| Summary: | 碩士 === 亞洲大學 === 資訊工程學系碩士在職專班 === 97 === Abstract
This research aims to establish a knowledge structure and Bayesian networks based mathematical teaching materials and computerized adaptive testing. First, we analyzed the content of the textbook, and established the expert knowledge structures of the content. According to the expert knowledge structures, indicator, sub-skills, and mistaken types which can be calculated in the Bayesian networks, items were designed. After the pre-test, ordering theory is used to decide the students' knowledge structure and those parameters were used in the item bank. At the same time, establish a knowledge structure based mathematical teaching materials which can be used by experimental teaching. After experiment, analyze the data by Bayesian probability statistics method, inspect the prediction accuracy of the sub-skills and mistaken types in the situations of adaptive test and completely tested.
Some findings are briefly outlined as follows:
1. After taking the experimental teaching, experimental group students' average grades were better than control group’s significantly (87.28>77.94). It shows that this teaching material is valuable for learning.
2. After experimental group students taking the remedial instruction, they have significant progress on their average grades (post-test > pre-test,94.76> 87.28). It shows that this teaching material is valuable for remedy.
3. The rate of saving items by Bayesian networks computerized adaptive testing (BNAT) is above 49% and prediction accuracy can reach over 94%.
4. With regard to the effect of BNAT, the prediction accuracy of the mistaken types and sub-skills is 93.1% in pre-test and 95.4% in post-test.
5.The compare the performance of the achievement about students among the different learning styles,and there was no sibnificant correlation between the learning styles and the achievement about students.
|
|---|
