Making University Education more like Middle School Computer Club: Facilitating the Flow of Inspiration Alexander Repenning University of Colorado Boulder Department of Computer Science Boulder, CO 80303 (303) 492-1349, 001 [email protected] Ashok Basawapatna University of Colorado Boulder Department of Computer Science Boulder, CO 80303 (720) 838-5838, 001 [email protected] Han Koh University of Colorado Boulder Department of Computer Science Boulder, CO 80303 (303) 495-0357, 001 [email protected] ABSTRACT The way programming is currently taught at the University level provides little incentive and tends to discourage student peer-to-peer interaction. These practices effectively stifle any notion of a ‘learning community’ developing among students enrolled in university level programming classes. This approach to programming education stands in stark contrast to the ‘middle school computer club’ approach; As part of 10 years of research projects aiming to teach programming to middle school children, it is observed that middle school students in computer clubs freely share programming ideas, code, and often query one another and provide solutions to the various programming problems encountered. To enable these interactions at the university level, a novel online infrastructure has been developed over the past 6 years through use in the Educational Game Design Class at the University of Colorado Boulder. The culmination of the submission system, entitled the Scalable Game Design Arcade (SGDA), seems to foster the flow of ideas among students yielding an effective open classroom approach to programming education. Categories and Subject Descriptors K.3.2 Computer and Information Science Education General Terms Algorithms, Design, Human Factors, Keywords University Programming Education, Middle School Programming Education, Scalable Game Design, Open Classroom, peer-to-peer interaction, flow of inspiration, Computational Thinking. 1. INTRODUCTION The current structure in most computer science classes at the university level resembles the so-called “Sage on the Stage” approach to learning [1]. A single lecturer in the front of the class talks to a group of students who are taking notes. Currently, the emergence of remote and on-demand class viewing obviates the need to physically be in a class wherein a teacher takes this ‘Sage on the Stage’ approach to teaching. This approach is further indicted by a recent study suggesting that students who on-demand remotely view these types of lectures actually retain more information and get a deeper understanding of the material as compared to students who physically attend the class [8]. Thus, the value of physically attending class is decreased when a teacher takes this approach. Moreover, the ‘Sage on the Stage’ approach to teaching makes no attempt to use the inherent characteristics of a student-filled classroom to enable a better learning experience. One could argue that with this teaching approach, a lecture wherein one student attends would be identical to a lecture wherein 50 students attend. Given that this is the case, it begs the question, why even have a physical classroom environment? This line of reasoning is unfortunate since the physical student-filled classroom environment lends itself nicely to teaching strategies that foster peer-to-peer student learning and the creation of a learning community within the classroom. Unfortunately, most undergraduate computer science classes make no attempt to create any kind of learning community. In computer science classes, for example, save for a possible “computer lab”, assignments are completed outside of class with little or no motivation or incentive for peer-to-peer interaction [2]. Furthermore, collaborations, sharing of ideas, and looking at fellow students’ code is actually frowned upon and often considered cheating; for example, it is not uncommon for teachers to run automatic checks on assignments to ensure everyone’s code is strictly their own. This has the unfortunate side effect of inhibiting the proliferation of ideas among students drastically reducing the opportunity for students to learn from one another. Since students are actively discouraged from looking at fellow classmates’ work, they are not pushed to do more based on what their peers have accomplished on a given assignment [3]. This has a negative effect not only on individual student achievement, but also, reduces the level of work produced by the class as a whole Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. WCCCE’09, May 1–2, 2009, Burnaby, BC, Canada. Copyright 2009 ACM …$5.00.over the duration of a given course [4]. Some classes try to allow for peer-to-peer interactions through group projects. However, in our experience and as documented by others, group projects often do not yield true group collaborations as students divide the work up and only work together to reassemble the project when everybody’s individual part is complete [7]. Thus, most group projects only have limited peer-to-peer interaction and learning. The above issues motivate the following question: what changes can be made such that students effectively learn from and inspire one another in undergraduate computer science classes? Surprisingly enough, an interesting solution approach to this problem presents itself in the seemingly chaotic structure of middle school computer clubs. For the last 10 years, as part of several NSF funded projects, middle school students were taught the fundamentals of programming using a rapid game prototyping environment called AgentSheets [5]. In our most recent project called Scalable Game Design1,