A Study on Learning Effect of Remedial Instruction Using Two- tier Bayesian Network Adaptive Learning System - Taking “Ratio, Proportion and Direct Proportion” Unit as an Example

• Main Authors: Lin,Yuhua, 林玉華
• Other Authors: Shih,Shuchuan
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/29451962960760562635

碩士 === 國立臺中教育大學 === 數學教育學系在職進修教學碩士學位班 === 99 === Abstract 　　This study aimed at exploring the effects of of remedial instruction using two- tier Bayesian network adaptive learning system on students’ learning achievement. The research field focused on the “ratio, proportion and direct proportion” unit in sixth grade and three different remedial instruction models (experiment group I: adaptive computerized self-learning, experiment group II: information technology integrated instruction and control group: traditional teaching) were compared in this study. The teaching media used in the two experiment groups were the teaching animation designed using cognitive conflict theory to eliminate students’ misconceptions in the unit. In addition, the differences in the remedial instruction effect and the learning retention effect of different gender students in the experiment groups were discussed. Finally, the study also analyzed students’comments on the application of the two-tier Bayesian network learning system to self-learning or information technology integrated instruction. 　　The research took a quasi-experimental approach using pretest-posttest, nonequivalent group design. The subiects were 6th grade students chosen from an elementary school in Taichung city. Experiment group I used the adaptive computerized self-learning model, experiment group II used the information technology integrated instruction model and control group used the traditional teaching model. After the experiment was completed, there were 27 effective samples in each group. 　　The results of research were briefly outlined as follows: 1. In terms of remedial instruction effect and the learning retention effect, the effect of experiment group I was significantly better than the effect of control group. The other groups did not show significant differences. 2. In experiment group I, students with high-scores and low-scores also performed significant progress on their average scores. In experiment group II, students with low-score performed significantly better in the post-test than in the pre-test, but high-score students did not show any significant difference between the post-test and the pre-test. 3. The remedial instruction effect and learning retention effect were not significantly different for different gender students in two experiment groups. 4. The average progress rates of misconceptions in three groups were compared as follows: experiment group I (7.68%) > experiment group II (7.00%) > control group (5.08%). But there were no significant differences between groups. 5. The average progress rates of math concept in three groups were compared as follows: experiment group I (19.61%) > experiment group II (17.65%) > control group (12.64%). But there were no significant differences between groups. 6. More than 90% of the students in the experiment groups held non-negative opinion toward the application of the two-tier Bayesian network adaptive learning system. It showed that students respond actively to the application of the two remedial instruction models.