Part of the course requirement asked each team to design a Java program to simulate the evolution of any species. For example, one team simulated the evolution of zebra and how body strips of zebra adapted to various environments; while another team simulated how computer virus evolve to survive from anti-virus programs and successfully infect programs of other online computers. The program should include various evolution conditions, such as mutation, breeding, enemy, food supply, and color adjustment with environment. Each team had to upload the design task to the system four times in the stages of focus question, design event/object, design methodology and initial outcome, as well as final outcome. The uploaded tasks (in Chinese) were displayed at the following web site: sandy.cis.nctu.edu.tw/~colearn/
page1.html.
The web-based collaborative design system, CORAL Vee, (in Figure 2)
contained seven modules: user registration, system announcement, chat room, BBS, task demonstration, simulation experiment, and production. Figure 3 depicted
homepage of the electronic Vee diagram. The system was performed by retrieving and storing DBMS’s (Data Base Management System) information through CGI program.
Detail description about the system refers to Sun and Lin (in press).
(Insert Figure 2 and Figure 3 here)
Two raters, a doctoral student and a master student in computer science who had obtained two-year training of educational research, rated the design projects. They independently graded (scores range from 35 to 100) team projects in four stages and the inter-rater correlations were significant in all stages (t = 0.51 ~ 0.67, all ps < .01).
Two raters also worked together in content analysis about the online discussion.
Content analysis revealed three dimensions of online discussion: the participation rate (raw number of discussion sentences), the creativity of design (raw number of
innovative design ideas and how flexible is a team in designing evolution program), the cognitive procedure of design, and social cues. Any sentence could conceivably contain several ideas, so the base “unit” of the analysis was not a sentence, but a paragraph. When one paragraph contained two ideas, it was counted as two separate units. Two raters coded paragraphs separately along four dimensions and the
inter-rater consistency was 72% for the creativity of design, 75% for the cognitive procedure, and 76% for social cues. Hara, Bonk, & Angeli (2000) suggested that such coding agreement among raters is adequate given the subjectiveness of coding
criteria.
III. Findings
III.1. The high competent and less competent groups
In Table 1, two teams (nos. 4 and 5) gained highest average scores and thus were categorized as the high competent teams and they also produced substantial amounts of online discussion sentences (1816 and 807). While two other teams (nos. 11 and 5) gained lowest scores were named less competent teams and produced few online discussion sentences. Thus, those who performed better in design evolution programming project discussed more.
(Insert Table 1 here)
III.2. The amount of paragraphs that the highly competent and less competent groups generated in on-line discussion during programming design
Table 2 shows discussion contribution (raw number of paragraphs) made by each member in four design stages among teams. Obviously, some members produced far more paragraphs than others both in high or less competent teams. For example, in the 4th team if three members contributed equally to discussion, then each member were to generate 33% (82 paragraphs) of total paragraphs. However, the first member generated 45 paragraphs, the second 104 paragraphs, the third 97 paragraphs, and the difference was very significant (X2 = 78.78, p < .01). For the two competent teams, paragraphs produced in various stages were significantly different. The 8th team discussed more in the first and second stages in forming focus questions and design orientation; whereas the 4th team discussed more in the last two stages during design
outcome production. In addition, in various stages high competent teams discussed more than the less competent teams.
(Insert Table 2 here)
III.3. The process of web-based collaborative design
Figure 4 depicted the process of web-based collaborative design observed in the study. This flow chart illustrated the underlining cognitive and metacognitive
processes in discussion as well as their trial and error in designing a Java program of evolution simulation. In the first and second stages, team members mainly engaged in brain storming for a better or innovative focus question as well as design
objects/events. It is very likely that the creativity of a design product is determined in the initial step of designing process. Also, the coding result showed that both high and less competent teams could generate innovative ideas. However, the less competent team generated slightly fewer ideas. The 8th team generated 10 distinctive ideas, the 4th team 5, the 11th team 4, and the 5th team 3.
(Insert Figure 4 here)
However, in the third and final stage of designing, team members were mainly engaged in planning, monitoring, and regulation. When team members went a step further, they always check how well the original plan worked in terms of feasibility, time limit, and functionality. In sum, it is a very typical process of metacognition. The high competent teams displayed a very complicated metacognitive process and that is what the less competent teams lacked.
III.4. The social dynamics through online discussion
In examining the social cues of discussion, team members’ conversation was very formal and polite in the first several days. When the design process went further, they became more and more relax, less formal, and even fooling around. There is always a leader, some active but others passive, made the final decisions or voted critically. Leaders were always in a higher grade level (senior students) and performed leadership in a democratic style. All leaders showed respect to expertise and tried to maintain a volunteering but fair partnership in design.
Acknowledgements
The authors would like to thank the National Science Council of the Republic of China for financially supporting this research under Contract No. NSC
89-2520-S-009-016.
IV. References
Collins, A., Brown, J. S., & Newman, S. E. (1989). Cognitive apprenticeship:
Teaching the crafts of reading, writing, and mathematics. In L. B. Resnick (Ed.), Knowing, learning, and instruction: Essays in honor of Robert Glaser (pp. 453-494).
Hillsdale, NJ: Lawrence Erlbaum.
Hara, N., Bonk, C. J., & Angeli, C. (2000). Content analysis of online discussion in an applied educational psychology course. Instructional Science, 28, 115-152.
Henri, F. (1992). Computer conferencing and content analysis. In A. R. Kaye (ed.), Coolaborative learning through computer conferencing: The Najaden papers (pp.
115-136). New York: Springer.
Novak, J. D., & Gowin, D. B. (1984). Learning how to learn. Cambridge University Press, Cambridge.
Schrage, M. (1995). No more teams: Mastering the dynamics of creative collaboration. New York: Currency Doubleday.
Sun, C. T. & Lin, S. S. J. (In press). Network-based Cooperative Design: Environment, Learning, and Evaluation. IEEE Transaction on Education.
Table 1
Scores and online discussion sentences of high and less competent teams in four stages of design process.
competent 4 85 50 80 90 76.25 807
11 100 80 45 40 66.25 114
Less
competent 5 80 45 60 60 61.25 340
Table 2
Numbers of online discussion paragraphs contributed by each team members in the high and less competent teams in four stages of programming design process and Chi square examination of frequency homogeneity.
Team
Total 217 112 65 49 442 155.28(3)**
X2 (df) 41.48(2)** 51.95(2)** 18.12(2)** 32.49(2)** 125.39(2)**
1 7 7 15 16 45
2 20 4 23 57 104
3 15 9 33 40 97
4
Total 42 20 71 113 246 78.78(3)**
X2 (df) 6.14(2)* 1.9(2) 6.87(2)* 22.53(2)** 25.34(2)**
Figure 1. The Vee Heuristic
Figure 2: The homepage of CORAL-Vee for the promotion of students’
collaborative design. Web address: sandy.cis.nctu.edu.tw/~colearn/ page1.html.
F o c u s Q u e s tio n s
A c ti v e I n te r p la y
C o n c e p tu a l M e th o d o lo g ic a l
W o r ld V i e w s P h ilo s o p h i e s T h e o r ie s P r in c ip le s C o n s tr u c ts
C o n c e p tu a l S tr u c tu r e s S ta te m e n ts o f R e g u la r itie s C o n c e p t D e f in itio n s
C o n c e p ts
V a lu e C la im s
K n o w le d g e C la im s I n te r p r e ta tio n s E x p la n a tio n s G e n e r a liz a tio n s R e s u lt s
T r a n s f o r m a tio n s F a c t s
R e c o r d s o f E v e n ts R e c o r d s o f O b je c ts E v e n ts /O b je c ts
Figure 3: Homepage of Vee diagram for collaborative design.
31