An Intelligent Tutoring System for Organic Chemistry

Project: Research project

Description

Overview
Over 70,000 students take organic chemistry each year in the USA, but it continues to be one of the most challenging classes for many students, and is often a roadblock course for entry into the health professions, particularly for minorities. Students often try to learn the subject by rote memorization and other passive techniques, but these are doomed to failure due to the extensive course content. Different kinds of active learning strategies are starting to be implemented in organic chemistry courses, but we suggest that a well-designed adaptive electronic homework system is the best active learning tool that can cover the entire general organic chemistry course material.

Building upon content and first generation electronic homework websites that we developed previously with support of a TUES Type I grant, we will develop and disseminate an adaptive intelligent tutoring system for organic chemistry. Critical components will be a student model that includes both cognitive and affective factors, a Bayesian network to evaluate student's knowledge, and a meta-cognitive coach as a software agent that will monitor the status of students problem solving skills and knowledge. The coach will offer unsolicited advice at just the right moments to motivate and use the most effective learning strategies for that particular student. Our research into the student-centric factors that result in high engagement with electronic learning tools for organic chemistry has identified self-efficacy and self-determination as being the most important. The coach will attempt to identify students with poor self-regulation so that it can offer appropriate advice to promote the active development of organic chemistry problem solving skills. The reputation of organic chemistry is such that many students start the course expecting to fail. The coach will also attempt to identify these students and offer growth mindset training.

The tutoring system will be developed and disseminated along with associated course materials that guide a method of instruction based on problem solving from fundamental organic chemistry principles (e.g., Lewis acid/base theory; retrosynthetic analysis). These materials together with the tutoring system are grounded in constructivist pedagogy, wherein students construct an understanding in a strongly guided setting (but not by inquiry), by applying fundamental principles to explain the properties and reactions of organic molecules.

The Intellectual Merit of this research lies in testing how a student model that incorporates both cognitive and affect factors such as mindset can increase learning in organic chemistry, and in determining the optimal form of meta-cognitive coaching to help motivate and students and facilitate learning, particularly for students with fixed mindset. The research will also help to reveal the factors that contribute to success in a challenging college level science course.

The Broader Impact of this research is the development and dissemination of an electronic learning system that may significantly improve organic chemistry students learning, which may reduce attrition and ultimately increase the flow of a diversity of students through the pre-health pipeline. The research should also contribute to the broader development of intelligent tutoring systems across different kinds of college science disciplines.
StatusFinished
Effective start/end date9/15/158/31/19

Funding

  • National Science Foundation (NSF): $591,958.00

Fingerprint

chemistry
student
coach
homework
learning strategy
electronic learning
learning
electronics
coaching
first generation
self-regulation
self-determination
science
health
reputation
self-efficacy
website
profession
minority
instruction