Presenter: Jeremy A. Morton
Advisor(s): Dr. Larry Silverberg
Author(s): Jeremy A. Morton
Graduate Program: Mechanical Engineering

Title: Theory and Application of New Methods in On-line Assessment for Engineering

Abstract: Online education has been a major area of development over the past decade. Today, online evaluation and tutorial systems are widely used in introductory level mathematics, physics, and chemistry classes. These systems are used less frequently in higher level classes and in the field of engineering, apparently due to limitations in functionality. Currently, online problems are limited to such forms as true or false, multiple-choice, and the binary (right or wrong) grading of equations. Engineering problems entail the entry of complex equations, and as such binary grading is not suitable. Traditionally the partial credit grading of equations is a crucial aspect of engineering education. Toward this end, a system called Expert TA was developed. Using Expert TA a student enters the full equations associated with each problem. When Expert TA is being used in a tutorial mode, feedback is given that describes any errors contained within a given equation. In homework mode the equations are analyzed to identify errors, partial credit is given that depends on the nature of each error, and feedback for each error is generated and delivered to the student. Expert TA evaluates an equation using two sets of mathematical algorithms. The first set of algorithms symbolically simplifies the equation into a “normalized” form. The second set of algorithms decomposes the equation into elements and generates a topological mapping for each element’s attributes. Using the deconstructed equation, errors are identified and partial-credit grading is performed on the basis of the types of errors found. Expert TA, because it emulates engineering instruction, represents an important step in the development of online systems for engineering education.