Facilitating Transformative Learning Through A Course On Introductory Programming

Abstract

Introductory programming courses are part of the first-year curriculum in all the newlineundergraduate engineering programs in India. In most tier 2 and tier 3 institutions, students newlinelearn programs instead of learning programming. The system hence has an integrity problem newlinethat needs to be addressed and without disturbing the running system. We adopted the action newlineresearch methodology to address the situation. We implemented an additional lab protocol to newlineaddress the problem. We classified student submissions as repeat programs (copied looking at newlinereference programs), remember programs (reproduced after referring to a written program), newlineand write programs (programs originally written). We defined integrity as a match between newlinethe student declared level of the program and the actual program level. We developed tools to newlineidentify integrity violations, and students were counseled on integrity when there were newlineviolations. We found that there was a significant reduction in the number of repeat newlinesubmissions because of the additional protocol. However, there was no significant change in newlinethe remember and write submissions. newlineTo improve the number of students who can write programs, we improved the instruction newline newlinedesign to include metacognitive instruction. We identified seven performance levels: copy- newlinepaste, copy, remember without understanding, remember with understanding, write without newline newlinemodularization, and write with modularization and identified metacognitive strategies to help newlinestudents move up from their current level to the next level. We compared the performance newlineand effort of the students who underwent metacognitive instruction with the students who did newlinenot undergo metacognitive instruction. We found statistically significant improvement in newlineperformance and effort, but impact size was small. We attribute the low impact to the fact newlinethat overall learning culture has an overbearing influence on learning, and the impact of the newlineefforts of an individual instructor is small. A small number of students who genuinely newlineparticipated in metacognitive instruction showed significantly improved performance and newlinespent significantly more effort. newlineInstruction design consists of identifying outcomes, designing assessments, and designing newlineinstruction to achieve the outcomes. The current design has problems since outcomes and newlineassessments are not aligned. Fundamental design issues cannot be addressed by merely newlineimproving instruction. The current instruction method is based on lectures and assignments. newlineInstructors resist any change in instruction method since it is assumed that the change will newline newlinev newline newlineresult in additional workload. To improve the design and estimate instructor effort, we newlinedesigned and implemented a CS0 course with the objective that a student should be able to newlinewrite a modularized program at the level of ACM high school programming competition newlinelevel problems. The study was conducted in Design Based Research (DBR) framework. We newlineadopted extreme apprenticeship as the instruction method, and all the interactions between newlinethe instructor and students happened over text chat. We estimated that students need newlineinstructor interaction time between 4 to 15 hours. We also identified and classified newlineinteractions in cognitive, metacognitive, and affective domains. We observed that a major newlineshift in attitudes needed to happen before the students could progress to writing programs. newlineWe identified the following transformative outcomes; the student should understand the newlinedifference between understand and apply cognitive levels, understand the difference between newlinemastery goals and performance goals, understand the need for completion, and understand newlineknowledge construction. We conducted another cycle of CS0 with these transformative newlinelearning outcomes as focus. At the end of the course, we conducted semi-structured newlineinterviews with 20 students who participated in the course. Analysis of data shows that newlinestudents had transformative learning experiences related to the stated outcomes. The analysis newlineshows that students also reported transformative learning experiences related to acceptance, newlineself-efficacy and self-confidence, motivation, self-awareness, and attitude. newline