Paper ID #49218From Reflection to Insight: Using LLM to Improve Learning Analytics inHigher EducationDr. Nasrin Dehbozorgi, Kennesaw State University I’m an Assistant Professor of Software Engineering and the director of the AIET lab in the College of Computing and Software Engineering at Kennesaw State University. With a Ph.D. in Computer Science and prior experience as a software engineer in the industry, my interest in both academic and research activities has laid the foundation to work on advancing educational technologies and pedagogical interventions.Mourya Teja Kunuku, Kennesaw State University Ph.D. student at
instance, it is assumed that students learn debugging by havingexperience with debugging [13]. However, a study by Whalley and colleagues revealed thatstudents’ reflections on their experiences with debugging tend to be negative [14]. In this study,students expressed that exploring strategies such as print statements frequently will make themmiss the program’s general idea, forcing them not to follow a methodological approach [14].Although debugging is a challenging task, it is also an essential skill that students must master toacquire other computational thinking skills [15]. Consequently, exploration of students’debugging skills is essential to develop teaching and learning strategies that fully explode theiralready-in-place preferences and
Science Outstanding Mentor Award. ©American Society for Engineering Education, 2023 Labor Based Grading in Computer Science - A Student Centered PracticeAbstractInnovation in teaching in STEM fields was explored widely during the COVID pandemic in 2020. Thispaper describes the adaptation of labor based grading for computer science courses. Labor based gradinghas been developed for language and writing courses by shifting the grading focus from summative examsto formative and reflective assessments. The method was tested in several computer science courses withtwo different instructors during the 2020-2021 academic year. Students were surveyed to understand howthey perceived grading methods
Inclusion and Accessibility: The Impact of Inclusive Design on UX Career PreparationTaylor M. Smith, The University of Texas at Austin; Hansika Murugu, University of Maryland, College Park; & Earl W. Huff Jr., The University of Texas at Austin[THE SHIFT TOWARDS INCLUSION AND ACCESSIBILITY] 2 AbstractThis study examines the transformative impact of inclusive design education on informationscience students’ career goals and their understanding of technology development. Throughanalyzing student reflection journals and conducting follow-up interviews, the research exploreshow exposure to inclusive design principles
projects. Thisresearch also analyzes how adult learners interactively learn, reflect, and apply their AIknowledge to examples drawn from their workplace, while improving their understanding andreadiness to implement AI technologies effectively.Our three-day workshop centered around enriching and engaging learning about AI technologies,ethics, and leadership, featuring topics like supervised learning and bias, AI strategy, andgenerative AI. Apart from discussions, the workshops incorporated hands-on learning with digitaltools, robots, problem-solving scenarios, and a capstone project. Participants were 44 leadersfrom a large government organization. Their learning was measured through pre- andpost-questionnaires on AI leadership, knowledge checks
it well worth the effort. The opennessof project topics has led to student creativity and expression in class projects, including theembracing of their unique identities and exploration of more advanced materials under instructorguidance. Projects that address a gender-specific, interest-specific, or queer concern also letstudents (the project makers and their classmates alike) understand that computing applies inmany disparate domains and there is great value to a diversity of voices in technology. Thispaper describes the approach, general project design outline, the ethical reflection embedded inthe project, and experiences from several years of teaching (since Fall 2017). A list of studentprojects with brief descriptions is included so
insight into the effectiveness of theassignment and which parts are most difficult for students to understand. Students alsoresponded to the reflection prompt “What was the most surprising or interesting part of thisactivity.” The responses were analyzed for common themes, which were the usefulness ofvisualizing memory in understanding the concepts of stack frames and buffer overflow, theprevalence of buffer overflow vulnerabilities in publicly available code, and how easy it is toexploit a buffer overflow vulnerability. Thus, this assignment shows promise in helping studentsto understand a difficult concept, and in emphasizing the importance of avoiding buffer overflowvulnerabilities.IntroductionSoftware vulnerabilities in commercial products
]). The SerenePulse webapp harnesses awebcam or selfie camera to capture heartbeats by analyzing fluctuations in light intensity reflected from the skin,a fundamental principle of rPPG technology.HRV metrics In this research, we build upon previous research [16] that detailed on HRV metrics and stress analysis. Heartrate variability (HRV), calculated from the input of rPPG, inter-beat intervals is a crucial physiological markerthat offers insights into the autonomic nervous system’s (ANS) functioning [16]. It reflects the dynamic interplaybetween the sympathetic and parasympathetic branches of the ANS, highlighting the body’s adaptability to stressand relaxation states [39]. Among HRV metrics, SDNN is indicative of autonomic flexibility and
computing is the reality of the computing education “culture” in the U.S.being primarily one-note (e.g., white-men)—including faculty, students, and professionals—which instigates perpetual curricular and non-curricular hurdles for members of non-majoritygroups to overcome. To attain their fit within computing, students must navigate the computerscience culture by adopting norms and values that are reflective of the majority-group [22]. Notbeing able to adopt these norms and values impacts students’ fit within computing contexts and,ultimately, their retention.Culture is a compelling explanation for underrepresentation in computer science. This identifiedone-note cultural concern in computing contexts where non-majority computing students
write values on the cars. Several examples of Play-Doh trains constructed by students are shown in Figure 1. 3. Manipulation (20 minutes): Students are guided through a series of operations on their trains, simulating common linked list operations: • Insertion: Adding a new car to the beginning, middle, or end of the train. • Deletion: Removing a car from the beginning, middle, or end of the train. • Traversal: Following the links (toothpicks) from the head of the train to the end, simulating the traversal of the linked list. 4. Discussion and Reflection (10 minutes): Students engage in a guided discussion, reflecting on their observations and drawing connections between the Play-Doh
instructor reflects this as demonstrated by student performance on a linkedlist implementation assignment as well as an unbalanced binary search tree implementation assignment. Bothassignments showed improved performance; and quicker submission times with more than half students turningin the unbalanced binary search tree assignment before the deadline. Finally, more students chose animplementation project as their final project, between the choice of completing an implementation project or aproject studying a data structure not covered in the course, than have typically done in the past. On this lastpoint, I will share experiences with the breakdown of students opting for one or the other from many semestersteaching data structures across several
’ understandingof the overall module to see whether they meet the module objectives and a survey withopen-ended questions to help students reflect on their learning and experiences with the module,the second of which we discuss in more detail in the next section. Below are example quizquestions, with the correct answer choice italicized, relating to each of our three learningobjectives. • Question related to Objective 1: One student gives work to another, knowing that the student is going to copy the work directly and submit it for credit. Who has committed an academic violation? Answer choices (choose one): (1) Both the student that copied and the student that provided the material. (2) The student that provided the material. (3
to their field’s broader conversation[5]. Undergraduate engineering students who wrote voluntary reflective journal entry essays onassigned readings performed better on multiple-choice quizzes about the readings than those whodid not [6]. Writing tasks serve not only as a means of expression but also as a vehicle forlearning and broader intellectual exploration.In the meantime, the advent of generative AI (artificial intelligence) tools has transformedwriting practices in both K-12 and higher education. Generative AI is transforming K-12 andhigher education by enhancing students’ writing skills. In K-12, it provides real-time assistanceby improving grammar, syntax, and style, enabling students to complete assignments such asessays, summaries
simulation is running in Tinkercad Circuits. However, the output in the serialmonitor will reflect whichever Arduino was selected at the beginning of the simulation start. Forexample, while interacting with the potentiometer, only the analog circuit (lower one in Fig. 2)will display output in the serial monitor. In contrast, if the student clicks on the upper arduinobefore clicking the StartSimulation button, s/he will notice the serial monitor starts displaying0 (the default digital output when push button is not pressed) on the serial monitor. As soon ass/he presses the push button, the serial monitor will print 1 and then go back to the default 0state.Graph Output: Tinkercad allows to visualize the circuit output data in graph format. Though
]. With the increase in research publications, the focus on impact indicators has broadened,with citation counts remaining a widely accepted measure. Yet, they are not direct measures ofquality [7]. Despite controversies around these metrics, they continue to be used in academicdecision making. An additional metric that is being used more for evaluating scholarship are downloadcounts [8, 9]. Using downloads reflects a broader view of research impact, considering the actualusage and dissemination of scholarly works. While there is a correlation between downloadmetrics and citations [10], there are situations where this is not the case. For example, papers withfewer citations might be extensively downloaded and used by practitioners
explained to students as a note-taking tool to distill their understanding of Python. For students theyare meant to act as personal documentation of Python’s syntax and semantics. In this research, rulesnotebooks were a window into students’ conceptions of Python’s syntax and semantics. We view therules notebooks as reflective activity germane to the conclusion and discussion phases of IBL.5 Data and Preliminary AnalysisOur analysis of the data from this course is ongoing, and we are collecting more data on a second iterationof the course. This work-in-progress paper reports on preliminary review of rules notebooks as well aswritten assignments and observations from the course. Preliminary analysis of students’ rules notebooksindicated that
. DiscussionThis research aims to examine students’ in situ demonstration of the cognitive and behavioralskills associated with algorithmic thinking in an introductory computing course in engineering.Our findings indicate that while students are frequently able to produce working code that solvesa wide array of computing problems, their submissions do not always reflect the cognitive skills,such as algorithmic thinking, that are central learning goals in introductory CS education. Thesefindings lead us to question the utility and appropriateness of autograders for assessing andevaluating student learning, particularly as it relates to complex cognitive skills in CS education.Existing research suggests instructor feedback supports students’ learning beyond
from participating in the surveys and interviews in the study.Section 2 (Evening): 13 students were included, constituting 24.5% of the sample. Theseparticipants were 25 to 36 years old, and three declined to participate in the surveys andinterviews.Section 3 (Evening—Control): Ten students formed this section, representing 18.9% of the totalsample. All participants in this section, aged between 25 and 36, completed the surveys andparticipated in the interviews.The gender distribution among the participants was 11.32% female students and 88.68% malestudents, reflecting the typical demographics of industrial engineering programs in the region.This study set out to explore students' perceptions of the effectiveness of active learningcompared to
everyone. Theyare in this space after having years of experience in the degree, and I did not.” This sense ofinadequacy often stems from insufficient preparatory support at key academic transitions. Manystudents reported feeling unprepared for advanced coursework or graduate-level studies due togaps in their foundational education. Institutions could implement "bridge programs" thatprovide intensive preparatory support for students transitioning between academic stages,whether from high school to undergraduate programs or from undergraduate to graduate studies.The rise of external online courses as a supplementary educational tool reflects the inadequaciesof formal CS programs in meeting students' needs. Students often enrolled in platforms
Reflections, Review Review Review Review 4:15 - 4:30 Feedback, 4:30 - 4:45 Photos Reflection Reflection Reflection Reflection Closing and Thank You! 4:45 - 5:00emotional intelligence [30, 31], and effective communication skills [32].Introductory technical skills were covered early in the Guild workshop so that the participantscould start applying these skills and programming languages
experience from the workshop, content of theworkshop, and their experience within the team. • For team collaboration, the score was based on the synergy of the work and the team members. Each team was required to reflect on the collaboration among the team members in terms of what worked and what did not work for their interdisciplinary collaboration. • For the final presentation, we suggested a presentation structure and outlined the main components in the presentation. Each team member was required to participate in the presentation. In addition, each member was asked to present part of the materials that was outside their background (i.e., the work of other team members). As a result, participants in each team needed
class. In the control group,the use of LESs was minimized, while the treatment group had increased LESs. The exams forboth groups were very similar or exactly the same. A statistical analysis of the results using theMann-Whitney U test showed a statistically significant difference between the groups. The Exam2 scores for the control (minimal LESs) group (Mdn = 72.4%) differed significantly from thetreatment (increased LESs) group (Mdn = 77.7%), where U = 2421, z = 2.875, p = 0.004, r = 0.26.This work provides evidence that using LESs or other active learning approaches has consistentlyimproved student learning outcomes, as reflected by the exam scores.Benefits of Teamwork. Teamwork is essential for student development in terms of knowledgeand
thatassessment practices have on the student experience but limited research has examined this topic.This paper begins to fill that research gap by addressing the research question: How do courseassessment practices affect students’ perspectives of learning technical writing?I conducted an interpretive qualitative study, grounded in Lave and Wenger’s Situated LearningTheory and Social Theory of Learning, with 10 third and fourth-year computer science studentparticipants. I used reflective journal writing and beginning-of-term and end-of-term interviews togather rich data on the student experience. I generated themes from the data corpus via Braun andClarke’s reflexive thematic analysis and found that students are conflicted in their desire to
began in the fall of 2019; lectures were recorded during the spring of 2020 face-to-face sessions using lecture capture technology, then edited and adapted for the online course,which launched in the summer of 2020.One of the main challenges in transitioning to an asynchronous format was preserving the dynamicinteractivity that supported student learning and reflected the instructor's distinctive teaching style. Inthe face-to-face delivery of this course, the instructor regularly used graded clicker questions andpaused at frequent intervals during the lecture to implement various questioning techniques to engagestudents and monitor their comprehension. Recognizing the importance of this course component, thedesign team looked for ways to
- Cybersecurity Planning and Management (CPM)CPM-1: Examine the placement of security functions in a system and describe the strengths andweaknessesSource: Final Project Individual Reflection Question 2 which provided a network diagram andasked students to identify strengths and weaknesses. EAMU Vector (19,0,0,0)CPM-2: Develop contingency plans for various size organizations to include: businesscontinuity, disaster recovery and incident response.Source: Final Project Individual Reflection Question 3 which provided three scenarios and hadstudents answer how to achieve various goals. EAMU Vector (18,1,0,0)CPM-3: Develop system specific plans for (a) The protection of intellectual property, (b) Theimplementation of access controls, and (c) Patch and change
people and circumstances that differ from those with which students are familiar. Frequent, timely, and constructive feedback. Periodic, structured opportunities to reflect and integrate learning. Opportunities to discover the relevance of learning through real-world applications. Public demonstration of competence.While not all HIPs address each element to the same degree, the list provides a standard forjudging the quality of implementation. It could potentially be used to assess the quality of otherevidence-based curricular and co-curricular activities as well.The most common outcome studied across all high-impact practices is student retention andacademic performance (grade point average). For both measures the result is
Paper ID #49295BOARD #106: Investigating Factors Influencing Performance in an IntroductoryProgramming CourseAmanda Nicole Smith, University of Florida Amanda is an undergraduate student pursuing a Bachelor of Science in Computer Science at the University of Florida, with an expected graduation in Spring 2025. Her research interests focus on computer science education, particularly how educators can use machine learning models to provide real time intervention strategies to optimize individual student outcomes. This paper is a reflection of her commitment to improving educational strategies and fostering an inclusive
. Real-time assessments,such as quizzes or activities during lectures, were perceived as less engaging by some students, likelybecause of the pressure to respond immediately and the lack of time for reflection (unless specificallybuilt into the assessment). Overall, scaffolded projects emerged as the most consistently favored format,while multimodal and real-time assessments showed potential but may require further refinement tomeet diverse student preferences.Students perceive that redesigned assessments significantly improve their critical thinking skills, witha mean rating of 4.27 and a standard deviation of 0.90. Additionally, they believe these assessmentsenhance their ability to apply course concepts to real-world situations, as reflected
helped them learnfrom mistakes, while a significant portion (27.3%) disagreed. The percentage of respondentswith negative opinion dropped to 7.1% in 2024. In 2021, 18.2% of respondents disagreed thatthey were able to study at their own pace while this percentage dropped to zero percent in 2024.Students were getting familiar with the trial-and-error learning approach by taking more courseswith automated grading. In other areas, the differences in opinions between 2021 and 2024 werenot significant. Overall, respondents in 2024 were slightly more positive than 2021, probably dueto smoother experience related to equipment upgrade. Most respondents in both years werepositive about the lab experience.One issue reflected in the comparison may be
observed but no long-term career outcomeevaluation. These studies collectively demonstrate the positive impact of inquiry-based learningin scientific education, albeit with a need for more extensive, long-term evaluations.Dickerson et al. [20] employed a distinctive approach to foster reflection among engineeringstudents within the context of a digital circuits course. This method integrated computer-basedsimulation for digital circuit design with reflective thought prompts administered after a midtermexam for post-exam analysis and contemplation. The study also underscored the significance ofemploying thought-provoking question prompts designed to voluntarily elicit comprehensivereflections after a significant milestone event, such as a midterm