A Longitudinal Study of Student Performance and Curriculum Evolution Across Repeated Course Offerings Under the Same Instructor in Creative-Technology Education (Fall 2023 – Summer 2025)

A Longitudinal Study of Student Performance and Curriculum Evolution Across Repeated Course Offerings Under the Same Instructor in Creative-Technology Education (Fall 2023 – Summer 2025) Abstract This dataset presents a three-year longitudinal record of student performance across repeated course offerings in the field of creative-technology education.Collected from the Department of Multimedia and Creative Technology (MCT) at Daffodil International University (DIU), Bangladesh, it includes 755 anonymized student-course entries covering both theory and laboratory modules taught between Fall 2023 and Summer 2025. The dataset captures performance metrics under a single instructor, S.M. Monowar Kayser, ensuring uniformity in teaching methodology and grading policy.It provides a reliable empirical foundation for exploring curriculum evolution, grading consistency, learning outcomes, and AI-based educational analytics in multimedia and creative disciplines. Instructor S.M. Monowar KayserLecturer, Department of Multimedia and Creative Technology (MCT)Daffodil International University (DIU), Dhaka, [e-mail] Dataset Overview Attribute Description Institution Daffodil International University (DIU) Department Multimedia and Creative Technology (MCT) Semesters Fall 2023, Spring 2024, Fall 2024, Spring 2025, Summer 2025 Course Families 8 (2D Animation, Character Animation, Lighting & Rendering, Audio and Video Streaming & Editing, Architectural Visualization, Design Principles, Landscape Simulation, Physics-Based Animation) Course Instances 41 unique sections (theory and lab) Instructor Consistency All courses taught and assessed by the same instructor Assessment Framework Theory Courses Component Weight Description Attendance 7 % Regular participation in class activities Quizzes / Single Project (3×15 → Average) 15 % Short assessments or a creative mini-project Assignments 5 % Individual design or written work Presentations 8 % Visual or oral demonstrations of concepts Midterm Exam 25 % Mid-semester assessment Final Exam 40 % Comprehensive final evaluation Total = 100 marks (converted to percentage). Laboratory Courses Component Weight Description Attendance 10 % Lab participation and engagement Lab Performance 25 % Continuous assessment of technical skill Lab Report / Assignment 25 % Documentation of projects or tasks Lab Final Exam 40 % Final practical demonstration Total = 100 marks (converted to percentage). Data Composition Each row represents one student record per course per semester.All identifiers are synthetic (STU-YYY-NNN). Main Fields: Student_ID, Semester, Course_Title, Course_Family, Course_Type, Section Performance metrics (Attendance %, CT_Avg %, Midterm %, Final %, Lab_Final %) Aggregate scores (Grand_Total %, Final_Grade, Completion_Flag, Performance_Comment) Derived features for research: Cohort_Average %, Z_Score_Performance, Lab_Theory_Gap, Difficulty_Index Key Findings from Data Analysis Average Performance: 60 % overall; labs consistently 3–5 points higher than theory. Completion Rate: ≈ 93 %, indicating high academic success. Most Common Grades: B+ (125), A− (93), A (78). Performance Distribution: Medium (41 %) dominates followed by Low (27 %) and High (20 %). Curricular Progression: Later semesters show improvement in average marks and lower variance, reflecting effective curriculum refinement. Research Applications This dataset can be used for: Longitudinal performance tracking in design and technology education Comparative studies of theory vs practical learning outcomes AI and ML model training for grade prediction and performance analytics Curriculum evaluation and educational policy design Instructor effectiveness and fairness assessment Because the data originate from a single, consistent instructor, they offer a controlled environment for statistical and machine learning research in education. Ethical Note All student information has been anonymized.No personal data or identifiers are included.The dataset complies with DIU’s ethical standards for research and data sharing. Keywords creative-technology · educational data mining · longitudinal study · curriculum evolution · student performance · learning analytics · multimedia education · theory-lab correlation · academic assessment · creative pedagogy · art and technology · educational data science · Bangladesh · DIU · performance forecasting · data-driven education · teaching consistency · AI in education · design education · academic analytics