better solution. Our aim was to provide high qualityassignments to students while minimizing time spent on logistics.Given the limitations of existing tools such as Microsoft Word, we chose to instead develop anew system for the following reasons: • We wanted to manage problems and associated solutions in a library. • We wanted the flexibility to use any of the features of LaTeX. • We wanted to focus time on pedagogical innovation and direct student engagement rather than the details of formatting assessments.Designing assessments requires the instructor to consider the purposes, format, and other detailsof the assignment [1]. Instructors often choose Microsoft Word for creating student learningassessments. Since it is a What-You-See
graded exam, the students were given a laboratory assignment inwhich they interacted with ChatGPT-3.5 to obtain feedback on their MATLAB exam. Qualitativedata on the students’ experiences with the use of ChatGPT as a tool in studying were collectedand analyzed. The results revealed that while students found the capabilities of ChatGPTintriguing, they remained skeptical in the output and reasoning given in regard to their MATLABassignment.1 IntroductionIn November of 2022, OpenAI introduced ChatGPT, a natural language processing model, to theworld. Two months later, it gained 100 million users, making it the fastest growing consumer appin history [1]. The name stems from the model’s dependence on the Generative Pre-trainedTransformer (GPT
able to effectively engage a broader audience.1. IntroductionThe number of jobs in software development is projected to increase substantially over the nextdecade [1]; this increased demand will require many new workers to learn how to developsoftware. Traditionally, many universities and colleges have provided computer science degreeprograms that will prepare future workers. However, more scalable approaches like MassiveOpen Online Courses (MOOCs) could be an alternative – a more scalable approach to preparingthe next generation of software developers that might reach a broader audience [2]. Thesecourses can help to address rising demand for computer programming education and expandaccess to educational opportunities [3]. Unfortunately, MOOCs
the virtual labs for the course PHYS 303 offered atOld Dominion University (ODU), the proposed development techniques can be readily extendedto other courses that utilize these common instruments, including courses offered by universitiesand high schools. A preliminary user study conducted with the first lab module in the coursePHYS 303 demonstrated the effectiveness of the virtual lab.1. IntroductionIn the evolving landscape of educational technology, virtual labs have emerged as an importanttool, offering an alternative to traditional laboratory experiences. With technology's continualadvancement and integration in educational settings, virtual labs are increasingly gainingprominence. This trend is particularly evident in the fields of
issues.Introduction and Literature ReviewFemale representation has continuously been an issue within computing, including computergaming. As women are stakeholders in educational software and make up roughly half of thepopulation, it is essential they see themselves being represented accurately and positively.Gender Representation Issues in ComputingWhen digital computers became a practical reality in the 1940s, women were the pioneers inwriting software for the machines. At the time, men regarded writing code as a secondary, lessinteresting task, as the real “glory” lay in hardware design [1]. When the number of coding jobsexploded in the 1950s and 1960s, employers looked for candidates who were logical, meticulous,and good at math. In this respect, gender
were mostly first- or second-year BSE-Mechanical orElectrical Engineering students. There were also a few fourth-year students from theEngineering Technology program who were taking the course as a technical elective.The paper presents modifications that have been made to the course to increase the overall digitalliteracy of our students. The most significant of these modifications was to change fromMATLAB to Python as the primary programming language [21,22]. This change was made forthree main reasons: 1) Python has more of the attributes common to “true” programminglanguages, such as explicit consideration of data types and references to external libraries, 2) Dueto its widespread use and open-source architecture, the quantity and range of
; teaching ES technical conventions; and building capacityfor project management and project documentation. Engineering students become more accuratein their evaluations of Technical Writing (TW), and better able to distinguish effective andineffective TW after working with these tools. Lastly, teaching students to use ML writing toolsallow engineering educators to effectively promote these learning outcomes in novel ways, whilesupporting professional preparation.1. BackgroundMany higher education institutions are penalizing or restricting students’ use of ArtificialIntelligence (AI) tools at the same time that professors and STEM practitioners are leveragingthem in practical ways. As higher education seeks to identify, control, and in some
, adaptability, user-centric design, and seamless integration, our aim is to forge a pathwaytowards a natural HCI that enriches the immersive and user-friendly aspects of virtual laboratoryenvironments.2. Gesture Recognition with YOLO2.1. YOLO FrameworkThe YOLO framework revolutionizes computer vision and object detection, offering real-time de-tection through a single-pass architecture [10]. Its unique approach enables swift and accurateobject recognition, minimizing computational complexity and supporting applications requiringinstantaneous responsiveness (Figure 1). YOLO's characteristics include: (1) Single pass, real-timedetection, emphasizing a streamlined process for efficient and instant object detection; (2) Unifiedframework for object
models to betterrecognize and highlight semantic errors, while simultaneously providing constructive andmeaningful feedback. Preliminary results from our research are highly encouraging,demonstrating advancements and highlighting the potential of large language models in databaselearning. By integrating these state-of-the-art computational tools into the learning environment,our study lays the groundwork for the creation of intelligent systems that offer nuanced andcontext-aware feedback. Such systems have the potential to enhance the educational experienceand support available to students.1 IntroductionThe de facto standard language for managing and querying relational databases is the StructuralQuery Language (SQL) [1]. Hence, individuals
∞Bots, havebeen used by organizations such as Black in Robotics, Girl Scouts, and Boys and Girls Club aswell as university professors, graduate students, undergraduate students, K-12 teachers, K-12students, and STEM enthusiasts around the world. It will be shown that the modularity of theFlower∞Bots make it suitable for a variety of applications as well as users with varying expertise.1. Introduction Robots are an ideal tool for recruiting diverse populations to STEM due to themultidisciplinary intersections that afford a variety of entry points. For example, robots canillustrate connections between electronics, kinematics, mechatronics, controls, programming, arts,and more. The flexibility of robotics means that it can be taught at a
assessment.1 IntroductionOver the last two decades, the field of computing has been concerned with diversifying thediscipline to better reflect the domestic composition of the US society at large [1]. Given thelucrative jobs in computing, this could be a tool to address the socio-economic disparities inexistence and help improve the social mobility of people from marginalized groups. But thecomputing discipline also benefits from diversifying its workforce. The common news of yetanother system implementing discriminatory practices (e.g., loans, sentencing, facial recognition)has shown that participation from a broader section of the population is a requirement for us toproduce better tools and services.Unfortunately, conversations about equity are
cultural, political, and educationalto economic aspects [1]. The Dominican Republic, as a Caribbean country in the process ofdevelopment, faces considerable educational and economic difficulties, and the effects of thisreality directly impact STEM education, mainly because alliances between academia, the privatesector, and the government are required to ensure the best inputs and practices [2], whichtranslates into investments that are significant for the budget available to educational institutions.This fact has prevented educational centers such as schools, technical institutes, polytechnics, anduniversities, both public and private, from effectively implementing STEM programs throughoutthe national territory, as equipping a single laboratory
Manufacturing Engineering at University of Southern California. His current professional interests include design thinking, collaborative engineering, technological innovation, and education reform. He has over 330 ©American Society for Engineering Education, 2024 ChatGPT and Me: Collaborative Creativity in a Group Brainstorming with Generative AIIntroductionThe emergence of generative AI (genAI), exemplified by ChatGPT, offers unprecedentedopportunities to the education system. However, as this technological advancement gainsmomentum, concerns surrounding hallucination [1, 2] and academic integrity [3, 4] have beenraised, casting doubt on its applicability in educational
teams work to complete. Both projects were introduced to students before the module oncircuitry, but the Introduction to Circuitry lab takes place well before the actual integration of theproject with course skills. Project 1 took place during the spring of 2022 and was comprised of awindmill generation system. The circuitry utilized in this project includes a basic resistor for thegenerative load along with a basic voltage measurement of this output. There is also a proximitysensor circuit that is used as a tachometer to measure the rotational speed of the windmill.Project 2 took place during the 2023 spring semester and was comprised of a water filtrationsystem. This project also incorporated measurement circuitry to monitor the system but
. Additionally, using scaffolding techniques helpslearners progressively develop programming skills. However, determining the appropriate size ofeach conceptual unit depends on factors such as the learners' aptitude and experience.In this paper, we present a data-driven approach to designing auto-graded activities in our online,interactive STEM textbooks, focusing on effectively breaking down complex concepts intosmaller, more achievable steps for learners. We analyzed two types of activities: 1) activities onchallenging topics as reflected by high struggle rates and 2) activities on introductory topics withlower struggle rates, but where students still needed assistance based on their feedback andincorrect submissions as they began learning programming
benchmarking process offers insights into the strengths and limitations ofLLMs in an educational context.This work serves as a call to action for educators across disciplines. By systematicallybenchmarking our curriculums against LLM capabilities, we can better understand the evolvingrelationship between AI and education. This understanding will allow us to refine our teachingmethods, emphasize uniquely human skills, and prepare our students for a future wherecollaboration with AI is expected. As we move forward, it is crucial that we, as educators, takecharge of shaping how these powerful tools are integrated into our classrooms and beyond. Thiswork will illuminate the need for curriculum-based learning outcomes at high levels in Bloom’staxonomy.1
state of Integrated Engineering is examined using established frameworks[1], [2], [3]. The review findings indicate that Integrated Engineering research, models, andcompetencies are rather broad, not well-focused, and connected to higher education andEngineering Education literature. We propose areas for future research and further suggest usinga heatmap conceptualization/framework to measure the interest of the model and competenciesin Integrated Engineering.Keywords: Integrated Engineering, Engineering Education, Review1. IntroductionThis theory/methods paper seeks to expand and enhance understanding of Integrated Engineeringmodels and competencies and offer areas for future work. We review the literature, and followingan interpretive and
between 2010 and the mid-2020s, technology is not merely a tool but an intrinsic part of their environment. They are true“digital natives.” Unlike their Millennial or Generation Z parents and teachers who adapted totechnology as it emerged, Gen Alpha seamlessly incorporates digital tools into their everydayroutines. This generational shift has created a pressing challenge: how to support children inengaging responsibly and creatively with technology while ensuring their developmental needsare met [1]. Efforts to address this challenge have evolved from simply restricting technologyaccess to actively guiding and managing children’s interactions with digital tools. However,the rapid pace of technological adoption by Gen Alpha has outpaced the
a scanner, detect if the gateopened or not, and then drive through the gateway if it opened.At the end of the semester, students were asked to complete a survey regarding their interest inengineering with respect to this new cornerstone project as well as interest based on different skills(programming, circuitry, etc.) used in ENGR 111. The survey included multiple Likert-scalequestions, including one particular question that asked, “How much did the opportunity to workwith a robot for this semester’s cornerstone experience impact your interest in the ENGR 111course?” The Likert-scale was: Not at all, Somewhat, Slightly, Very, and Extremely. Previousresearch shows a relationship between student interest and persistence [1][2][3]. The
into traditional classification models, including Naïve Bayes, K-Nearest Neighbors,and Logistic Regression. The performance of the hybrid model is evaluated in a pass/failclassification scenario. Experimental results show that our proposed CNN-based hybridclassification model outperforms the standalone traditional model in terms of classificationaccuracy. This study introduces an innovative approach in the educational domain,demonstrating that CNNs can provide a more robust and reliable method for predicting studentperformance, especially when predicting binary results like pass or fail.1. IntroductionEducational institutions view their students as valuable assets and are committed to fosteringtheir academic success. Academic performance is a
technique10 can often be successful, and can sometimes be implemented at thediscretion of educators, but is more often incorporated by curriculum or technology developersdirectly. In this work,we investigate the prevalence of avatars as potential role models through theexamination of 312 computing-infused Snap! programming activities created by secondaryteachers and high school interns for non-computer science K-12 classrooms.We seek to answer the following research questions: 1. How do computing-infused lessons created by teachers and high school interns differ in inclusion and usage of avatars? 2. How do creator and avatar demographics correlate?Related WorkAccording to Bandura’s theory of self-efficacy, similarities between a
assignments, making it easier to identify irregular behavior and struggling students so instructors can provide targeted interventions. his paper explores various approaches to effectively utilize the Student BehaviorTInsights feature, providing early use cases and recommendations. It is important to note that Student Behavior Insights is not a "plug-and-play" solution for detecting cheating but is intended for use at the instructor's discretion. Additionally, this feature can serve as a predictor of student struggles. We will present examples of how to use the feature to gain insights into: 1) a student who works earnestly, 2) a student who is utilizing an outside source for their work, 3) a student who
customization, this paper points to new frontiers for delivering student-centeredlearning experiences in engineering education and beyond.IntroductionAs education becomes increasingly complex and specialized, artificial intelligence (AI) offerstools to make teaching and learning more effective, engaging, and equitable [1]. Therefore, wesee artificial intelligence (AI) as a transformative force in education which has a large potentialto offer solutions to challenges posed by traditional, standardized instructional methods.Specifically, modern AI models offer the ability to generate new content in real-time, makingtruly adaptive learning [2] a possibility. These challenges are unique in the context ofengineering education due to the complexity and
these modes of AI as tools for problem-solving, highlight theircomplexities, and explore ethical considerations and biases stemming from data configuration.BackgroundAI has become a fundamental part of the educational curriculum [1]. Its growing importance inrecent years drove its integration into diverse fields such as healthcare, finance, and engineering[2]. Educational institutions are increasingly emphasizing AI to assist students with theknowledge and skills necessary for an advancing job market and to prepare them for a future oftechnological innovations.A key component of this preparation is advancing AI literacy, which refers to an understandingof the use and applications of AI without necessarily requiring deep technical knowledge [3
less effective at differentiating student performance. In contrast, manually createdquizzes offer greater depth, better alignment with course objectives, and foster critical thinking,though they require more effort to design. These findings offer evidence-based insights into thestrengths and limitations of AI in educational assessment. To address these challenges, wepropose strategies for leveraging AI-generated quizzes more effectively, such as incorporatingtargeted prompts and interactive workflows. Overall, this paper provides valuable insights andpractical recommendations to enhance the alignment of AI tools with educational goals andimprove the efficiency of quiz creation.1 IntroductionQuizzes and assessments are fundamental in higher
, and learning approaches in engineering fundamentals students. Thedata gathered from four semesters of engineering fundamentals courses, includes detailed logs ofeach submission, such as submission times, errors identified, critique counts, and resolution times.By analyzing patterns across different submissions, we tracked how student learning evolved overtime and varied across disciplines. This approach allowed us to identify effective strategies inautomated feedback design that cater to the diverse needs of learners from different educationalbackgrounds.1 IntroductionProviding rich, timely feedback to the students when they are learning is a significant challenge inlarge classrooms. This is especially true in first-year engineering
objective measures, show that the Datastorm challenges helpstudents grasp content better and help them improve their soft skills within the context of teamwork.The authors also present feedback from faculty showing the Datastorm challenges’ impact on thequality of information delivery and real-time evaluation options available to Computer Scienceinstructors.IntroductionComputer Science and computing-based majors at the university level face a variety ofchallenges.One significant issue is low student engagement. Many students are unwilling to invest the effortneeded to grasp complex concepts and develop the demanding skills required in ComputerScience [1, 2]. This is a widespread issue with even international surveys reporting lowerengagement rates
-dictive power on performance outcomes. Finally, we call for continued empirical research on theefficacy of LLM-based technologies in STEM education and propose future research directions inexploring their impact on teaching and learning.1 IntroductionThe introduction of OpenAI’s ChatGPT in November 2022 [1] triggered an unprecedented surgeof interest in applications of artificial intelligence (AI) based on Large Language Models (LLMs)and their underlying transformer architecture.In particular, LLMs appear to be exceptional in applications that involve human interaction, infor-mation retrieval, and summation, making them an attractive prospect for improving the effective-ness and accessibility of education in the digital age [2, 3, 4]. However
Flipped Classroom Applications in Engineering EducationThe flipped classroom model has emerged as a transformative approach in engineering education,addressing limitations of traditional teaching methods 1. By shifting theoretical content delivery topre-class activities and dedicating class time to active learning, this model fosters deeperengagement, better conceptual understanding, and enhanced collaborative skills. Studies haveshown its effectiveness across various engineering disciplines, including mechanical engineeringcourses such as statics 2 , rigid body dynamics 3 , and thermodynamics 4 . Bishop and Verlegerhighlighted that flipped classrooms promote active learning, enabling students to tackle problem-solving and design challenges
work.Previous Work Practical laboratory experiences including engineering labs and projects represent essentialelements of learning [1], [2]. As part of intensive laboratory experiences, robots have had alongstanding positive impact on education of students at all levels. Small, wheeled, programablemobile robots like LEGO Mindstorm series have been used as motivational tools to attract studentsto STEM fields in general [3], as well as to help students (and teachers) learn how to program [4]- [6]. However, at the practical level of industrial robot programming, the use of industrialmanipulators for teaching programming robotic tasks was often the only option. Expensivehardware, proprietary software, and required safety measures made programming of