Informatics in Education

Teaching programming is a complex process requiring learning to develop different skills. To minimize the challenges faced in the classroom, instructors have been adopting active methodologies in teaching computer programming. This article presents a Systematic Mapping Study (SMS) to identify and categorize the types of methodologies that instructors have adopted for teaching programming. We evaluated 3,850 papers published from 2000 to 2022. The results provide an overview and comprehensive view of active learning methodologies employed in teaching programming, technologies, programming languages, and the metrics used to observe student learning in this context. In the results, we identified thirty-seven different ALMs adopted by instructors. We realized that seventeen publications describe teaching approaches that combine more than one ALM, and the most reported methodologies in the studies are Flipped Classroom and Gamification-Based Learning. In addition, we are proposing an educational and collaborative tool called CollabProg, which summarizes the primary active learning methodologies identified in this SMS. CollabProg will assist instructors in selecting appropriate ALMs that align with their pedagogical requirements and teaching programming context.

The paper discusses an alternative method of assessing the difficulty of pupils’ programming tasks to determine their age appropriateness. Building a program takes the form of its successive iterations. Thus, it is possible to monitor the number of times such a program was built by the solver. The variance of the number of program builds can be considered as a criterion of the difficulty of the task. We seek to verify whether this variance is the greatest in the age group for which the task is most suitable. We created several series of programming tasks and offered them to 87000 pupils from 4th to 13th grade. For each task, we compared the optimal age group determined by the variance of the number of program builds method with the group determined by the correct answer ratio method. A strong correlation was observed in traditional microworlds Karel the Robot and Turtle. A moderate correlation was achieved in the new microworld Movie.

In education, we have noticed a significant gap between the ability of students to program in an educational visual programming environment and the ability to write code in a professional programming environment. The aim of our research was to verify the methodology of transition from visual programming of mobile applications in MIT App Inventor 2 to textual programming in the Android Studio using the Java Bridge tool as a mediator of knowledge transfer. We have examined the extent, to which students will be able to independently program own mobile applications after completing the transition from visual to textual programming using the mediator. To evaluate the performance of students, we have analysed qualitative data from teaching during 1 school year and determined the degree of achievement of educational goals according to Bloom’s taxonomy. The results suggest that students in the secondary education can acquire advanced skills in programming mobile applications in a professional programming environment, when they have knowledge of visual programming in an educational programming environment, and a suitable mediator is used to transfer such knowledge into a new context.

Students' performances in introductory programming courses show large variation across students. There may be many reasons for these variations, such as methods of teaching, teacher competence in the subject, students' coding backgrounds and abilities, students' self-discipline, the teaching environment, and the resources available to students, all of which can affect student performance and outcomes. Our observations in teaching programming courses (at Al-Imam Muhammad Ibn Saud Islamic University in Riyadh) are that many students (up to 50% per course) drop out. There is a strong belief by many instructors that such a high dropout rate is due, at least in part, to students underestimating the effort needed to finish this course and not following instructions as recommended. This paper reviews the factors that affect student performance in an introductory programming course (CS1) and aims to discover correlations between various assessment methods, students' participation and their final performance measured. It analyses mark distributions across various assessment methods to identify which assessment method best predicts final exam marks and overall marks, and gives recommendations for assessment in introductory programming courses.

The paper analyses the problems in selecting and integrating tools for delivering basic programming knowledge at the university level. Discussion and analysis of teaching the programming disciplines, the main principles of study programme design, requirements for teaching tools, methods and corresponding languages is presented, based on literature overview and author`s experience. A pressure from labor market, students and other sources to emphasize practical skills over deeper, long-term programming concepts is described. A model of teaching introductory programming disciplines at a higher logical level, using C#, is presented as a summary of the accomplished analysis, and also taking into account the recommendations of the ACM (Association for Computing Machinery) association for typical teaching programs. Also, design principles for building introductory programming courses, aligned with such teaching approach, are presented. This model has already been trialed at Vytautas Magnus University.