| Summary: | 碩士 === 亞洲大學 === 資訊工程學系碩士在職專班 === 98 === Abstract
The research takes the “Linear Inequalities in one variable” in mathematics field of junior high school, and designing the test items and digital materials of learning is based on Bayesian Networks and Unit Knowledge Structure, and experimented on junior high school students to prove its effects. In this teaching experiment, subjects were divided into the experimental group and the control group. Students in the experimental group were under the teaching of Information Technology Integrated into Instruction with self-compiled digital materials, and received Information Integrated Remedial Instruction according to the error patterns and misconception of Computerized Adaptive Diagnostic Test. Students in the control group were under traditional teaching with textbooks and received remedial instruction by test reviews. After teaching phase, students in these two groups took Computerized Adaptive Diagnostic Test. The results of the test are aimed to investigate the effects of learning and remedial teaching of the two groups, and estimate the effects of Computerized Adaptive Diagnostic Test. The research findings are as follows.
1. In learning feedback questionnaire analysis, over ninety percent of the students think Information Technology Integrated into Instruction makes their classes more lively, more active, and more learning motivation promoted. Eighty percent of the students think Computerized Adaptive Diagnostic Test is helpful.
2. The effect of the experimental group is significantly superior to that of the control group, showing this digital material can raise students’ learning effects.
3. The effect of the remedial instruction of the experimental group is significantly superior to that of the control group, showing this remedial material and remedial teaching mode are more helpful for students than traditional remedial instruction.
4. Under these two different teaching modes, the average of postponed test in the experimental group is a little higher than that in the control group, though not reaching significant difference.
5. Computerized Adaptive Diagnostic Test based on Bayesian Networks can make teachers and students understand learners’ error patterns and misconceptions so that adaptive remedial teaching can be carried out.
6. In the unit of “Linear Inequalities in one variable”, the average of the recognition rate of optimum threshold is 88.73% , revealing good inferential accuracy of Bayesian Networks.
7. In the unit of “Linear Inequalities in one variable”, the average of the question-saving ratio in the pre-test, post-test, and postponed-test of Computerized Adaptive Diagnostic Test is 30.63%., and the average of its predictive accuracy reached over 93.67%.
8. In the unit of “Linear Inequalities in one variable”, the average of the consistency of every test is 95.18%, revealing high consistency between Computerized Adaptive Diagnostic Test and whole-test diagnosis in the aspect of error patterns, sub-skills, and competence indicators.
Key words: Linear Inequalities in one variable, Information Technology Integrated into Instruction, Bayesian Networks, Computerized Adaptive Diagnostic Test, Information Integrated Remedial Teaching
|
|---|
