Content uploaded by Yunfei Hou
Author content
All content in this area was uploaded by Yunfei Hou on Sep 07, 2022
Content may be subject to copyright.
The Adaptation of Online Project-based Learning in
Computer Engineering Education Settings
Yunfei Hou, Fadi Muheidat, Amir Ghasemkhani, Qingquan Sun,
Haiyan Qiao, Miranda McIntyre and Montgomery Van Wart
California State University San Bernardino, San Bernardino CA 92407, USA
Abstract. While many universities offered some online courses in computer en-
gineering prior to the COVID-19 pandemic, few offered project-based courses in
a fully virtual mode. This has changed with pandemic-related restrictions. When
converting to an online-only mode, active-learning projects and lab components
of computer engineering courses pose additional challenges because they require
face-to-face interaction and specialized equipment. In this paper, we present how
we adapted a year-long capstone course to be fully online at a public university
in California, U.S., and report our findings based on students’ perceptions. Our
major findings are: 1) students’ satisfaction rates of online courses improved over
the one year of online-only learning, and 2) students tended to prefer synchronous
lectures, but the difference in preference between synchronous vs. asynchronous
teaching was not significant.
1 Introduction
The coronavirus pandemic has challenged many aspects of the educational system.
Moving from face-to-face instruction to online learning has produced questions about
the long-held, fundamental ideas about instruction, attendance, evaluation, the role of
technology and human interactions, etc. Online learning offers valuable advantages
compared to face-to-face such as time and space flexibility, long-term access to written
and recorded lectures, opportunities for self-regulated learning, and efficient pro-
grammed learning. However, adapting project-based courses (e.g., computer engineer-
ing design capstone course) to an online learning structure requires careful planning
and teaching presence to support successful learning outcomes such as teamwork skills,
effective communication, self-directed learning of new knowledge, and the ability to
apply engineering design solutions need to be satisfied.
The goal of this paper is to summarize and discuss the lessons learned from teaching
project-based courses that are fully online. The results can help better design computer
engineering courses in remote or hybrid learning modalities after the COVID-19 pan-
demic. We will present the implementation of project-based learning in a one-year-long
capstone course for undergraduate computer engineering students. We will discuss the
pedagogical considerations based on the student outcomes mandated by the Accredita-
tion Board for Engineering and Technology’s Engineering Accreditation Commission
(ABET EAC).
2
The paper consists of two parts: 1) how we transformed the design of project-based
learning courses to be fully online, and 2) an analysis of the students’ perceptions of
online learning. In order to measure students’ perceptions regarding online learning
experiences and learning outcomes, we conducted two surveys in June 2020 and March
2021. The surveys covered both the first and last terms that the university moved to
fully online instruction. This allowed us not only to capture students’ opinions but also
investigate how their opinions and expectations evolved during the pandemic.
2 Related Work
Computer science and engineering curriculums include both theory and hands-on skills.
To accomplish the important latter goal, schools set aside lab components for most
courses. In those courses, the faculty set aside time to assist students in understanding
the concepts through hands-on experience, either by simulation or experimental boards
and systems [1]. Project-based learning relates to learning theories, cooperative learn-
ing, acquiring knowledge, thinking process, and problem-solving skills. It is learner-
centered, makes them curious and enthusiastic about exploring more. Most students are
keen to learn by utilizing real tools rather than just covering theoretical concepts with
pen and paper [2]. Project-based learning can be applied in particular courses or
throughout a curriculum. It may be coupled with traditional teaching or implemented
through individual or small group settings, and may last from a few weeks to a year [1].
Computer science and engineering students also benefit from the hardware experience
to establish a clear picture of the relevant relationship among many hardware compo-
nents. Through hands-on activities, the students can quickly grasp the concepts of com-
puter operations, examine the CPU structure while executing programs, and acquire the
ability to adjust expert skills in their careers [3]. The practical activity can also make
them enjoy learning [2]. Learning-by-doing is a learning paradigm where students tend
to be more engaged by hands-on activities based on physical lab components than sim-
ulation-only activities. [3]. Some [4] went one step further by proposing that we move
to project-based learning and allow the students to design and implement their projects
in alignment with the assignment objectives and goals, rather than the traditional guided
lab assignment. In [4], the authors found that the “create your project” was more enjoy-
able and useful by engaging their personal interest. The ultimate goal is to establish a
good learning environment for students to gain experience in developing projects. In
this paper, we acknowledge the importance of hands-on and project-based courses. Un-
like much of the previous literature, however, we are interested in the students and
faculty perspectives on transforming these assignments into an online format.
3 The Engineering Design Course
At the computer science and engineering program used in this study, the engineering
design capstone course is offered through two consecutive semesters and entitled
“Computer Engineering Design”. These courses include a lecture section where stu-
dents are introduced to the engineering design process, a lab section where students
3
practice their knowledge of computer engineering concepts through empirical lab as-
signments, and a final project where students work in teams for two semesters. The
final projects are defined based on real-world engineering applications which require
careful system architectural design and evaluation, concurrent hardware and software
design, system integration, testing and validation. The final student products are a
working prototype and project documentation. The computer engineering capstone
course targets five different ABET student learning outcomes [5] including: 1) an abil-
ity to apply engineering design to produce solutions that meet specified needs with
consideration of public health, safety and welfare, as well as global, cultural, social,
environmental, and economic factors; 2) an ability to communicate effectively with a
range of audiences; 3) an ability to recognize ethical and professional responsibilities
in engineering situations and make informed judgments, which must consider the im-
pact of engineering solutions in global, economic, environmental, and societal contexts;
4) an ability to function effectively on a team whose members together provide leader-
ship, create a collaborative and inclusive environment, establish goals, plan tasks, and
meet objectives; and 5) an ability to acquire and apply new knowledge as needed, using
appropriate learning strategies. In what follows, we present the implemented strategies
for this year-long computer engineering capstone course to enhance students’ learning
in a virtual environment.
Logistical support: Due to the COVID-19 pandemic, students did not have access
to face-to-face lab instructions and lab equipment. The university of this study was fully
online from March 2020 to August 2021. To alleviate the logistical hurdles of the lab
assignments, we offered no-cost lab supplies to students during the pandemic on an as-
needed basis. First, a list of required equipment for all the lab assignments was provided
to students at the beginning of the semester. Then, they were asked to notify the in-
structors with a list of items they needed to be mailed to them from the department
supply. As a result, all students had access to the necessary equipment for at-home lab
assignments during online synchronous sessions.
Flexible project selection: The scope of projects for the computer engineering cap-
stone course revolves around the computer engineering concepts gained from under-
graduate courses. In the past, the selected software-hardware co-design projects re-
quired some level of face-to-face interactions between team members. This posed a
challenge during the pandemic since the students could not meet in-person due to health
safety issues. To address this challenge, we offered software-only projects (e.g., facial
recognition to match photos in the database) by relaxing the hardware component inte-
gration. The focus was on software modules which can be integrated into hardware
components in different design problems (e.g., facial recognition module for unlocking
a door), see Fig.1. As a result, students had two options for selecting their final projects:
hardware-software or software only.
Facilitating virtual teamwork activities: The lecture and lab sessions of the com-
puter engineering capstone course were held synchronously through Zoom meetings.
Students were given in-class and asynchronous assignments alongside their final pro-
ject to work in teams. The main challenge was to provide a suitable virtual environment
for teams to work and manage their tasks similar to a face-to-face environment. In this
regard, facilitating virtual teamwork activities to enhance the students' learning process
4
is critical. Students were introduced to strategies for effective teamwork through a lec-
ture on team dynamics, structures, ethics, and conflict management at the beginning of
the first semester. Then, team members were asked to define their own ground rules for
their teams. To simulate face-to-face interactions, Zoom breakout rooms were utilized
to put student in private team sessions while instructor support is provided.
Regular instructor-student meetings: In a virtual learning environment, a lack of
face-to-face interaction can make it difficult for team members to receive adequate
feedback from instructors. When left on their own, virtual team members are more
likely to get into ambiguity about their roles, project outcomes, expectations, etc. There-
fore, regular instructor-team meetings are a powerful strategy that enables instructors
to build a relationship with students in a virtual environment, address their needs, set
clear expectations, and help students to achieve their best performance. Specifically,
instructors required regular updates about the progress of each team and set clear ex-
pectations and goals for each team’s project outcomes. Additionally, instructors moni-
tored each student’s progress and provided necessary guidance to students. In sum, stu-
dents had many opportunities to seek clarification or assistance with their projects and
assignments during meetings, as well as virtual office hours, email etc.
Taking Advantage of new online learning platforms: The main challenge with
online learning for computer engineering courses is to provide hands-on experience for
hardware-related assignments. To address this issue, online learning platforms such as
Tinkercad were leveraged to create models of electronic circuits with embedded micro-
controllers (e.g., Arduinos). Students developed their codes for the microcontrollers on
a web-based interface, practiced the correct wiring of different components, and tracked
the real-time results. Subsequently, students were given an option to work and submit
their hardware-related assignments using the online platforms. Moreover, other readily
accessible online design and prototyping tools (e.g., Figma) were introduced to design
UI/UX of the final project prototype.
Fig. 1. Sample Projects. COVID-19 Face mask recognition lock (left). A door lock fixed with a
keypad and face scanner. To ensure that the person accessing the door is "fit" to work, an appa-
ratus connected requires the face to be scanned to see if guests are wearing masks properly. It
also checks the temperature and has a hand sanitizer dispenser. Vitals on the Go (right). A wrist-
band with NodeMCU Esp8266 development board to communicate with an application in the
user’s cell phone. This design tracks and monitors a person’s health indicators such as temperature
and heart beat and notifies the user of any unusual change in the health conditions.
5
Adjusting course contents for online learning: The course content for the com-
puter engineering capstone course was prepared based on the pre-selected projects for
the year. The main theme was about cyber physical system (CPS) applications such as
building a parking lot monitoring system on a college campus with IoT sensors and
surveillance cameras, and building a smart home appliance with sensing, control, and
communication capabilities. In this regard, lectures were mainly focused on detailed
implementation of CPS’ components including control, sensing and communication
sub-systems since students need to have these features in their projects. However, it is
challenging to cover the detailed implementation of the physical components for CPS
applications in a virtual environment which usually requires face-to-face interactions
and specialized equipment. To tackle these challenges, instructors prepared the course
content with more emphasis on the data processing/analysis, UI/UX design, and app
development components (e.g., flutter app development, python GUI, etc.) and relaxed
the requirements related to face-to-face meetings, prototype packaging/casing, and
hardware integration. In addition, asynchronous tutorials were prepared for the im-
portant hardware related topics to cover the detailed design of physical components
(e.g., wireless communication protocol with Node MCU and Raspberry Pi). Finally,
due to time and space flexibility in the virtual environment, résumé and job-interview
workshops were also offered as part of the course by inviting alumni speakers to prepare
the students for the future job-hunting process.
4 Survey Results
Two surveys were created to measure STEM students’ perceptions of online learning.
They were conducted in the Spring 2020 [6] and Spring 2021 terms when the university
was in the first and last mandatory online teaching terms. The university in this study
is the region’s largest public, comprehensive university, with a student population over
20,000. Hispanic minorities make up over half of the student population. 58% of un-
dergraduate students receive financial aid (Pell Grant recipients). The surveys were sent
to the School of Computer Science and Engineering with about 900 major students. We
received a total of 179 and 156 responses in Spring 2020 and Spring 2021, respectively.
In the Spring 2021 sub-sample, we had 86.5% male (13.5% female), 42.9% Hispanic,
and the average student age was 24.8 years (range = 18 to 43). Major results of this
study can be summarized into three categories: Teaching Presence, Course Format and
Online Modality.
4.1 Teaching Presence
Teaching Presence refers to students’ perceptions about the quality of lectures, as well
as directions and feedback received from instructors. To assess student perceptions, we
asked about their satisfaction with online courses in general and with project-based
courses specifically, and then compared the results from the Spring 2020 and Spring
2021 terms.
6
The empirical data indicated that the overall satisfaction rate of online teaching had
improved over the past year during the COVID-19 disruptions (refer to Fig.2). Students
that indicated they were satisfied (including both extremely and somewhat satisfied
categories) with online courses increased from 43.5% in Spring 2020 to 48.1% in
Spring 2021. While the percentage of dissatisfied respondents remained roughly the
same (29.9% in Spring 2021, and 29.5% in Spring 2021), it is worth noting an encour-
aging trend at the two ends of spectrum in Fig. 2: Students that were Extremely Dissat-
isfied dropped by 3.8%, and student that were Extremely Satisfied increased by 10.9%.
This confirmed our observation that both students and faculty have better adapted to
the online learning environment.
In the Spring 2021 survey, we specifically asked students how satisfied they are with
the project-based courses. Among the 156 respondents in the Spring 2021 survey, 72
of them were taking online project-based courses, i.e., the Engineering Design courses
and Senior Design courses. We compared these 72 students’ perceptions of project-
based courses with their overall online courses in Fig. 3, which indicated slightly less
satisfaction on project-based course. In the case of project-based courses, 51.4% of the
students indicated that they were satisfied, whereas 54.2% students were satisfied with
online courses in general. There were also more percentages of respondents that were
dissatisfied with project-based courses (i.e., 25%) when compared to online courses in
general (i.e., 23.6%). The results confirmed our observation that students were not keen
on online project-based courses, due to the difficulties on team collaboration and the
lack of equipment and facilities.
4.2 Course Format
In the online learning environment, we were interested in understanding students’ pref-
erences between synchronous and asynchronous lectures, differences in the composi-
tion of lecture time, and grading policies. The past year provides a special opportunity
to experiment with different course formats which can help guide online teaching after
the pandemic.
When considering the preference regarding synchronous lectures (i.e., live virtual
class meetings) vs. asynchronous lectures (i.e., pre-recorded), the results were slightly
favored asynchronous. In Spring 2020, 47.5% of respondents preferred asynchronous
Fig. 2. Question: Overall, how satisfied are
you with your online courses?
Fig. 3. Question: Overall, how satisfied are
you with your online project-based courses?
7
lecture; after one year of fully online learning, this number increased to 50.7% in Spring
2021. In comparison, 41.3% of respondents preferred synchronous lectures in Spring
2020, and that rose slightly to 42.3% in Spring 2021. See Figure 4. There was no sig-
nificant correlation by class standing (the logistic regression p≥.231).
In a follow up question, we asked respondents how much of their class time they
prefer to spend on lectures vs. activities (e.g., group discussion, practice exercises).
17.9% of the respondents selected 100% lecture, 31.4% selected 80% lecture, 32.7%
selected 60% lecture, 12.2% selected 40% lecture and 5.8% selected 20% or less lecture
out of all the class time allocation.
A major challenge identified in this study was the grading and assessment policy.
When asked about how online testing reflects their understanding of course material,
41.7% of respondents indicated that they were satisfied with online testing, 25.6% an-
swered neutral and 32.7% indicated unsatisfied. The high percentage of negative re-
sponses, unfortunately, was not surprising. The university implemented a temporary
policy that allowed students to choose between credit/no credit and letter grades on
their transcript. Instructors recognized the importance of grading student performance
while considering COVID-19 impacts, and some deployed a hybrid grading approach
to motivate students. Most of the instructors, however, maintained a traditional grading
policy. We have identified designing fair and flexible grading policies in distance learn-
ing as a future work.
4.3 Online Modality
Online modality refers to issues related to the use of technology tools and online func-
tionalities. We found significant improvement in students’ satisfaction rate regarding
the availability of online lab resources. The qualitative response indicated the rapidly
deployed fully-online classes led to a high percentage of negative responses in Spring
2020 (refer to Fig. 5). In contrast, the percentage of respondents who were satisfied
with the online lab resource doubled to 49.6% by Spring 2021. This improvement in
lab resources seemed to contribute to the higher satisfaction rate in Spring 2021 as
shown in Fig.2. In general, students were satisfied with the technology tools provided
by the university, which includes Blackboard, a learning management system; Zoom,
Fig. 4. Question: Which type of online classes
are you most comfortable with?
Fig. 5. Question: Overall, how satisfied are
you with the availability of resources in your
online labs?
8
Google Cloud and Adobe Creative Cloud. Our results also suggested that students un-
derstood the challenges caused by the COVID-19 crisis and appreciated the depart-
ment’s effort to alleviate these challenges with equipment mailed to their homes.
5 Conclusion
Our study showed cautiously optimistic results in online teaching during the pandemic.
Given that online and hybrid education will likely be a strategic priority after COVID-
19, we hope this study helps faculty better prepare project-based courses in virtual en-
vironments. Specifically, for computer engineering capstone courses, the paper sum-
marizes our efforts on how we have better provided students with hands-on experience
in solving practical problems and prepare them to adopt a data-driven mindset. Addi-
tionally, based on the survey results, we make the following recommendations to im-
prove project-based online teaching:
1. In terms of student preferences, faculty can choose between synchronous and asyn-
chronous teaching modalities. However, students clearly appreciate alternatives so a
synchronous lecture that is subsequently posted is generally preferred above a single
mode. Most preferred is a synchronous lecture and a high-quality, pre-recorded lec-
ture which becomes more realistic as course materials are built up over time. In pro-
ject-based settings, instructors may want to consider less all-class sessions and sub-
stitute shorter individual team briefings.
2. It is important to offer a variety of interactive opportunities with the students—full
class, small group, virtual office hours, blog sites embedded in the course, and so on
—, and to provide feedback in a very timely fashion. Mentor and peer support are
also valuable.
3. Students in project-based courses are just as eager to have reminders and “ticklers”
about assignments and deadlines as students in theory-based classes or components.
However, given the nature of the medium and the flexibility inherent in project-
based learning, instructors may consider providing auxiliary guidance to students on
self-paced and time management strategies.
References
1. McManus, J. W., Costello P. J.: Project based learning in computer science: a student and
research advisor's perspective. Journal of Computing Sciences in Colleges (2019).
2. Kojmane, J., Aboutajeddine, A.: Engineering Junior: A hands-on activity to enhance tech-
nological learning in an engineering dynamics course. In: 2016 International Conference on
Information Technology for Organizations Development, (2016).
3. Lee, S. J., Jung, A., Park J., Yun, M.: Cost-efficient Hands-on Learning Design for Com-
puter Organization Course. In: 15th International Conference on Computer Science & Edu-
cation on Proceedings, (2020).
4. Cassens, M., Reimer,Y.: Engaging CS1 Students With Project Based Learning. In: 2018
IEEE Frontiers in Education Conference, (2018).
5. ABET Engineering Accreditation Commission,https://www.abet.org/accreditation/accredi-
tation-criteria/, last accessed 2021/5/29.
6. Hou, Y., Muheidat, F., Usher, T., Prado, W., Guo, X., and Van Wart, M.: Evaluation of the
COVID-19 shock on STEM laboratory courses. In: IEEE EDUCON 2021, (2021)
