A quantitative study of GUI versus text-based object-oriented instruction

Introductory object-oriented programming is considered difficult to teach and a number of different methodologies have been proposed to address this difficulty. One method, which is reported in this paper, involves the use of graphical user interfaces (GUIpsilas). The research question developed for this study is as follows: Does any statistical difference exist in test scores between a control group using text-based object-oriented programming (OOP) and a test group using GUI-based OOP? To answer this research question, quantitative data was collected through the use of common questions on the final CS2 exam for both groups. Data was collected over a period of 3 class offerings for five multiple-choice questions and one question requiring programming. No statistical difference between the groups was found for either the 5 multiple-choice questions or the open programming question. An analysis of the final CS2 course letter grade showed no significant difference between the test and control groups. Further, students who participated in the first year of the study were tracked through data structures, the course which follows CS2 in the curriculum sequence. A statistical analysis of the final letter grade for found no significant statistical difference between control and test groups.