Using Machine Learning to Assess Critical Thinking Skills Related to Health and Nutrition

Critical thinking CT skills, such as ability to analyze information, form logical conclusions, and problem solve, are essential for students and measure for educational outcomes because they demonstrate that students can consider facts, recognize gaps in evidence, and evaluate choices when making decisions Gambrill, 2006. With this in mind, 133s at the University of Rhode Island developed and implemented an online critical thinking development and assessment program in introductory nutrition and agricultural science courses. While this has shown promise, implementation is limited by its reliance on human-scorers. To address this our proposal will test the viability of automated assessment. Utilizing data gathered from the current online program we will test a range of natural language processing NPL approaches and compare the results to those of the human-scorers.