methods, and the web engineering class sessions were taught using a CFT model.

Throughout the semester, students completed three assignments:

• Assignment 1 (group assignment) - required students to use a WWW search engine to find some sites on the Web; coverage of core HTML tags and structure of HTML documents/web sites; creation of a simple homepage; design of a multi-page web site; create a new web site for the Matakohe Kauri Museum, improving on the existing site; load the web pages onto a web server.
• Assignment 2 (individual assignment) – required students to develop a Web site to teach a chosen topic, containing at least 50 Web pages and structured according to the CFT principles. Students could develop their applications using HTML, SHTML or XML. They were taught for 120 minutes about the CFT principles.
• Assignment 3 (group assignment) - required students to design, implement and evaluate a multimedia application; in addition, they were required to give a 15-minute presentation of their application to the class.  Implementation of the application used an authoring package (Director) or Java.

This classroom application of CFT was different from the “standard” application of CFT in that the lecturer chose the information path to be followed, while Spiro’s model focuses on the learner as the person responsible for selecting the information path.  In this environment, instruction focused on the review of case studies selected to further illustrate themes previously explained.

To evaluate the relative effectiveness of CFT as an instructional method, the students were given an examination, presumably at the end of the semester, as it covered web engineering and the other topics taught during the course.

What were the results?

The authors “informally evaluate[d] the effectiveness of the CFT as a teaching practice,” using boxplots to depict results rather than calculating other measures of central tendency.  The median scores for both sections of the exam were similar, but the boxplots revealed that students’ results on the exam subjects taught using CFT produced a flatter, more widely-dispersed distribution than the their results on the exam subjects taught traditionally. 

Were the conclusions consistent with the methodology and results?

Considering, again, that the use of CFT in a classroom setting rather than in hypermedia instruction has not previously been studied, the conclusions appear to be consistent with the methodology and results.  The authors are careful to note that there could have been several factors that contributed to the greater variability, and indicate that their future work will focus on developing a set of tools to aid in generating web pages using a CFT model.

What interested you most about the article? What questions did it raise?

I liked this article because it helped me gain a better understanding of CFT; I struggled with our in-class discussion of CFT, since I wasn’t thinking of it in terms of its utility when designing hypermedia lessons but rather as a method to be used in a traditional classroom.  I read it a few days after we covered the other material, and it gave me a different lens to use when going back and re-reading the Spiro articles. 

I’d like to learn more about CFT’s use in a traditional classroom setting, since it seems like it would take a great deal of effort to adapt a hypermedia-based lesson to a classroom, but doing so would also seem to provide a way to reach more students and to increase the level of collaboration among students.