Effectiveness and Design of PBL-Based Project Approach for Non-Major University Computing Courses

Abstract

The purpose of this study is to analyze the learning outcomes and educational implications of introducing a new Problem-Based Learning project approach in a computing course for non-major students. The Problem-Based Learning project approach, designed by the research team, provides a teaching and learning strategy in which students collaborate to analyze and extend code within real-world problem contexts while systematically exchanging feedback. The study was conducted as a case study involving two groups of students: a control group comprising 30 students following traditional instructional methods and an experimental group consisting of 31 students employing the proposed approach. Educational outcomes were evaluated through both quantitative and qualitative analyses. The findings demonstrate that the Problem-Based Learning project approach facilitated a significantly more effective learning experience, fostering key skills such as communication, collaboration, and problem-solving. Furthermore, substantial improvements were observed in instructional effectiveness and the structure of practical activities. These findings suggest that the Problem-Based Learning project approach mirrors real-world practices, equipping students with essential computational thinking skills and a deeper understanding of foundational topics such as Python, data structures, and algorithms. This study provides valuable insights for the development of instructional strategies in computing courses for non-major students, emphasizing the importance of creating systematic learning environments that bridge theoretical knowledge and practical application through real-world problem-solving activities.

1. Introduction

With the advancement of digital transformation and artificial intelligence (AI) technologies, the importance of Computational Thinking (CT) has expanded [1]. CT refers to the ability to analyze problems and solve them through algorithmic and procedural structures. Beyond simple programming, it has become an essential mindset for systematically addressing complex problems across various fields and is widely regarded as a core competency required in the AI-driven era. Python, a versatile tool widely used across disciplines, plays a role similar to that of Excel in the past. Its applications in data analysis, automation, and artificial intelligence demonstrate that computing skills are no longer exclusive to developers. In the future, non-major students will also need problem-solving skills grounded in CT and AI-driven thinking to meet the demands of various fields [2].

In response to these needs, many universities have integrated computing courses into their general education curricula. These courses are designed not only to teach programming skills but also to help students understand the importance of problem-solving processes using concepts such as data structures and algorithms. Rather than training students to become developers, these courses aim to equip them with fundamental CT and collaboration skills necessary to work effectively with developers and solve problems together. This approach enables students from diverse academic backgrounds to logically address real-world problems that they may encounter in their personal and professional lives [3,4].

Previous studies have primarily focused on the theoretical approaches and effectiveness of Problem-Based Learning (PBL) and Project-Based Learning (PJBL) in computing education for non-major learners [5,6,7,8,9]. These studies often emphasize providing non-majors with introductory experiences in computing and artificial intelligence, enabling basic engagement with the subject matter. However, there is a noticeable lack of research on how computing education can be tailored for non-majors, particularly in foundational areas such as programming, data structures, and algorithms, and how these can be integrated with learners’ respective disciplines to foster interdisciplinary convergence.

This research gap poses challenges for the design of learning environments that equip non-major learners with practical problem-solving skills. To address this, our study aims to design computing education programs for non-majors that focus on solving real-world problems and to evaluate their effectiveness. This study introduces a PBL-based Project-Based Learning (PJBL) approach, which extends the traditional PBL framework by incorporating long-term project development and implementation. PBL serves as the foundation, focusing on analyzing and resolving smaller, well-defined problems through collaborative inquiry and decision-making. Building on this foundation, the PJBL model requires students to design, plan, and execute complex, real-world projects that demand the application of problem-solving strategies in authentic contexts [10].

The PBL-based PJBL model employed in this study goes beyond individual problem-solving tasks. Students not only analyze and extend existing code written by their peers to apply it to new problems but also engage in feedback exchanges with their peers on the extended code. This process helps students understand how data structures and algorithms are applied to real-world problem-solving and provides them with essential collaborative coding and communication skills.

Building on previous research, this study compares the outcomes of a PBL-based PJBL project evaluation with those of traditional written and practical exams. Through this comparison, the study seeks to derive insights into the learning outcomes and educational implications of computing courses for non-major students. Furthermore, it aims to propose instructional strategies that enable non-major students to collaborate and solve problems effectively using computational thinking.

The research questions addressed in this study are as follows:

• How should the instructional design for computing courses based on PBL-based PJBL learning be structured?
• What kinds of learning experiences do students gain from expanding and providing feedback on their peers’ code in PBL-based PJBL computing courses?
• What are the effects of PBL-based PJBL computing courses on students’ sense of achievement, problem-solving abilities, and collaboration skills?

2. Theoretical Background: Problem-Based Learning and Computing Education

2.1. Problem-Based Learning (PBL)

Problem-Based Learning (PBL) was developed by Barrows and Tamblyn at McMaster University’s medical school in Canada during the 1960s [10]. It was designed to address the limitations of traditional lecture-based education, which merely transmits knowledge without fostering meaningful application in real-world contexts. PBL emphasizes student-led exploration of problem situations, enhancing students’ problem-solving abilities [11,12].

PBL also promotes collaboration and communication among learners, encouraging active participation and fostering creativity, critical thinking, and problem-solving skills. Rogers emphasized that instructors should act as facilitators rather than mere knowledge transmitters [13]. In this role, instructors support students in independently exploring problems and seeking solutions while providing guidance when necessary [4]. Thus, in PBL-based project courses, instructors must create an environment where students can collaborate effectively with peers.

The learning process of PBL, as outlined by Torp and Sage [14], can be summarized as follows: 1. Learners assume roles relevant to the presented problem situation. 2. They encounter unstructured problems. 3. Learners identify what they already know and determine what additional knowledge is required to solve the problem. 4. They explore further and define the issues that need to be addressed. 5. Learners evaluate possible solutions and select the most appropriate one.

The essence of PBL lies in discovering and defining unstructured problems and extending the relevant solutions. This study employs the PBL process outlined by Torp and Sage [14] to design computing classes, enabling students to collaborate with peers and deepen their understanding of code analysis and algorithm development.

In addition, several studies have demonstrated the benefits of applying PBL in computing education. For instance, Lee [6] developed a PBL-based computer instruction model focused on broadening learners’ thinking processes. Similarly, Han [9] explored the effectiveness of PBL in programming education, emphasizing how PBL enables learners to structure and solve unstructured problems independently, highlighting its educational value. Furthermore, Lee [15] conducted research on the practical application of PBL in introductory computer engineering courses, confirming that PBL-supported instruction positively impacts the development of creativity and problem-solving skills required for computing professionals. This body of research collectively underscores the educational significance of PBL in fostering creativity, collaboration, and problem-solving capabilities, which are critical competencies for learners in computing disciplines and align with the demands of interdisciplinary and real-world problem-solving contexts.

2.2. The Importance of Project-Based Learning and Collaborative Learning

Project-based learning (PJBL) has been studied by various scholars since Kilpatrick’s work in 1918 [16,17,18]. Project-based learning is recognized as an effective method for fostering collaboration and enhancing creative problem-solving skills [19]. It encourages self-directed learning by engaging students in solving problems connected to real-life contexts [20]. According to Oh [2], the three core attributes of project-based learning are self-directedness, integration with life contexts, and the creation of tangible outputs.

A series of studies applying PJBL in computing education have demonstrated its educational effectiveness compared to traditional lecture-based methods [21]. Choi and Heo [7] and Kim [22] reported that while PJBL requires more effort, it yields substantial educational benefits, such as deeper engagement and long-term knowledge retention. Herrington and Oliver [8] emphasized that PJBL focuses on solving real-world problems and fostering creativity and problem-solving skills, rather than merely acquiring fragmented knowledge. Similarly, You [23] advocated for project-based software learning, arguing that rote instruction in programming syntax is inefficient. Instead, You recommends providing learners with opportunities to identify exploratory problems, collaborate on solution strategies, and creatively solve challenges.

PJBL aligns with the broader goals of computing education, particularly in fostering computational thinking and problem-solving skills [24]. Its emphasis on teamwork, communication, and the practical application of knowledge prepares students for professional and research environments, equipping them with the ability to collaboratively address complex problems. Building on this foundation, this study integrates PBL with Project-Based Learning to design instructional strategies. Students are encouraged not only to solve individual problems but also to extend their peers’ programs and apply them to new challenges. Through this process, students exchange feedback, strengthening their collaboration skills and accumulating teamwork experience [25]. This approach not only aligns with the attributes highlighted in prior studies but also expands on them by providing opportunities for peer learning and interdisciplinary applications. The ultimate goal is to equip students with the collaborative and computational thinking skills necessary to address real-world problems effectively.

2.3. Computational Thinking and Computing Education in General Education

Computational Thinking (CT), a concept introduced by Wing [1], involves analyzing problems and developing algorithms to solve them using computer science principles. CT includes processes such as abstraction, decomposition, algorithm design, and pattern recognition. These skills serve as essential tools for solving problems across various disciplines and real-life situations.

Python, akin to Excel in the past, has become a versatile tool that is widely used in data analysis, artificial intelligence, and numerous other fields. This trend underscores the growing importance of CT as a fundamental problem-solving framework for non-majors. The goal of computing education in general education is not to train students to become developers but to equip them with CT and collaborative skills necessary for working with developers and solving problems effectively.

Research on CT in computing education has provided valuable insights into its application for both majors and non-majors. Park and Choi [26] discussed the dual objectives of university-level software education: training software specialists and providing general computing education for non-majors. Na [27] conducted a study comparing the learning outcomes between humanities and science students, reporting minimal differences in performance and concluding that software education is universally beneficial regardless of academic background. Interestingly, Na’s research also highlighted that students, irrespective of their major, recognized the importance of software education for future career and personal development. You [28] explored the use of different computational tools for learners from varying academic backgrounds, such as humanities and arts students using Entry, while science students employed Entry alongside educational robots like Playbots. This approach reflects the importance of tailoring educational tools to the specific needs and aptitudes of different learner groups. Park and Choi [26] also emphasized the pivotal role of instructor expertise in teaching programming languages and the need for interdisciplinary education to enhance creativity and problem-solving skills for non-major learners.

The PBL-based project learning model applied in this study aims to help students understand the principles of data structures and algorithms and apply them to solve problems. Through activities such as analyzing, extending, and providing feedback on peers’ code, students strengthen their collaborative skills and foster a self-directed learning attitude. These experiences help students apply CT in real-world situations and develop problem-solving abilities through collaboration.

Building on these findings, this study also underscores the importance of instructors serving as facilitators in a PBL or PJBL setting. By leveraging instructor expertise and combining PBL with project-based approaches, this study aims to promote creative problem-solving and collaboration among non-major students. This integrated approach has the potential to enhance both practical problem-solving skills and interdisciplinary applications of CT in general education.

2.4. Connection Between Previous Studies and This Study

Previous studies on PBL and project-based learning have reported their effectiveness in enhancing problem-solving abilities and learning motivation [3,4]. They also emphasize the importance role of collaboration and feedback in fostering creative thinking and communication skills.

Building on these studies, this research employs an instructional design that integrates PBL with project-based learning to analyze the learning outcomes in computing education for non-major students. The goal is to enable students to acquire CT and collaboration skills, providing non-majors with the ability to apply these skills across various problem contexts.

3. Methodology

This study adopts a case study approach with a hybrid instructional design integrating elements of Problem-Based Learning (PBL) and Project-Based Learning (PJBL). The PBL component focuses on guiding students through the exploration and analysis of unstructured coding problems, fostering critical thinking and inquiry-based learning. Once students define the problem and explore possible solutions, the instructional design transitions into PJBL, where the emphasis shifts to project execution. In this phase, students collaboratively extend and refine code, applying their problem-solving insights to create functional project deliverables. The hybrid nature of this approach ensures that students acquire theoretical problem-solving skills (PBL) in tandem with practical project-based execution skills (PJBL) [19,24]. This combination aligns with the goals of this study, whichs aim to provide non-major students with a comprehensive understanding of computational thinking (CT) and its application in real-world contexts.

3.1. Participants and PBL-Based Project Course Design

This case study was conducted with students enrolled in an introductory computing course. The course was part of the liberal arts curriculum for non-majors at a national university in Seoul, Korea, during the first semester of 2024. The course aimed to teach students the fundamental concepts of computational thinking (CT), algorithms, and programming, enabling non-major students to develop problem-solving skills and apply them to real-world challenges. To analyze the impact of the Problem-Based Learning (PBL) project approach, the learning outcomes of students from the first semester of 2024 were compared with those of students from the second semester of 2023.

At the beginning of both semesters, students were surveyed about their prior experience with Python programming, data structures, and algorithms. The majority of students demonstrated basic exposure to Python programming, typically acquired through high school informatics classes or introductory university courses. However, their experience was limited to fundamental concepts including syntax, basic operations, and simple problem-solving. They had not been exposed to advanced concepts, such as object-oriented programming or structured approaches to data structures and algorithms.

The study was conducted with two groups of students within the scope of the case study: a control group of 30 students and an experimental group of 31 students. Both groups received identical theoretical instruction, covering foundational Python programming, data structures, and algorithms. The key difference lay in their final assessments. The control group followed a traditional assessment model, culminating in an algorithm coding exam that evaluated the application of concepts learned throughout the semester.

In contrast, the experimental group participated in the proposed PBL project approach, which emphasized collaborative problem-solving and code analysis. Students in the experimental group were tasked with identifying algorithmic problems, proposing their own problems, and extending the solutions provided by peers. Finally, they engaged in peer feedback, systematically reviewing and critiquing the extended code submitted by their classmates. This iterative process emphasized collaboration and enabled students to practice critical coding and communication skills.

This iterative process of problem creation, extension, and feedback enabled students to deepen their understanding of data structures and algorithms while practicing collaborative coding and communication skills. This design facilitated a direct comparison of the educational outcomes of the proposed approach with traditional methods, particularly in enhancing practical problem-solving abilities and fostering teamwork. As shown in

Table 1. Comparison of course and evaluation methods.

Category	Control Group	Experimental Group
Theory	Basic concepts of data structures and algorithms	Same
Practice	Algorithm implementation and problem-solving exercises	Same
Final Evaluation	Programming proficiency test	Problem-based project evaluation
Number of Students	30	31

This table compares the differences in evaluation methods, grouping, and student distribution for both semesters.

, the two groups followed identical weekly programming assignments, ensuring consistency in learning activities. However, the final evaluation methods diverged significantly.

Table 1 highlights the primary distinctions between the control and experimental groups, particularly in the format of the final evaluation. While both groups received identical theoretical and practical instruction, the control group completed a programming proficiency test, whereas the experimental group engaged in a problem-based project evaluation. Additionally, the table includes the number of students in each group, emphasizing the balanced distribution of participants. The iterative process of problem creation, extension, and peer feedback in the experimental group enabled students to deepen their understanding of data structures and algorithms while practicing collaborative coding and communication skills. This comparison framework facilitated an analysis of how these differing evaluation methods impacted learning outcomes, problem-solving abilities, and teamwork experiences.

The number of participants, and the distribution of academic years and departments, were consistent across the two semesters. This ensured comparability in evaluating the impact of the instructional methods and final evaluation formats.

3.1.1. Problem-Based Project Design

The problem-based project course design was developed from previous studies, incorporating the principles of Problem-Based Learning (PBL) and collaborative learning [14]. The course was structured into four key phases: (1) problem creation and implementation, (2) problem expansion and application, (3) peer feedback, and (4) presentation evaluation. A summary of the phases is shown Figure 1.

• Phase 1: Problem Creation and Implementation

In the first phase, students designed and solved self-created programming problems using algorithms. This phase aimed to strengthen students’ ability to define, structure, and implement problems using CT principles. Each student created a problem and implemented its solution using Python, ensuring that the code was modular and reusable. The problems, along with their corresponding code and explanations, were shared on an online platform for peer access and collaboration.

• Phase 2: Problem Expansion and Application

In the second phase, students selected a peer’s problem to extend or modify it. This process required students to analyze existing code and apply creative solutions to develop new functionalities. The goal was to deepen their understanding of algorithms and foster the ability to work with and build upon others’ code. Students were encouraged to create standalone programs by adding new features to existing solutions, simulating collaborative development environments.

• Phase 3: Peer Feedback

During the third phase, students reviewed the extended problems and provided detailed feedback using a structured rubric. Although the students were non-majors with limited technical expertise, they were guided to evaluate their peers’ work based on the following key aspects:

• Code Accuracy: Assess whether the code meets the problem requirements and is free of bugs or errors.
• Code Efficiency: Evaluate the time and space complexity of the algorithm, ensuring no unnecessary computations and considering more efficient approaches.
• Readability and Style: Check if the code is well organized, uses meaningful variable and function names, and adheres to consistent coding standards.
• Documentation and Comments: Ensure that appropriate comments are included to aid in understanding, clarifying the algorithm’s intent and logic.
• Modularity and Reusability: Verify whether the code is modular, employing well-structured functions or classes to enhance maintainability and reusability.
• Creativity and Innovation: Identify creative or unique approaches in solving the problem, highlighting non-standard solutions where applicable.
• Security and Exception Handling: Examine the code for input validation, proper exception handling, and the absence of security vulnerabilities.
• Testing and Validation: Assess whether the provided test cases comprehensively validate the code and whether it functions correctly across varied inputs and scenarios.

Recognizing that non-major students might lack the proficiency to assess all aspects thoroughly, the rubric was designed to offer clear guidance and a foundation for meaningful feedback. Students made earnest efforts to apply these criteria, focusing on the areas they felt most confident in while gradually developing a more holistic understanding of the coding principles.

This structured rubric not only ensured consistency in feedback but also encouraged students to critically engage with their peers’ work, deepening their understanding of coding principles and fostering collaborative problem-solving. By analyzing the strengths and weaknesses of each solution and offering constructive suggestions for improvement, students honed their critical thinking and interpersonal skills. This phase was pivotal in bridging individual work with collaborative enhancement, simulating professional software development practices.

• Phase 4: Presentation Evaluation

In the final phase, students submitted video presentations summarizing their problem creation, problem extension, and feedback processes. The presentations, lasting 3–5 min, offered students an opportunity to reflect on their learning and showcase their achievements. This phase aimed to enhance students’ communication skills and foster a sense of accomplishment. Additionally, the presentations served as a platform for peer feedback, promoting knowledge sharing and collaborative learning.

3.1.2. Instructor’s Role and Facilitation Strategies

The instructor played the role of a facilitator, guiding students to actively engage in self-directed learning. The instructor’s primary responsibility was to ensure students remained focused during the problem exploration process, providing feedback and guidance as needed. The instructor also encouraged students to collaborate effectively during the peer feedback and presentation phases, emphasizing the importance of code quality and creative problem-solving.

The facilitation strategies implemented in this course were consistent with the principles of PBL, as described by Torp and Sage [14]. The instructor’s role was to promote collaboration and self-directed learning, ensuring students developed both technical and interpersonal skills necessary for real-world problem-solving.

3.2. Data Collection and Analysis Methods

To assess the effectiveness of the problem-based project design, both quantitative and qualitative data were collected and analyzed. The project-based evaluation results from the first semester of 2024 were compared to the programming test results from the second semester of 2023, which primarily assessed individual coding and algorithmic skills. The focus of the analysis was to determine the impact of the evaluation method on students’ computational thinking, algorithmic understanding, and collaboration skills.

The project course was evaluated based on students’ contributions throughout four phases: problem creation, problem extension, peer feedback, and presentation. Key evaluation criteria included creativity, originality, code accuracy, efficiency, and the quality of feedback provided by students. These metrics were compared with the results of the programming test conducted in 2023.

To further assess the overall effectiveness and satisfaction of the course, an official university-administered survey was utilized. The survey included multiple-choice and open-ended questions, addressing topics such as course preparation adequacy, teaching effectiveness, and the appropriateness of assignments and feedback. The survey comprised both multiple-choice and open-ended questions. While the multiple-choice responses were analyzed using Likert scales for quantitative analysis, the open-ended responses were analyzed using thematic coding to identify recurring themes and provide qualitative insights. Although the open-ended responses were not included in the formal evaluation scores, they were reviewed to assess the reliability and consistency of student feedback. A summary of the survey items is presented in

Table 2. Course evaluation survey items.

Evaluation Questions
The lecture was generally satisfactory.
The course content and preparation were adequate.
The teaching methods were effective.
The lecture plan was helpful for course selection and learning.
The feedback on assignments and tests was satisfactory.
The practice activities were well organized and facilitated learning.
Students were encouraged to actively participate, and feedback was appropriate.

This table summarizes the survey questions used for evaluating the course.

.

The course evaluation survey consisted of two categories of questions: general questions applicable to all courses and specific questions tailored to practice-based sessions. The general questions assessed the overall satisfaction of the students with the course, the adequacy of the content and preparation, the effectiveness of the teaching methods, the relevance of the lecture plan, and the appropriateness of feedback on assignments and tests. In addition, specific questions for practice-based sessions evaluated whether hands-on activities were well organized, enhanced student understanding, and encouraged active participation with constructive feedback.

By combining multiple data sources and evaluation methods, this study ensures a robust and multifaceted assessment of the educational impact of the problem-based project design. Quantitative measures, such as test scores and Likert-scale responses, were complemented by qualitative insights from open-ended survey questions, offering a holistic understanding of the course’s strengths and areas for improvement.

4. Results

A comparative analysis of course evaluation results from the second semester of 2023 and the first semester of 2024 within the scope of this case study reveals consistent improvements across all evaluation metrics. These findings highlight the positive impact of implementing Project-Based Learning (PJBL) in non-major computing education.

As shown in

Table 3. Comparison of evaluation scores: control group vs. experimental group.

Evaluation Item	Mean (C)	Std Dev (C)	Mean (E)	Std Dev (E)	Mean Difference	t-Value	p-Value
1. Overall Satisfaction	4.57	0.75	4.74	0.78	0.17	−2.147	0.039
2. Preparation and Quality	4.70	0.71	4.87	0.73	0.17	−2.126	0.041
3. Teaching Method Effectiveness	4.43	0.79	4.77	0.86	0.34	−4.187	0.0001
4. Feedback Appropriateness	4.47	0.76	4.77	0.81	0.30	−3.724	0.001
5. Organization and Alignment	4.37	0.80	4.74	0.85	0.37	−4.581	0.00003
6. Participation and Feedback	4.51	0.76	4.78	0.81	0.27	−3.167	0.004

Note: (C) = Control Group; (E) = Experimental Group. Significant p-values are highlighted in bold.

, the mean scores for all evaluation items improved significantly between the control group (C) and the experimental group (E), with t-values confirming the statistical significance of these changes (p < 0.05). Table 3 presents a comparison of evaluation scores between the control group (2023) and the experimental group (2024). The groups followed identical instructional content but differed in their assessment methods. Significant p-values are highlighted in bold. The category “Overall Satisfaction” increased from 4.57 in 2023 to 4.74 in 2024, with a mean difference of 0.17 (t = −2.147, p = 0.039). Notably, when compared to the university-wide averages of 4.53 in 2023 and 4.54 in 2024, this course consistently outperformed institutional benchmarks. “Teaching Method Effectiveness” demonstrated one of the most notable improvements, with the mean score rising from 4.43 in 2023 to 4.77 in 2024, reflecting a mean difference of 0.34 (t = −4.187, p = 0.0001). This result underscores the capacity of PJBL to enhance students’ perceptions of instructional quality. Similarly, the category “Systematic Organization and Alignment” exhibited the most significant enhancement, increasing from 4.37 to 4.74 (mean difference = 0.37, t = −4.581, p = 0.00003), highlighting PJBL’s role in structuring a coherent learning experience. Other areas such as “Feedback Appropriateness” and “Participation and Feedback” also saw substantial gains, with mean differences of 0.30 (t = −3.724, p = 0.001) and 0.27 (t = −3.167, p = 0.004), respectively. These metrics reflect the emphasis on clear, constructive feedback and active engagement in PJBL environments.

Figure 2 visually represents the evaluation score improvements between the control group (C) and the experimental group (E). The control group (C) followed traditional instructional methods, while the experimental group (E) engaged in a PBL-based project approach. Notably, the standard deviations ranged from 0.7 to 0.8 for most items, indicating a high degree of response consistency. However, slight increases in standard deviation for certain metrics in 2024 suggest that PJBL may evoke diverse reactions due to its innovative and collaborative approach.

In summary, the course not only exceeded the university-wide averages but also showed statistically significant improvements across all evaluation items. These results confirm the positive impact of PJBL on enhancing student engagement, satisfaction, and learning outcomes in non-major computing education.

The results of this study provide insights into how project-based learning influences students’ problem-solving abilities, collaboration experiences, and computational thinking. These findings contribute to developing effective computing education models for non-major students, emphasizing the significance of collaborative problem-solving in real-world contexts.

5. Discussion

The comparative analysis of course evaluation metrics between the control group (2023) and the experimental group (2024) in this case study shows consistent improvements across all items. These findings strongly indicate that implementing Project-Based Learning (PJBL) positively impacted student learning experiences. For instance, significant gains were observed in teaching method effectiveness (mean increase from 4.43 to 4.77), systematic practice organization (mean increase from 4.37 to 4.74), and overall satisfaction (mean increase from 4.57 to 4.74). These results align closely with prior research highlighting the benefits of PJBL in fostering systematic and engaging learning environments [21,29,30,31].

This consistent improvement underscores the efficacy of PJBL in not only increasing student satisfaction but also fostering a more structured and coherent educational experience [5]. By emphasizing long-term projects and collaborative problem-solving, the PJBL approach in this study successfully bridged the gap between theoretical concepts and practical application, particularly for non-major students.

The analysis of open-ended responses highlights both the strengths and challenges of the PJBL approach. Students reported that the integration of coding and algorithm concepts into practical projects greatly enhanced their understanding of computational problem-solving. Comments such as “The combination of practice sessions and projects made the course structure feel more systematic” suggest that the alignment of theoretical knowledge with hands-on projects was a crucial element in enhancing learning outcomes.

Additionally, students appreciated the diverse assessment approach, combining written exams with long-term projects. For instance, one student noted, “The final project helped me understand not just coding basics, but how to create practical outputs that I can actually use in real-world scenarios and potentially apply in my major field”, highlighting the role of PJBL in demonstrating the practical utility of learned skills and fostering confidence in their future applicability.

The formal comparison between the control group (2023) and the experimental group (2024) reveals notable differences in learning outcomes. While both groups received identical theoretical instruction and programming exercises, the experimental group showed significant improvements across all evaluation metrics due to the introduction of PJBL. Key findings include the following:

• Teaching Method Effectiveness: The experimental group’s mean score increased significantly compared to the control group (4.77 vs. 4.43), reflecting enhanced perceptions of instructional quality.
• Systematic Practice Organization: The PJBL model’s structured phases led to notable improvements (mean score of 4.74 for the experimental group vs. 4.37 for the control group).
• Participation and Feedback: The experimental group benefited from collaborative feedback and coding extensions, resulting in higher engagement (4.78 vs. 4.51).

This comparison highlights that the PJBL approach fostered deeper student engagement, improved problem-solving skills, and enhanced collaboration experiences compared to the traditional instructional model.

Despite these positive outcomes, certain challenges were identified. Students expressed concerns about the workload, especially during project extension phases. Comments such as “The final project was too demanding for a general education course” point to the need for adjustments to the course design to alleviate the perceived burden. This finding aligns with Bell and Horowitz [32], who emphasized the importance of providing clear guidelines and structured feedback to manage workload effectively in PBL settings.

Another concern was the variability in the quality of peer feedback. While peer feedback is a key component of PJBL, some students noted that “peer feedback lacks expertise, and guidance from tutors or instructors is necessary”. These challenges suggest that incorporating expert feedback or structured guidance during peer evaluation phases could improve the overall effectiveness of collaborative activities.

Additionally, difficulties in understanding and expanding upon others’ code during the project extension phase were reported. Comments such as “The code used to solve the problem was not suitable for extension” emphasize the need for reference materials and example code to support students in this process. Providing such resources could enhance the scalability and adaptability of PJBL activities in future iterations.

While the standard deviations for most metrics indicated consistent responses, a slight increase in variability was observed for certain items in the first semester of 2024. This suggests that students had diverse experiences with the PJBL approach. Prior studies, such as Chen and Yang [29] and Zhang and Ma [33], highlight that variability in PBL outcomes often arises from differences in task complexity, prior knowledge, and the alignment of instructional design with student needs.

To address this, future course iterations could integrate diagnostic assessments at the beginning of the semester to gauge students’ initial knowledge and readiness for PJBL activities. This approach could inform tailored scaffolding strategies, reducing disparities among students. Additionally, incorporating structured milestones and supplementary resources throughout the course, as recommended by Fini et al. [30], could ensure more equitable and supportive learning environments.

The findings of this case study suggest several improvements for future iterations of the course:

• Enhanced Feedback Mechanisms: Incorporating structured expert feedback alongside peer evaluations could address the variability in feedback quality and provide students with more actionable insights.
• Workload Adjustments: Reducing the intensity of project phases or distributing the workload more evenly throughout the semester could alleviate the burden on students, making PJBL more accessible.
• Support for Code Reuse and Extension: Providing example codes and reference materials could help students adapt and expand upon existing solutions more effectively.
• Diagnostic Assessments: Implementing early assessments to identify varying levels of prior knowledge could inform more personalized support strategies.

This case study contributes to the growing body of literature on PJBL by demonstrating its potential effectiveness in non-major computing education. By emphasizing collaborative and structured project phases, this study highlights how PJBL can enhance creative problem-solving, computational thinking, and collaboration skills among students who do not major in computing.

However, this study is not without limitations. The absence of teacher and administrator perspectives limits the comprehensiveness of the findings. Additionally, while the analysis focused on short-term outcomes, longitudinal studies are needed to assess the long-term impact of PJBL on students’ problem-solving and teamwork abilities. Future research should incorporate these perspectives and explore strategies for further optimizing PJBL implementation.

In summary, this case study provides strong evidence supporting the use of Project-Based Learning (PJBL) as an effective instructional method for non-major computing education. By addressing the challenges and incorporating targeted improvements, future courses can further enhance the learning experience, ensuring that PJBL remains a valuable tool for bridging theoretical and practical knowledge in diverse educational settings.

6. Conclusions

This study evaluated the educational outcomes and implications of a newly developed PBL-based project approach applied in a computing liberal arts course for non-major students. The instructional design integrated problem-based and project-based learning strategies to enhance students’ computational thinking, collaboration, and problem-solving abilities. By engaging students in problem creation, code extension, and peer feedback in realistic scenarios, the course aimed to foster both individual skills and collaborative capabilities.

A comparative analysis was conducted between the control group (students assessed using traditional methods) and the experimental group (students assessed using the PBL-based project approach). The results showed that the PBL-based project approach significantly outperformed the traditional instructional method in several key areas, including teaching effectiveness, systematic practice, and student satisfaction. The course also exceeded university-wide averages, further supporting the value of the proposed approach.

The findings highlight the potential of PBL-based project designs to bridge the gap between theoretical knowledge and practical application, particularly for non-major students. Students demonstrated improved problem-solving skills, enhanced collaboration experiences, and a deeper understanding of core concepts such as Python programming, data structures, and algorithms. These outcomes affirm the effectiveness of integrating PBL with project-based activities to create meaningful, engaging, and systematic learning environments.

Building on this case study, future research could refine the instructional model by addressing identified challenges, such as workload management and variability in peer feedback quality. Additionally, longitudinal studies are needed to assess the long-term impact of PBL-based projects on students’ professional skills and interdisciplinary problem-solving abilities. By continuing to develop and evaluate such models, computing education for non-majors can better align with real-world demands and prepare students for diverse professional and academic contexts.