Meta: Exploring the potential of a minimally adaptive intelligent tutoring system to support comprehension of grade appropriate texts
Killips, Laura (2019)
2019
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
http://urn.fi/URN:NBN:fi:amk-2019121226188
http://urn.fi/URN:NBN:fi:amk-2019121226188
Tiivistelmä
Background: In spite of evidence that demonstrates the impact explicit strategy instruction can have on students’ ability to comprehend text, few educators include these lessons in their classes. As a result, students graduate without the appropriate literacy foundation. Given the time commitment necessary to develop teacher practice and the urgency of this issue, it is useful to consider alternative methods of support.
Purpose: The study’s objective was to explore the potential of a minimally adaptive intelligent tutoring system that provides explicit strategy instruction for comprehension monitoring. The question was whether or not such a system could raise student awareness of their understanding as they read, which would give evidence as to the system’s potential to support comprehension.
Methodology: The research was designed as a proof-of-concept, randomized control study. Students in both the control group and the intervention group were given a pre-test that assessed metacognition and their ability to paraphrase text. Half of those students then participated in the intervention where they received explicit instruction in comprehension monitoring through an intelligent tutoring system prototype. Then students in both groups completed a post-test to measure growth.
Results: Results were mixed. A paired-sample t-test was used to determine that there was not a statistically significant difference in the change from pre-test to post test in students’ ability to note a planted error; however, there was a statistically significant difference in the students’ ability to notice vocabulary breakdowns. Additionally, there was evidence of accelerated growth on several measures for the intervention students as compared to the control students.
Conclusions: There is evidence to suggest that a minimally adaptive intelligent tutoring system has the potential to support comprehension of grade level text. In order to further explore the extent to which such a system can provide sup-port, the student model should be built out further in order to capture all parts of the metacognitive process.
Purpose: The study’s objective was to explore the potential of a minimally adaptive intelligent tutoring system that provides explicit strategy instruction for comprehension monitoring. The question was whether or not such a system could raise student awareness of their understanding as they read, which would give evidence as to the system’s potential to support comprehension.
Methodology: The research was designed as a proof-of-concept, randomized control study. Students in both the control group and the intervention group were given a pre-test that assessed metacognition and their ability to paraphrase text. Half of those students then participated in the intervention where they received explicit instruction in comprehension monitoring through an intelligent tutoring system prototype. Then students in both groups completed a post-test to measure growth.
Results: Results were mixed. A paired-sample t-test was used to determine that there was not a statistically significant difference in the change from pre-test to post test in students’ ability to note a planted error; however, there was a statistically significant difference in the students’ ability to notice vocabulary breakdowns. Additionally, there was evidence of accelerated growth on several measures for the intervention students as compared to the control students.
Conclusions: There is evidence to suggest that a minimally adaptive intelligent tutoring system has the potential to support comprehension of grade level text. In order to further explore the extent to which such a system can provide sup-port, the student model should be built out further in order to capture all parts of the metacognitive process.