Altogether there are 72 students participating in this research, all of them undergoing an experimental course of multi-media animation. The students are divided into three groups, in which they shall respectively establish a personal concept map of the designated unit after the curriculum. In order to diminish the external interference, the whole process of modeling and interconnection experiment proceeds under anonymity. Besides, in order to enhance students’
understanding of the concept maps, a class will be given beforehand to teach them how to construct concept maps.
3.2 Models
3.2.1 The model of modeling and interconnection design
The following is the statement of modeling and interconnection constructional model in concept maps system(Figure.1). The teaching method is divided into three units, and each student should construct his own concept map in accordance with his appointed lesson. After completing his own, each student shall observe all the others’ concept maps through the system, make a comparison and judgment in another two units and pick a person as the choice to do the concept interconnection. The Choice will determine whether to interconnect with the chooser or not by the quality of the chooser’s concept maps, furthermore give comments on the concept maps. If the Choice denies concept interconnecting with the chooser, the chooser
has to find in the same unit a willing substitute, revise his own constructed concept map on the basis of the integration of concept interconnection, then go through the modeling and interconnection again to pick a second Choice. For example, A1, who constructs his concept map in Unit 1, chooses B2 in Unit 2 as his choice 1 and C12 in Unit3 as his choice 2. C12 in Unit3 consents to interconnect with A1 on the concept maps and gives some comments on A1’s concept maps. But B2 in Unit 2 considers A1’s concept maps to be of poor quality and refuses to interconnect with him, and B2 also comments on A1’s concept maps. In this case, A1 has to choose B12 in Unit2 to be his Choice 1. Accordingly, everyone needs to pick out concept maps from two different units to do the concept interconnection and join them into a complete curricular cognition construction. Then, modify his own concept maps according to the concept combination and the feedback from peers. Finally, make again a choice to do concept interconnection in the second round of modeling and interconnection.
3.2.2 Assessment design
To achieve the target of this research, five phases have been devised in this experiment(Figure.2): (a) IMPROVE pre-test, (b) peer judgment, (c) peer feedback, (d)
Unit1 Unit2 Unit3
survey of interconnection & systematic user, (e) IMPROVE post-test. The experiment time totals a succession of four weeks, three hours within each week.
The IMPROVE meta-assessment utilized in this research has adopted the four perspectives of meta-cognition put forward by Mevarech & Kramarski (1997), together with the curriculum, to design its questionnaire. Each perspective has eight related questions, which are used to find out and assess the students’ level of awareness on curricular concepts.
For example: 1. Comprehension: Have you ever thought of the terms for constructing Flash concept maps? 2. Connection: Have you ever thought of the correlation between ActionScript and programming? 3. Strategic: Have you ever considered how to edit ActionScript? 4.
Reflection: Have you ever thought of referring to others’ concept maps in order to introspect your own for improvement? To assess the reliability of this whole measurement, we use Crobach’s α cofficient to execute an internal unanimous reliability analysis. The results of the analysis are respectively .8736. Besides reliability analysis, we also adopt factor analysis to measure the validity (see the appendixB for details).
In the first stage of modeling and interconnection, we employ giving peer judgment to IMPROVE pre-test
The first week The second week The third week The fourth week time Figure.2：methods used in the study
achieve peer inter-grading. As for the standard, we adopt “Novak and Gowin’s concept map construction” to undertake an open-ended questioning, in order to prevent students from scrawling the concept maps or sloppily giving out judgment. For the curricular achievements, there will also be a grading part on personal concept maps and peer judgment(see the appendixB for details). After modifying concept maps in the second stage, we use peer feedback examination to assess how much the students have gained through peer feedback.
There are two parts in this examination: 1. To know the quality of modeling and interconnection. For instance, “Did I make my choice according to his/her ability of ‘Novak and Gowin’s concept map construction’?” 2. To understand how much the students care about feedback. For instance, “I will modify my own concept maps because my classmates’
judgment tells me to do so?” They are respectively asked in an open-ended way (see the appendixB for details). Finally there will be three experts to judge the quality according to the answers to above-mentioned examination.The third stage, the survey of modeling and interconnection & systematic user, investigates in a closed method (see the appendixB for details) , hoping to understand if students will change their choices in the second interconnection, for example, “Are your second choice and first choice the same person?”, and to know if using the system is helpful to the students and their points of view on it, like
“Do you think it helpful for integrating the concepts by model and interconnecting with your peer’s concept maps?”