| Summary: | 碩士 === 國立臺中教育大學 === 教育資訊與測驗統計研究所碩士在職專班 === 106 === This study aimed to develop a one-on-one mathematics intelligent tutoring system (ITS) for addition and subtraction of fractions with different denominators based on the one-on-one dialogue design of AutoTutor, and assess its learning effect. In the one-on-one dialogue math AutoTutor system, the computer agent asked several questions which specifically designed for this study, and through a series of natural language interactive dialogues between the student and the computer agent, the student's answer was compared with the expected answer and the misconception to provide appropriate feedback for each student. These feedbacks helped students clarifing the misconception and constructing the correct concept of mathematical knowledge.
Participants of the study were fifth graders in Taichung city. All 91 students were divided into the experimental group and the control group. The experimental group used the one-on-one dialogue AutoTutor system for remedial instruction, whereas the control group used traditional classroom learning for remedial instruction. The teaching materials for both groups were the same. The pre-test and post-test were conducted before and after the lessons. The postponed test were condute and after seven weeks. Finally the differences of learning effect and learning interest were analyzed.
The results of the study showed that the average scores of post-test and postponement-test were better than the average score of pre-test both. That was to say, there were obvious progressions in the two different remedial instructions. Comparing the difference of two remedial instructions in the post-test, it also showed that for the students with medium and low ability, or the more difficult concept about subtraction of fractions with different denominators, the learning gains in the post-test of one-on-one dialogue math AutoTutor system were all significant superior to those of the traditional classroom teaching. However, postponed-test was no significant difference in terms of learning progress between the traditional teaching.
Further analyzing the difference of the two different remedial instructions in learning interest, it can be found that the students of the control group did not achieve significant differences in learning interest before and after teaching, but the experimental group did. In the experimental group, there were all significant progressions regardless of ability levels or gender.
In conclusion, the one-on-one dialogue math AutoTutor system could enhance learning interest and had more positive teaching effect. Students could concentrate on learning, and got better learning gains through the one-on-one dialogue math AutoTutor system.
|
|---|
