Teaching Data Structure Using a Programming by Example System: Effects on Students Achievement and Cognitive Load

In an attempt to help alleviate the complexity of programming, the technique called Programming by Example (PBE) has been developed. Although businesses and educational institutions have embraced PBE systems, it has not been demonstrated that these systems can equally facilitate program learning and achievement by programmers in lieu of the traditional programming courses. The goal of this research was to determine the effects of the use of a PBE system with students who had experience in programming and design. This investigation examined the effects of teaching students a Data Structures course using a PBE system and analyzed their achievements and measured cognitive load. Data from this study were collected at Middlesex County College from four sections of Data Structures using C++ Programming. Data were gathered from the participants during the experimental semester by using a student programming self-efficacy survey and four programming achievement tests. The researcher made use of a C++ and Java programming expert panel to help develop and determine the validity of the content of the student self-efficacy survey and the four Data Structures achievement tests. In addition, students were asked on each test to subjectively evaluate their cognitive load pertaining to the material covered by each question. The data were examined for significant interactions among the students\u27 prior programming languages experiences, exposure to using the PBE system, cognitive load, and achievement. However, due to violations of the MANOY A assumptions, only a descriptive analysis for the results was possible. The descriptive analysis indicated only minimal differences in the participants\u27 levels of student achievement and cognitive load using PBE. The results of the MANOY A tests showed that student achievement and cognitive load were not correlated. The researcher made the following observations based upon the descriptive statistics regarding the combined quiz average scores and cognitive load z scores for all four quizzes. When taught using Agent Sheets, those students possessing high prior experience scored higher combined quiz averages and z scores than those students possessing low prior experience. In addition, those students, who were presented with Agent Sheets, possessing high prior experience scored lower combined quiz averages but higher z scores than those students classified with the same experience level and did not have exposure to the PBE course content