This Innovative Practice Full Paper reflects on a pilot experiment aimed to fill a perceived need for training faculty and to sustain systematic and rigorous research in the area of engineering education in India. The pilot was carried out in the form of a year-long online course with thirteen webinars, taught by two faculty members, one from the United States and one from Australia. One of the faculty is a computer engineer who has published research in engineering education outlets in the past decade, and the other faculty member is a psychologist who has extensive experience in conducting human subjects research. The participants in the course are faculty teaching in engineering institutions in India. Some have a PhD while others have a master’s degree in an engineering discipline. The course was designed to cater to an audience that had no prior experience with human subject research and no prior experience in action research. The content covered are the significant components of a research framework, which includes problem definition, research design, methods of data collection and analyses, ethical issues in research and report writing. The long-term goal for this course is to spread awareness of best practices and to increase the overall quality of papers in this area. A survey was administered to the 40 participants in the course, and information was obtained that is deemed helpful to future offerings of the course. This paper summarizes the findings and aims to present them to an international audience with the intent of soliciting inputs from the audience and engage in a discussion that might inform future efforts in this direction in other countries where engineering education might be a nascent discipline.

This Work-in-Progress Innovative Practice paper presents the design, environment, and preliminary results of implementing a virtualized environment for an upper-division course offering in Software Engineering. The Internet-of-Things (IoT) is an emerging paradigm rapidly gaining adoption in everyday consumer computing. Several recent publications and educational platforms recognize the potential for IoT to serve as this rich platform for computing education. However, most of these contributions focus on early undergraduate courses in small hands-on settings using general purpose hobbyist kits based on Arduinos or Raspberry PIs. This paper presents a new and novel IoT educational platform based on virtualization technology. The platform was designed specifically for scalable and complex IoT systems-oriented challenges appropriate for upper-division undergraduate study. This platform was utilized in an online setting at Arizona State University for the first time in Fall 2019. This paper presents insights from this experience, based on student in-class surveys, formal course evaluations, and the instructor’s perspective, along with a roadmap for maturing the platform.

The role of error messages in the context of teaching programming, specifically assembly language programming to students who have limited prior programming experience was investigated. Assemblers and compilers provide feedback to a programmer in the form of error messages, and these error messages influence the programmer's mental model of computing. The current study investigated how an error message affects students' approach to understanding the error and fixing the error. Three error message types were developed - Default, Link and Example, to better understand the effects of error messages. The Default type provides an assembler-centric single line error message, the Link type provides a program-centric detailed error description with a hyperlink for more information, and the Example type provides a program-centric detailed error description with a relevant example. A think aloud programming exercise was conducted to capture the student programmer's knowledge model. Different codes were developed to analyze the data collected as part of think aloud exercise. After transcribing, coding, and analyzing the data, it was found that the Link type of error message helped to fix the error in less time and with fewer steps. Among the three types, the Link type of error message also resulted in a higher ratio of correct to incorrect hypotheses made by the participants, and a correspondingly higher ratio of correct to incorrect steps taken by them to fix the error.

Online education comes in various flavors - skills centered short-duration training, massively open online courses (MOOCs), and more recently, the offering of full online degree programs. In the past 4 years at Arizona State University, the faculty created an online software engineering degree program equivalent to an existing on-campus program, and produced its first graduates in Spring 2017. The challenges in creating this program were significant, but surprisingly the main challenges were not the ones that the faculty anticipated at the outset of the program's development. This paper shares the lessons learned from the development of the online degree program, with an emphasis on the gap between faculty expectations and fears versus the actual issues that needed to be addressed.

This paper investigates the effectiveness of including questions within instructional multimedia content to improve student performance on a related programming assignment. An experiment was conducted where one set of students was provided with an instructional video without any embedded questions and another set of students was provided the same video with embedded questions. The findings of this paper demonstrate that the technique of embedding questions showed promise in improving student performance on a programming assignment.

This paper presents an initial investigation on the effect of non-pre-exposure to an instruction set architecture (ISA). In particular, a specialized ISA based on the Progressive Learning Platform (PLP) is implemented in the computer architecture course. Prior research has demonstrated the benefits of using PLP in the computer engineering curriculum. However, it is possible that the PLP ISA could hinder learning by requiring extra work for students to master it (extraneous load), if they have not had prior exposure to it. To investigate this, the current study implemented a quasi-experimental design with two groups (students knowledgeable with PLP from a previous course, and new users) and a pretest to determine differences in students' familiarity with the common terms in computer engineering, pre-requisite knowledge for a computer architecture course, and course knowledge. Both sets of students implemented the PLP CPU in behavioral Verilog in the computer architecture course. Results of the evaluations revealed significant learning from pretest to posttest by students in both groups on all measures. Moreover, no group differences were seen, indicating that pre exposure to an ISA (specifically PLP ISA) might not be necessary for successful course implementation. This is promising, considering that many students at 4-year colleges in the USA transfer from other institutions, and may have exposure to different instruction set architectures in their prerequisite courses. The sample size for this study is too small to draw a firm conclusion, but these preliminary findings merit further exploration of this topic.

Peer learning principles have been successfully applied to novice programmers. Pedagogies such as Pair Programming, Peer Testing, Peer review of code or tests, or, more generally Peer Instruction, have repeatedly demonstrated their effectiveness in improving both individual performance and retention rates. This paper proposes to supplement the existing literature by investigating how students interact with one another during collaborative programming tasks. More specifically, we are interested in comparing the learning principles used during student-student interactions with those used during student-instructor or student-teaching assistant dialogs.

Agile methods are the fastest rising software lifecycle process methods in software engineering. Educators are converting traditional and project-base courses to agile in response, but this is a daunting task with few structured teaching resources methods available to reduce the burden on the educator. In professional practice, agile methods have been particularly effective in empowering experienced software engineers through its focus on empirical process control and constant feedback loop. These process traits are difficult to simulate in an academic setting, as student developers are inexperienced, synchronous meeting times are few and far between, and obtaining meaningful constant feedback a laborious undertaking. This workshop will present a comprehensive approach to teaching Agile methods that is itself agile, employing a highly iterative, continuous feedback-driven process. Pedagogical and assessment strategies will be shared, and the presenter will facilitate a best practices interactive discussion to draw out lessons learned from workshop participants. Specific agile practices with supporting labs from the popular Scrum and eXtreme Programming (XP) process models will be presented. The workshop will also encourage interaction amongst participants to share best practices and lessons learned. Research directions related to the application of agile principles to teaching and learning will be discussed.

In this paper, we describe a novel platform for Computer Science and Engineering education that helps students to connect software concepts with the underlying hardware. Students often learn about programming languages, computer architecture, assembly languages and compilers in isolation, and fail to see the connections between the hardware and the software. The Progressive Learning Platform (PLP) was designed for students to anchor their conceptual learning about microprocessors and computer architecture, and for them to see the connections between assembly language and trade-offs in architecture. The long-term vision for PLP is to be a free and scalable platform for face-to-face and online education in computing worldwide.

Multimedia applications are fast becoming one of the dominating workloads for modern computer systems. Since these applications normally have large data sets and seem to have little data-reuse, a commonly held belief is that they have poor memory behavior compared to traditional programs and that current cache architectures cannot handle them well. It is therefore important to quantitatively characterize the memory behavior of these applications in order to provide insights for future design and research of memory systems. However, very few results on this topic have been published. We analyze the cache behavior of a group of representative multimedia applications. These programs include a subset of the popular MediaBench suite and several large multimedia programs running on the Linux, Windows 98, and TruUnix operating systems. We performed extensive measurement and trace-driven simulation experiments. We then compared the memory utilization of these programs to that of the SPEC2000 applications. We found that multimedia applications actually have better cache behavior than SPEC2000 programs. The following three factors contribute to the high cache hit rates of multimedia applications: First, most multimedia applications apply block-partitioning algorithms to the input data and work on small blocks of data that easily fit in the cache. Second, within these blocks, there is significant data reuse as well as spatial locality. Third, a large number of references generated by multimedia applications are to their internal data structures, which are relatively small and can easily fit in reasonably sized caches.