Integrated Software for Artificial Intelligence Tutoring and Assessment in Science

The researchers are first developing the procedures for grading student work in an intelligent tutoring system that they have already created. Actual student work samples will be used to test and validate the system's procedures and to conduct an initial external review of the system with a small group of high school chemistry teachers. When the system is fully developed, it will be field tested with chemistry teachers in several high schools. In this quasi-experimental study, 10 to 12 teachers will implement the tutoring system with the assessment component in one chemistry class. In a second class, each teacher will continue normal instruction, not using any functions of the tutoring system. With a third class, each teacher will use the assessment system, but not the tutorial, and with a fourth class, each teacher will use only the tutoring program. Using a pre-post design, student achievement will be compared across the four groups. Because it is not possible to have a randomly selected group in this test, statistical analyses will be used to determine preexisting similarities and differences between the groups. The end result of this project will be a fully developed intelligent tutor and assessment system for high school chemistry with evidence of the potential of this system to improve high school students' chemistry achievement.