A flowchart-based intelligent tutoring system model to improve students' problem-solving skills / Danial Hooshyar

• Main Author: Danial, Hooshyar
• Format: Thesis
• Published: 2016
• Subjects: QA76 Computer software
• Online Access: http://studentsrepo.um.edu.my/6244/6/danial.pdf; http://studentsrepo.um.edu.my/6244/

Many students fail to succeed in programming courses or face difficulties. Lack of problem-solving skills is one of the most important factors contributing to this challenge. Several researchers believe that forming accurate mental models may yield improvement in novice programmers’ problem-solving skills and should thus be a key goal of any introductory programming course. The flowchart has always been deemed ideal in forming accurate mental models of imperative programming concepts. Another concern is the lack of assistance when students encounter problems, which may lead to demotivation. In order to address this concern, one-to-one tutoring provided by an Intelligent Tutoring System (ITS) is known to be effective. Although numerous ITSs have been developed for the programming field, none are designed to enhance problem-solving skills of novice programmers by focusing less on language and syntax and more on solution designing activities in the shape of flowchart development. Hence, the goal is to address the aforementioned gaps in this thesis by developing and evaluating a novel Flowchart-based Intelligent Tutoring System model (FITS) to produce improvement in students’ problem-solving abilities and help them learn basic and imperative computer programing concepts. The decision-making process in FITS is managed by a Bayesian network to handle uncertainty based on the probability theory. Additionally, an online formative assessment game called Tic-tac-toe Quiz for Single Players (TRIS-Q-SP) is incorporated into FITS to promote student motivation in case timely guidance and interaction are deficient. Unlike other existing ITSs related to computer programming, FITS not only promotes the idea of navigating online learning materials and updating the Bayesian network by applying an online game-based formative assessment, it also offers an adaptive and personalized flowchart development environment. The aim of iv FITS is to improve problem-solving ability besides suggest learning goals along with appropriate reading sequences to students. Therefore, FITS can offer students an accurate mental model of execution, as it visualizes the solution development for a programming problem by converting the given problem statement into a relevant flowchart while actively engaging users in the process. Since a flowchart-based multi-agent system and an online formative assessment game are incorporated into the domain model and student model of the proposed ITS, FITS contributes to two different components of intelligent tutoring systems. FITS also expands and improves on many existing ITSs aimed at teaching programming. At the end of the study, the prototype of FITS was evaluated by university students. According to the results, students who used FITS showed higher scores for the post-test than the pre-test with a learning gain of 60% compared to 36%. A two-tailed paired t-test with a 95% confidence interval was performed against the null hypothesis. The p-value of two-tailed paired t-test of 0.000 was obtained, showing strong evidence against the null hypothesis. Therefore, from the result of this t-test, it can be concluded that the scores in the post-test are significantly higher from the scores in the pre-test and the use of FITS in practice is supported. The students’ opinions about FITS were collected via questionnaires and the results signified that the students really liked FITS, the online game and the personalized flowchart development environment as a learning approach.