Contribution: A Class–Lab–Field (CLF) framework for structural health monitoring (SHM) education centers on ArduinoNode, an open, ultralow-cost (~U.S. ${\$}50$ /node) wireless Internet of Things (IoT) platform students build, program, and deploy, narrowing the analysis–field gap from sensing to interpretation. Background: SHM courses foreground structural models and algorithms over instrument working principles and operational procedures; many Arduino labs lack coordinated outdoor deployment. Inspectable, affordable kits for teaching instrument internals remain scarce. Intended outcomes: Operational literacy along the SHM pipeline: accelerometer fundamentals, instrument limits, and links among acquisition, synchronization, logging, and defensible interpretation of noisy field data. Application design: CLF sequences classroom node preparation, lab work on a structural model, and field multisystem monitoring for compressed programs. Hands-on sensing comes first; SHM theory depth scales with contact hours. Findings: CLF: APESS2025 ( $N=56$ ): 93.3% satisfaction, 100% recommend; ~60% report sensor/SHM/IoT gains, 40% WSN gains, and 60% rate hands-on emphasis much or significantly above typical theory-heavy courses. ArduinoNode: Footbridge data (ArduinoNode, smartphones, commercial SHM) compared with a finite-element (FE) model yield 5.1% and 7.2% errors on the first two modes under dual-link, sub-10-ms-class synchronization.
Contribution: Despite their logistical advantages, multiple-choice questions (MCQ) and multiple-true–false (MTF) formats are often underused or viewed skeptically in science, technology, engineering, and mathematics (STEM) education due to concerns about superficiality, a lack of depth, and vulnerability to guessing. This study presents redesigned MCQ and MTF formats that aim to improve objectivity, fairness, and efficiency while preserving conceptual rigor. Background: Open-ended questions are traditionally favored in STEM due to their ability to assess reasoning and partial understanding. However, they pose challenges for consistent and efficient grading, especially in large courses. Meanwhile, the literature shows that well-constructed objective items can match open-ended questions in diagnostic power, yet most existing MCQs fail to meet quality standards, and MTFs remain underutilized. Intended outcomes: The intervention seeks to: 1) ensure objective, deterministic grading; 2) reduce the impact of guessing; 3) support partial credit where appropriate; and 4) offer an efficient solution without compromising the depth of assessment. Application design: We propose a ten-option MCQ structure designed to reduce random success rates while enabling richer distractor design, and a sum-based MTF variant where students report the sum of weights assigned to true statements. The combination of both formats enables automatic grading, discourages guessing, and allows nuanced performance measurement. These were applied to two exams in an undergraduate Electrical Circuits II course over a semester. Findings: Results from 52 students indicate that the proposed formats produce balanced precision and effective discrimination between items. Partial credit allowed for better scoring without inflating the results. A parallel evaluation using state-of-the-art large-language models revealed low performance on the proposed questions, suggesting that such structured formats cannot yet be reliably replaced by AI-based grading of open-ended STEM responses. These findings suggest that thoughtfully redesigned objective questions can serve as efficient, fair, and pedagogically rigorous alternatives to open-ended formats in theoretical or foundational STEM courses where conceptual precision is paramount, and should be complemented by project-based or laboratory assessments.
Contribution: This study presents PlagiAct-V3, an empirically explored and stabilized instrument to assess science, technology, engineering, and mathematics (STEM) students’ attitudes toward plagiarism in the era of generative AI. Background: The rapid rise of tools like ChatGPT has reshaped academic dishonesty. Existing instruments on plagiarism and generative AI (GenAI) lack a robust underlying theoretical basis, clear multidimensional structures, and comprehensive psychometric validation, especially in STEM and European contexts. PlagiAct addresses this gap by integrating four complementary theories, adding GenAI-specific items, and exploring a multidimensional structure with 725 STEM students. Research Questions: Which factors influence STEM students’ attitudes toward plagiarism in digital contexts? What is the factorial structure of plagiarism attitudes? Methodology: PlagiAct-V2 was developed through a literature review, expert content validation, and pretesting, and was administered to 725 engineering students from five Spanish universities. Internal consistency was examined using Cronbach’s alpha and McDonald’s omega, while exploratory factor analysis was used to examine the structure of the attitude scales, yielding eight empirically derived dimensions. These analyses led to the final PlagiAct-V3 instrument. Main limitations include convenience sampling, potential selection bias, and the need for periodic revalidation as GenAI practices evolve. Findings: Results provide evidence of satisfactory internal consistency and exploratory structural validity. The findings highlight the multidimensional nature of plagiarism attitudes in the digital age and offer an empirically grounded basis for designing targeted interventions to promote academic integrity.
Contributions: This article proposes a new knowledge tracing (KT) model called heterogeneous information network embedding with metapath KT (MPHINE-KT) to model the knowledge state of learners and predict the accuracy of learners’ future academic performance. To mine the transfer relationships between different knowledge points in sequences, the learning experience of learners is designed. The effects of individuality and commonality between learners’ learning experience and abilities through heterogeneous information networks are analyzed, learners’ individual attributes are introduced, and the problem of data sparsity is alleviated. The design improves the prediction effect of KT models on learners’ future performance. Background: Online education platforms use KT models to analyze learners’ online study. These models assess and predict learning effects, providing multidimensional feedback and guidance. They fail to adequately extract the features of the transfer relationships between different knowledge points for learners. Moreover, they lack discrimination regarding the differences in the characteristics of individual attributes of learners, and does not fully integrate the multiattribute features of learners, resulting in the sparsity of input data. This restricts the KT model’s ability to predict learners’ performance in answering future exercises. Intended Outcomes: The MPHINE-KT model analyzes the historical interaction sequence data of students and exercise problems, fully explores the relationships among learners, learning experience, and learning abilities, and more accurately predicts whether learners can answer new exercise problems correctly in the future, thus providing support for key educational aspects such as personalized education. Application Design: The MPHINE-KT model first designs a learner’s learning experience recognition module based on a learner behavior graph for capturing learners’ complex representations in the learning process, and then constructs a learner heterogeneous information network. Finally, as an additional input, the learner’s feature vector is put into the gated recurrent unit (GRU) temporal prediction model, so as to improve the accuracy of predicting learners’ future learning performance. Findings: Experiments are conducted on several real-world datasets, and the experimental results prove that the proposed model has a better performance compared to existing popular KT models.
Contribution: This study investigates how the accuracy of artificial intelligence (AI)-generated explanations affects electrical engineering students' solution correctness, confidence, inquiry behavior, and reasons for accepting or rejecting ChatGPT's answers when solving a real-world probability problem grounded in Bayes' theorem and susceptible to base-rate neglect. Unlike prior controlled or guided research, it captures authentic, unstructured student-AI interactions in genuine learning contexts. Background: Although students are generally aware that large language models can produce errors, little is known about their ability to recognize such errors or the extent to which ChatGPT's feedback shapes their confidence during problem-solving in real-world contexts. Research Questions: This study explores: 1) how engineering students solve a real-life problem with ChatGPT shaping their accuracy; 2) how using ChatGPT affects students' confidence in their own answers; and 3) what types of questions students ask ChatGPT and why they accept or reject its responses. Methodology: Using a within-subjects design, N = 107 students solved a real-world task, first unaided and then with ChatGPT. We analyzed ChatGPT's accuracy, students' solution accuracy, written explanations, self-reported confidence, interaction logs, and the reasons students accepted or rejected ChatGPT's answers. Findings: Unaided accuracy was 14%. With ChatGPT, accuracy rose to 62% when its guidance was correct but dropped to 6% when it was incorrect. Few students used verification-oriented prompts, and many accepted plausible but incorrect answers without checking. Exposure to erroneous outputs increased confidence in wrong solutions, revealing a low level of critical engagement and highlighting the importance of integrating AI literacy with probability instruction.
$Contribution$ : This study extends the concept of integrative thinking (InT) from the areas of systems engineering and management to educational robotics. The study defines and classifies InT competencies that high school and undergraduate students employ while learning to construct, program, and operate robots in technology-rich settings. $Background$ : Modern engineers need to apply InT to analyze, design, and operate complex systems across knowledge domains and relationships. Educational robotics provides opportunities for developing InT competencies but approaches to fostering and assessing these competencies have not been sufficiently developed. Research. $Questions$ : What integrative thinking competencies are observed in students involved in experiential learning with robots? What characteristics of learning settings support students’ use of integrative thinking? $Methodology$ : A multiple case study approach was employed with four groups of high school and undergraduate students (N = 198). Data were collected through observations, student reports, logbooks, and interviews, and analyzed using grounded theory methods. While the limited sample constrains generalizability, the study provides exploratory evidence. $Findings$ : The analysis revealed four InT competencies in robotics activities: interconnectedness, functionality, externalities, and alternatives. These competencies were observed across both educational levels and aligned with established frameworks in management and systems engineering. Learning settings involving mechatronic projects and system integration tasks were found to effectively elicit InT. The findings suggest that educational robotics can scaffold integrative thinking development from secondary to higher education, preparing students for the interdisciplinary challenges of modern engineering.
Contribution: This article presents an innovative intelligent system designed for online DC motor experiments. Leveraging the high-precision DC motor control hardware, the system integrates virtual load simulation with neural network-based intelligent evaluation. This integration not only simulates the real-world loads accurately, but also enables real-time assessment of learning outcomes. Background: Conventional DC motor experiments are typically carried out on fixed platforms in laboratories and evaluated through paper-based reports. This approach has several drawbacks, such as spatiotemporal constraints, limited load types, and the lack of process-based assessment. Intended Outcomes: To address these issues, an online DC motor experiment system supporting remote operation has been developed. This system provides real-time feedback to help users understand DC motor operations under various loads. Moreover, the system identifies operational weaknesses and offers specific suggestions for improvement. Application Design: The hardware implementation of the system employs an STM32 controller running a PI control algorithm, which can simulate the mechanical properties of various loads precisely. This is complemented by a web-based platform developed with ASP.NET, which enables remote operation and real-time data visualization. Furthermore, an intelligent evaluation module leverages neural network models to perform automated assessments and provide feedback on the experimental process. Findings: The system enhances the learning experience by combining an intuitive interface with versatile load simulations under standardized procedures. Furthermore, its intelligent assessment generates automated, data-rich reports, offering key support and making it a practical tool for advancing outcome-based education.
Contribution: This study explores the potential benefits and challenges of integrating artificial intelligence technologies, specifically large language models (LLMs), into robotics programming education at the undergraduate level. An LLM is embedded within the virtual simulation environment, functioning as an AI assistant to support students and promote scalable, responsive, and cost-effective robotics education. Background: In robotics curricula, the high costs and safety risks associated with physical robots present significant barriers to scalability and accessibility. Meanwhile, limited instructional resources hinder timely feedback and individualized support. Research Questions: This study investigates the following research questions. RQ1: to what extent does the integration of LLMs impact students’ acquisition of robotics programming skills? RQ2: how do students perceive their learning experiences when interacting with LLM-assisted instruction within a virtual simulation environment? Methodology: An LLM is integrated into a custom-developed virtual simulation platform tailored for undergraduates. After the instructional cycle, student performance between the randomized experimental and control groups is analyzed. The 101 TAM questionnaires are collected and evaluated using reliability and validity testing to examine the effectiveness of the simulation platform. Findings: Results indicate that the LLM supports students in bridging conceptual gaps and improving engagement in robotics programming. The observed performance improvement suggests that LLM assistance facilitates learning processes. The survey reflects strong student acceptance of AI-enhanced instruction with positive attitudes. However, further research is required to adapt LLMs to specific course content and learning objectives.
Background: Educators commonly use formative or summative assessments to evaluate learners’ understanding and identify misconceptions. In the context of programming, existing approaches typically do not provide time-efficient diagnosis of specific misconceptions. Adaptive testing can provide an opportunity to overcome these limitations by customizing the item selection process individually for each learner. Contribution: This article presents a theoretical derivation of an approach for diagnosing misconceptions related to control flow. Based on the notion of shadowed misconception, an adaptive test to identify these misconceptions is constructed, and an evaluation of this test in the context of an introductory programming course (N = 97) is presented. Research Questions: (RQ1) How can an adaptive test be designed to diagnose known misconceptions of control flow as isolated as possible? (RQ2) How reliable is an adaptive testing instrument, designed around the isolated detection of misconceptions, at predicting novice learners’ performance? Methodology: A formative assessment was constructed based on a theoretically derived decision-tree framework of typical misconceptions related to control flow. To evaluate the assessment, a mixed methods approach was employed through a study involving novice university-level programmers. Subsequently, the instrument was analyzed in terms of internal consistency and the presence of various misconceptions. Findings: The results indicate that the adaptive nature of the test reduces the number of items necessary for a diagnosis by 33%. The results also demonstrate that learners’ responses to unseen items can be reliably predicted with an average accuracy of approximately 86%. Furthermore, qualitative data confirms that the test is able to detect a variety of known misconceptions.
Traditional assessment methods in basic mechanics courses often prioritize final outcomes over learning processes, leading to passive student engagement and limited development of practical skills. In response to these limitations, a three-phase assessment framework was designed that integrates formative evaluation, gamification, and real-time feedback throughout the teaching cycle. This model comprises three phases: 1)adaptive preassessment (e.g., class participation and quizzes), 2)competency reinforcement (e.g., group simulations and error correction tassks); and 3) comprehensive application (e.g., theory visualization and dual-teacher evaluation). Implemented in a semester-long mechanics course (N = 120), the staged formative performance assessment (SFPA) cohort showed higher participation (+28%) and higher average exam scores (+7.8 points) than the comparison cohort. Qualitative feedback revealed enhanced teamwork abilities (reported by 85% of students) and stronger motivation through gamified elements like ”lucky draws” and ”beauty discovery” activities. However, challenges such as increased teacher workload and subjective grading criteria were identified. This study contributes a scalable framework for engineering education reform, emphasizing the synergy between process-driven assessment and student-centered learning. Its findings provide actionable insights for educators seeking to balance academic rigor with engagement in foundational STEM courses.
The study of power substations is fundamental to the training of electrical engineers, as these facilities are crucial to power system operation, ensuring the reliability and efficiency of energy distribution. However, traditional teaching methods often face limitations in offering practical and immersive experiences. Given this gap, this work presents a Virtual Reality Substation Laboratory (VRSL), developed based on a 69/13.8-kV substation at the Federal University of Cear & aacute;, Fortaleza, Brazil. The VRSL provides a free-to-use educational platform that allows teachers and students to explore the structural and operational aspects of power substations, including equipment operation and fault simulations. Its integration into two substation-related courses over a three-year period engaged more than 120 students by combining virtual reality (VR) with conventional lectures. The triangulation of quantitative and qualitative analyses demonstrates substantial improvements in academic performance, student engagement, and comprehension of complex substation concepts. By bridging theoretical knowledge and practical simulation, the VRSL enhances engineering education and prepares students for real-world challenges in power systems.
Contribution: This article presents a virtual-real combination experiment system for multirotor autonomous aerial vehicle (AAV) assembly and swarm cooperation, featuring a novel multirotor AAV assembly and swarm cooperation virtual simulation experiment platform (MAAVASC-VSEP) with physical experiments. The experiment system enhances student engagement with cutting-edge AAV swarm technology. Background: The teaching of AAV swarm technology faces challenges such as complex multidisciplinary integration, abstract concepts, and limited experimental resources. Existing AAV courses often focus on AAV technology, lacking a comprehensive, interdisciplinary approach to swarm technology. Intended outcomes: The experiment system cultivates students' exploratory thinking and problem-solving skills, helping them understand the fundamental principles of AAV swarm technology and improving multidisciplinary engineering application abilities. Application design: Combining MAAVASC-VSEP with the corresponding physical experiments, the experiment system includes three inquiry-based tasks: multirotor AAV assembly and parameter tuning, AAV swarm link budgeting and networking, and swarm cooperation and obstacle avoidance. Findings: Student performance evaluation and student survey demonstrate that the experiment system effectively improves students' understanding of AAV swarm technology. Students in the experimental group outperformed students in the comparison group through more fault-tolerant exploratory attempts, indicating the system's significant effectiveness in fostering self-directed learning and problem-solving skills.
“Alice in Codeland” is a gamified course designed to introduce Web programming through HTML, CSS, JavaScript, and PHP, incorporating game mechanics inspired by Haro Aso’s “Alice in Borderland.” The course structure leverages gamification principles, integrating elements such as a standard deck of French playing cards linked to programming exercises, team collaboration, and a dynamic scoring system to foster a competitive yet collaborative learning atmosphere. Each card represents a different level of challenge and is associated with a specific Web programming language, enabling students to progressively build their skills based on the card’s difficulty. The Mad Hatter character, inspired by Carrol’s “Alice’s Adventures in Wonderland,” serves as a mentor and guide, presenting challenges and insights to facilitate the development of programming skills. The game incorporates a comprehensive evaluation system that not only assesses the correctness of submitted exercises but also factors in team performance, adherence to deadlines, and the ability to tackle challenges of varying difficulty. This course integrates narrative elements and gamification to engage students in Web development, emphasizing autonomous learning, creativity, and teamwork.
Experiential learning in the context of an analog electronics lab involves immersing students in hands-on circuit design, testing, and troubleshooting—moving beyond traditional theory-centric instruction. Conventional methods often emphasize theoretical concepts at the expense of practical understanding, limiting the student’s ability to grasp the complexities of analog circuits. This article presents an innovative methodology for conducting analog electronics laboratory sessions, aimed at promoting experiential learning and increasing student engagement. The approach integrates real-time simulations, project-based learning, and collaborative problem-solving to bridge the gap between theory and application. Implemented over two consecutive semesters with a cohort of 71 second-year engineering students, the redesigned lab encouraged critical thinking, creativity, and deeper conceptual understanding through hands-on experiments and interactive demonstrations. Student feedback indicates notable improvements in conceptual clarity, motivation, and practical skills. The findings suggest that this innovation-driven and experiential framework can significantly enhance the effectiveness of analog electronics education within engineering curricula.
Contribution: This article describes an effort to quantitatively characterize the "complexity" of engineering programs across the United States using a burgeoning framework called curricular analytics. A new dataset is introduced to facilitate these analyses for researchers and practitioners, tied to an existing data-sharing agreement. Several avenues for future research using this dataset are outlined, such as connecting curricular data with student course-taking data. Background: As curriculum development and evaluation adopt a more data-driven approach to understanding how to best retain and graduate engineers, curricular analytics offers the engineering education community a method for uncovering the complexities of curricula that may be deterring students from the field. Curricular analytics involves representing program requirements as a network to assign a measure of complexity, called "structural complexity," which has been empirically shown to correlate with completion rates. Research question: How does the complexity of engineering programs vary across institutions, disciplines, and time? Methodology: The research question was approached using a quantitative research design. The sampling frame consisted of the multiple-institution database for investigating engineering longitudinal development (MIDFIELD), a data-sharing agreement encompassing 21 institutions (and still expanding). Plans of study were collected from 13 institutions for five disciplines-Mechanical, Electrical, Chemical, Industrial, and Civil-spanning a decade since their most recent record in MIDFIELD ( $n$ = 494). Then, network analysis was applied to calculate the structural complexity of the programs. The results were explored using descriptive statistics, boxplots, and plotting the complexities longitudinally. Findings: Across the years 2012-2022, a range of 307-372 was observed for structural complexity, with a mean of 325 and a median of 323. In the sample, Chemical Engineering was found to be the most structurally complex discipline, followed by Mechanical Engineering. The remaining disciplines were more tightly clustered together. Over time, chemical engineering and civil engineering marginally increased in complexity by 4%, whereas electrical and mechanical engineering decreased by 2% and 0.7%, respectively. Industrial engineering exhibited the most significant decrease of 11%.
Background: The 360 degrees assessment has been widely used in corporate human resources evaluation of employees, focusing on various employers' desired abilities. These tools are not designed to consider a learning environment and agents' assessment capabilities. Contribution: In the proposed approach, applied to higher education, the 360 degrees assessment aims at a holistic range of agents in the teaching and learning process, including peer, self, and professor evaluation. Research Questions: The primary objectives of the proposed method were active participation in the assessment process, student engagement, and the development of problem-solving abilities and critical views. Methodology: The methodology applies multiple feedback practices and uses rubrics and digital information and communications technologies (DICTs) to assist the assessments. The proposed assessment was implemented over three semesters in nine classes of the digital electronics laboratory (DEL), a course in the Electrical Engineering undergraduate curriculum. Findings: A student survey shows a positive perception of the proposed methodology, and descriptive statistical analysis indicates high student engagement and performance, constituting empirical evidence regarding the approach's efficacy.
As automotive systems increasingly incorporate electronics, software, and sensors, they become complex and widely distributed cyber-physical systems. This complexity makes them more susceptible to cyber-attacks. Nevertheless, despite its great need, the cybersecurity of automotive systems needs to be better understood, even by key stakeholders. This article presents an immersive virtual environment (IVE) platform that enhances the understanding of cybersecurity in automotive systems, focusing on ranging sensor attacks. By utilizing virtual reality (VR), the platform provides a hands-on experience for users to explore and comprehend cyber-attack implications. User studies were conducted to evaluate the effectiveness of the platform, revealing statistically significant improvements in participants’ knowledge, engagement, and self-efficacy related to automotive security. The findings underscore the potential of immersive learning tools in automotive security education, with IVE demonstrating a substantial impact on participants’ comprehension and interest in automotive security.
Contribution: Although noncognitive factors play crucial roles in facilitating the conceptual change process, many prior studies have focused primarily on cognitive variables. This study proposes and tests a model to examine the relationships among learners' metacognitive self-regulation (MSR), situational interest, deep engagement, and shallow engagement in the conceptual change process. Background: Misconceptions that students bring into the classroom hinder the learning of basic science and engineering concepts. Such misconceptions are prevalent across multiple domains and, if left unaddressed, can impede understanding of more advanced concepts. Understanding the role of noncognitive factors in conceptual change is, therefore, essential for advancing STEM education. Research Questions: 1) To what extent does situational interest affect students' conceptual change in electric circuits? 2) To what extent does MSR affect students' conceptual change in electric circuits? 3) How comparable are the effects of situational interest and MSR on students' conceptual change in electric circuits? 4) Are the effects of situational interest and MSR on conceptual change mediated by learning strategies? Methodology: The study analyzed data from 164 undergraduate engineering students enrolled in an electric circuit analysis course. A pretest-posttest design with surveys and a computer-based instructional activity was used, and relationships among constructs were examined using structural equation modeling (SEM). Findings: Results showed that MSR and situational interest had significant total effects on conceptual change. Deep engagement partially mediated the effect of MSR and fully mediated the effect of situational interest, while shallow engagement was not significant. These findings extend the cognitive reconstruction of knowledge model (CRKM) to engineering education and suggest that fostering both situational interest and MSR, in conjunction with supporting deep engagement, can help students overcome misconceptions.
Contribution: Prior studies have emphasized the importance of examining nonlinear control systems in engineering courses, where implementing a proportional-integral-derivative (PID) controller is a fundamental approach to understanding system stabilization and dynamic response. This study demonstrates the effectiveness of a ball and beam system as an open-source educational tool that enables students to engage in real-time PID tuning and system optimization through practical experimentation. Background: PID controllers are fundamental to automatic control courses. While the traditional class provides mathematical foundations, experimental learning through direct interaction with a physical system enables deeper understanding. With its unstable dynamics, the ball and beam system is particularly well-suited to illustrate feedback control principles tangibly and interactively. Intended outcomes: This study seeks to assist students in improving their capacity to optimize PID gains to minimize overshoot and decrease settling time to attain the setpoint. The course’s final laboratory activity was centered around a student competition, which promoted teamwork and reinforced the theoretical and practical concepts of control systems in engineering. Application design: The ball and beam system consists of a servo motor-actuated beam and a time-of-flight (ToF) sensor mounted at the beam’s hinged end for real-time position feedback. Based on the students’ configuration, the Arduino microcontroller processes the sensor data and dynamically adjusts the beam’s tilt to maintain the ball at the desired position. Findings: The competition-based approach fostered a highly motivated learning environment, as students were driven to develop the most effective PID tuning strategy. Learning with hardware enhanced student engagement and provided a multidisciplinary experience.
