1. The Use of Multiple Representations in a Web- Based and Situated Learning Environment 指導教授： Ming-puu,Chen 報 告 者： Yun-fang,Chou 報告日期： 2007/3/31 Hsu, Y. S., & Hwang, F. K. (2002). The use of multiple representations in a web-based and situated learning environment. In Proceedings of World Conference on Educational Multimedia, Hypermedia and Telecommunications 2002 (pp. 816-820). Chesapeake, VA: AACE.

2. Introduction This study attempts to –develop a web-based lesson according to the theory of situated learning. –examine how the Internet supports student situated learning. –investigate how multiple representations support students with different learning preference on the construction of their scientific concepts. –how students benefit from multi-representation simulations in a situated learning environment.

3. Simulations with Multiple Representations(1) Why simulations? Many scientific experiments of natural phenomena take days or months to produce needed data or occur in real time that is too short to observe. Simulations alter the natural time scale to make phenomena more visible to the learners (de Jong & Njoo, 1992).

4. Simulations with Multiple Representations(2) Why simulation is ineffectiveness? 1.Due to learner’s inability to overcome the difficulties of interacting with a simulation on their own (de Jong et al., 1994). Pedagogical support, such as: –reinforcing productive conjectures, –discussing successful learning strategies, and relating phenomena to the discipline could be missing. 2.Learner difficulties could be due to concepts embedded in the simulation that are too abstract or a design of the simulation that is not meaningful to learners. 3.Learners may not know how to use problem-solving strategies, such as the generation and testing of hypotheses, model-based reasoning and metacognition.

5. Simulations with Multiple Representations(3) Simulations should encompass formalized and manipulatable underlying models (de Jong, 1991). The representations of models should be dynamic, visual, realizable and multiple-representational.multiple-representational? Raghavan and Glaser (1995) stated: “Visual models that are dynamic and interactive are presented not only to concretize abstract ideas but also as reasoning tools that give students the leverage to solve problems in a variety of contexts” (p. 37).

6. Simulations with Multiple Representations(4) multiple-representational? Multiple-linked representation between the animation, graph, and symbol forms facilitates students’ meaningful learning (Goldenberg, 1995). Multiple representations of the same phenomena provided by computer simulations make students to move flexibly among different modes of representation in a multiple-representation learning environment (Kafai, 1995). This allows students to choose their own modes of information representation on the computer screen and their own ways to solve problems (Windschetl, 1995).

7. Simulations with Multiple Representations(5) multiple-representation? Many cognitive researchers assert the importance of representation in problem solving and conceptual change (Greeno,1989; La, 1983; Newell & Simon, Diehl, Ranney, & Schank, 1972). From epistemological view (Confrey, Smith,Piliero & Rizzuti, 1991), multiple representations have the potential –to highlight different aspects of the concept, –to lead to a convergence across representations that may improve or strengthen the depth of understanding, –to promote examination of the conflict among representations, and –to allow for assessing how changes in one representation affect another.

8. Simulations with Multiple Representations(6) Computer-based multiple representations Computer-based multiple representations includes –the representations with literal features that correspond to those in the real world (e.g., video and images etc.), –the representations designed to correspond to the symbolic expressions (e.g., animations dynamic graphs, tabular data and equations etc.) used by experts, and –the representations designed to correspond to the conceptual entities and events (e.g.,the modeling process) in the minds of experts (Kozma, et al., 1996). Computer-based multi-representations may connect with conceptual entities and mental models if students are actively exploring the phenomena.

9. Method / Participants Method: quasi-experimental method along with semi-structured interviews. Participants: –second-year senior high schools in Taipei. –There were four classes who enrolled in Earth Science class at School A : 44 males and 42 females. School B : 16 males and 30 females. –Totally, the participants were 132. –After excluding invalid data, 110 (49 males and 61 females).

10. Instrumentation Web-based lesson: (called as Lesson Rainbow) a test –the concepts related to rainbows, humidity, and condensation( 冷凝 ). –There were 23 items on the test, which were validated and was examined for reliability (The Cronbachα= 0.76) before the formal experiment. a questionnaire –with 34 Likert-type items was used to conduct student opinions about the design of this web-based lesson. –Four dimensions :interface design, situation design, the design of learning tools and the overall design of Lesson Rainbow. –The Cronbachα= 0.87. a semi-structured interview –investigate student understanding of the relative concepts and to collected their opinions about the lesson. –20 students (8 males and 12 female s) were selected for interviews based on specific purposes. Each student was interviewed for 20-30 minutes.

11. Lesson Rainbow

12. Procedure 1.The preparation stage (01/2000~11/2000): –to develop Lesson Rainbow and the instruments. –After a pilot study, the instruments were validated and revised for the experimental stage. 2.The experimental stage (12/2000~1/2001): –Before the experiment : a pretest. –In the experiment : a training session for 1hour & completed Lesson in 2 hours. –After the experiment, a posttest on the concepts & the questionnaire were administrated. –The selected students received follow-up interviews a few weeks later. 3.The data analysis stage (1/2001~5/2001): –Data analysis and concluding remarks were the major jobs in this stage.

13. Data Analysis 1.Paired t-test ： –compare the pretest and posttest scores on concepts. 2.Repeated ANOVA ： –test the hypotheses that stated male and female students had a significant difference before and after the experiment. 3.Descriptive statistics ： –analysis of each item in the questionnaire. 4.Chi square ： –test if there was a significant difference among the four dimensions in the questionnaire. 5.Qualitative data –to show a deep understanding of student learning processes and perspectives using the web-based lesson, Lesson Rainbow.

14. Results and Discussion

15. Interview: The web-based lesson was more interesting than textbooks because of the animations and interactions. The online discussions made students feel more involved in the learning activities. Five of the twenty students suggested that the teacher should participate in the online discussion with them. Few of them need teachers’ assistances because they are lack of skills on communication and reflective thinking.

16. Conclusions The animations for the simulated authentic situations could promote their learning motivation and immerse them in an interesting context for meaningful learning. In order to reach effective cooperation, students needed to share ideas, adventure and argue with others so that they could come to reasonable interpretations of the subjects they were studying (Blumfnfeld, et al., 1997) Students can compare the varied views of a topic and enhance cohesive understanding of science from a well-designed online asynchronous discussion. Online asynchronous discussion designed for a situated learning environment can promote knowledge integration.