One of the greatest achievements of Scratch as an educational tool is the eager willingness of programmers to use existing projects as the starting point for their own projects, a practice known as remixing. Despite the importance of remixing as a foundation of collaborative and communal learning, the practice remains poorly understood. Without a clear picture of how and why Scratch programmers remix a project as a starting point of their own projects, this programming community would remain in the dark about which programming practices encourage and facilitate remixing. The designers of programming environments for blocks lack feedback on how the remixing facility is used in the wild. To gain a deeper insight into remixing, this thesis presents the results of a comprehensive study of this practice in Scratch that investigates the following heretofore unexplored dimensions of remixing: (1) the prevailing modifications that remixes perform on existing projects, (2) the impact of the original project's code quality on the granularity, extent, and development time of the modifications in the remixes, and (3) the propensity of the dominant programming practices in the original project to remain so in the remixes. Our findings can be used to promote those programming practices in the Scratch community that encourage remixing while also improving this practice's effectiveness, thus benefiting the educational and end-user programming communities. / Master of Science / The Scratch programming language has become an intrinsically important tool in introductory CS education. A visual, block-based language, Scratch is web-based, featuring an enormous online programming community, through which projects are eagerly shared. One of the unique learning provisions of Scratch is the ability to easily start a project by modifying someone else’s project, a practice referred to as remixing. Despite the central role that remixing plays in enabling the communal and collaborative learning styles in the Scratch community, the practice of remixing remains inadequately understood. This knowledge gap leaves the Scratch community in the dark about which programming practices encourage and facilitate remixing, as well as deprives Scratch environment designers from actionable feedback on how the remixing facility is used in the wild. To address this problem, this thesis reports on the results of an exploratory study of remixing in Scratch that investigates three heretofore unexplored dimensions of this practice. First, we study the general remixing trends in terms of how remixes modify the original projects. Second, we infer the impact of a project’s code quality on the modifications in its remixes and the development time. Finally, we investigate whether programmers adopt the techniques and practices of the remixed projects. Computing educators can apply our findings to enhance the educational effectiveness of Scratch by encouraging the practice and magnitude of remixing.

Much of software engineering research and practice is concerned with improving software quality. While enormous prior efforts have focused on improving the quality of programs, this dissertation instead provides the means to educate the next generation of programmers who care deeply about software quality. If they embrace the culture of quality, these programmers would be positioned to drastically improve the quality of the software ecosystem. This dissertation describes novel methodologies, techniques, and tools for introducing novice programmers to software quality and its systematic improvement. This research builds on the success of Scratch, a popular novice-oriented block-based programming language, to support the learning of code quality and its improvement. This dissertation improves the understanding of quality problems of novice programmers, creates analysis and quality improvement technologies, and develops instructional approaches for teaching quality improvement. The contributions of this dissertation are as follows. (1) We identify twelve code smells endemic to Scratch, show their prevalence in a large representative codebase, and demonstrate how they hinder project reuse and communal learning. (2) We introduce four new refactorings for Scratch, develop an infrastructure to support them in the Scratch programming environment, and evaluate their effectiveness for the target audience. (3) We study the impact of introducing code quality concepts alongside the fundamentals of programming with and without automated refactoring support. Our findings confirm that it is not only feasible but also advantageous to promote the culture of quality from the ground up. The contributions of this dissertation can benefit both novice programmers and introductory computing educators. / Doctor of Philosophy / Software remains one of the most defect-prone artifacts across all engineering disciplines. Much of software engineering research and practice is concerned with improving software quality. While enormous prior efforts have focused on improving the quality of programs, this dissertation instead provides the means to educate the next generation of programmers who care deeply about software quality. If they embrace the culture of quality, these programmers would be positioned to drastically improve the quality of the software ecosystem, akin to professionals in traditional engineering disciplines. This dissertation describes novel methodologies, techniques, and tools for introducing novice programmers to software quality and its systematic improvement. This research builds on the success of Scratch, a popular visual programming language for teaching introductory students, to support the learning of code quality and its improvement. This dissertation improves the understanding of quality problems of novice programmers, creates analysis and quality improvement technologies, and develops instructional approaches for teaching quality improvement. This dissertation contributes (1) a large-scale study of recurring quality problems in Scratch projects and how these problems hinder communal learning, (2) four new refactorings, quality improving behavior-preserving program transformations, as well as their implementation and evaluation, (3) a study of the impact of introducing code quality concepts alongside the fundamentals of programming with and without automated refactoring support. Our findings confirm that it is not only feasible but also advantageous to promote the culture of quality from the ground up. The contributions of this dissertation can benefit both novice programmers and introductory computing educators.

Computing courses struggle to retain introductory students, especially as learner demographics have expanded to include more diverse majors, backgrounds, and career interests. Motivational contexts for these courses must extend beyond short-term interest to empower students and connect to learners' long-term goals, while maintaining a scaffolded experience. To solve ongoing problems such as student retention, methods should be explored that can engage and motivate students. I propose Data Science as an introductory context that can appeal to a wide range of learners. To test this hypothesis, my work uses two educational theories — the MUSIC Model of Academic Motivation and Situated Learning Theory — to evaluate different components of a student's learning experience for their contribution to the student's motivation. I analyze existing contexts that are used in introductory computing courses, such as game design and media computation, and their limitations in regard to educational theories. I also review how Data Science has been used as a context, and its associated affordances and barriers. Next, I describe two research projects that make it simple to integrate Data Science into introductory classes. The first project, RealTimeWeb, was a prototypical exploration of how real-time web APIs could be scaffolded into introductory projects and problems. RealTimeWeb evolved into the CORGIS Project, an extensible framework populated by a diverse collection of freely available "Pedagogical Datasets" designed specifically for novices. These datasets are available in easy-to-use libraries for multiple languages, various file formats, and also through accessible web-based tools. While developing these datasets, I identified and systematized a number of design issues, opportunities, and concepts involved in the preparation of Pedagogical Datasets. With the completed technology, I staged a number of interventions to evaluate Data Science as an introductory context and to better understand the relationship between student motivation and course outcomes. I present findings that show evidence for the potential of a Data Science context to motivate learners. While I found evidence that the course content naturally has a stronger influence on course outcomes, the course context is a valuable component of the course's learning experience. / Ph. D. / Introductory computing courses struggle to keep students. This has become worse as students with more diverse majors take introductory courses. In prior research, introducing fun and interesting material into courses improved student engagement. This material provides a compelling context for the students, beyond the primary material. But instead of only relying on fun material, courses should also rely on material that is useful. This means connecting to students’ long term career goals and empowering learners. Crucial to this is not making the material too difficult for the diverse audience. To keep more students, we need to explore new methods need of teaching computing. I propose data science as a computing context that can appeal to a wide range of learners. This work tests this hypothesis using theories of academic motivation and learning theory. The components of a learning experience contribute to students’ motivation. I analyze how the components of other existing contexts can motivate students. These existing contexts include material like game design or media manipulation. I also analyze how good data science is as a context. Next, I describe two projects that make it simple to use data science in introductory classes. The first project was RealTimeWeb. This system made it easy to use real-time web APIs in introductory problems. RealTimeWeb evolved into the CORGIS Project. This is a diverse collection of free “Pedagogical Datasets” designed for novices. These datasets are suitable for many kinds of introductory computing courses. While developing this collection, I identified many design issues involved in pedagogical datasets. I also made tools that made it easy to manage and update the data. I used both projects in real introductory computing courses. First, I evaluated the projects’ suitability for students. I also evaluated data science as a learning experience. Finally, I also studied the relationship between student motivation and course outcomes. These outcomes include students interest in learning more computing and their retention rate. I present evidence for the potential of a data science context to motivate learners. But, the primary material has a stronger relationship with course outcomes than the data science context. In other words, students are more interested in continuing computing if they like computing, not if they like data science. Still, the results show that data science is an effective learning experience.