Dissertation/Thesis Abstract

Pen-Based Interfaces for Intelligent Statics Tutoring Systems
by Lindsey, Levi Scott, M.S., University of California, Riverside, 2013, 222; 1547829
Abstract (Summary)

Here we present two intelligent tutoring systems for statics, the sub-discipline of engineering mechanics concerned with the analysis of mechanical systems in equilibrium under the action of forces. These systems are pen-based: one runs on Windows tablet PCs and the other on LivescribeTM smartpens with specially-designed paper worksheets. It is common for novice students to attempt to solve problems without understanding the fundamental concepts involved. For example, they may attempt to solve a new problem by adapting the solution to an example problem. This approach can lead to errors as novices often categorize problems on the basis of surface similarity rather than the structural—i.e., conceptual—similarity. Our new instructional model guides students in explicitly examining the structural elements that govern the solution. For example, before the student draws forces on a free-body diagram, the system requires the student to explicitly identify all interaction points, points at which other objects apply forces to the body. The student must then identify what kind of interaction occurs at each interaction point before representing them by force arrows. The system critiques the student's work for each of these steps and provides appropriate tutorial feedback. This instructional design has a number of benefits. It helps students to identify the structural elements that guide the solution process, which is important for problem-solving transfer. It also enables the system to accurately diagnose student errors. Because each step in the reasoning is explicitly recorded, the system can unambiguously determine the cause of an error and provide focused tutorial feedback. Also, the use of natural pen-based interfaces unburdens the student from extraneous cognitive load inherent in more traditional interfaces. We conducted two studies to evaluate these systems. The first included 43 students enrolled in Statics (ME 10) at UCR, while the second included 10 students enrolled in Introduction to Mechanical Engineering (ME 2). The results suggest that students find the systems to be useful for learning statics. However, the tablet-based system is more effective than the smartpen-based one, with the former leading to large and statistically significant learning gains in the second study.

Indexing (document details)
Advisor: Stahovich, Thomas
Commitee: Vahid, Frank, Zordan, Victor
School: University of California, Riverside
Department: Computer Science
School Location: United States -- California
Source: MAI 52/03M(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Educational technology, Computer science
Keywords: Intelligent tutoring systems, Pen-based computing, User interfaces
Publication Number: 1547829
ISBN: 9781303507175
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest