Analysis and Improvement Paths of Curriculum Goal Achievement in Data Visualization based on the OBE Concept

Fei Hou, Zhihui Li, Shaoze ZhangInternational Journal of Education and Humanities 2026 · DOI

TL;DR

This paper explores how to modernize Data Visualization education using the Outcome-Based Education (OBE) framework. By shifting the focus from traditional content delivery to measurable student competencies, the authors propose a systematic "reverse-design" approach to bridge the gap between classroom theory and the practical demands of the big data industry.

What Problem Does This Paper Address?

In the rapidly evolving field of Data Science, the "Data Visualization" course often struggles with a disconnect between academic instruction and industrial reality. Traditional teaching models frequently prioritize lecturing through a syllabus (content-oriented) rather than ensuring students can actually execute complex projects (outcome-oriented).

The authors identify several specific pain points:

  • The "Scattered Theory" Trap: Students often learn tool syntax (like Matplotlib or Seaborn) in isolation without understanding the underlying design logic or data preprocessing requirements.
  • Evaluation Imbalance: Grading often relies heavily on exams, which fail to capture a student's ability to manage a project, work in a team, or communicate insights effectively.
  • Technological Lag: Curriculum content often misses cutting-edge trends like AI-assisted visualization and multi-modal data handling.

Methods

The authors apply the OBE Concept, which is built on a "reverse-design" logic: first define the desired student outcomes, then build the curriculum and evaluation systems to meet them. They categorize these outcomes into three dimensions:

  1. Quality Goals: Cultivating ethics, social responsibility, and a "craftsmanship spirit" regarding data authenticity.
  2. Knowledge Goals: Mastering core principles and libraries (Matplotlib, Seaborn, Pyecharts).
  3. Ability Goals: Practical skills in data cleaning, interactive graph drawing, and cross-disciplinary communication.

To achieve these, they propose a four-dimensional improvement path:

  • Content Reconstruction: Moving to a "basic-core-comprehensive" modular structure.
  • Method Innovation: Utilizing "flipped classrooms" (pre-class videos) and project-driven teaching.
  • Diversified Evaluation: Implementing a "process + outcome" model that includes peer reviews, oral defenses, and refined quantitative scoring for project reports.
  • Practical Strengthening: Increasing in-class lab hours and integrating real-world enterprise projects.

Datasets

While this is a pedagogical research paper rather than a machine learning study, the authors highlight several teaching case libraries used to evaluate student achievement:

  • 315 Gala Data: Used for value mining and data ethics discussions.
  • New Retail Intelligent Sales Data: A practical project for business visualization.
  • Radio and Television Big Data: Used to test students' ability to handle large-scale, industry-specific datasets.
  • Public Health/Epidemic Data: Used to explore social responsibility in data representation.

Key Insights and Ideas

  • Measurability via Bloom’s Taxonomy: The paper emphasizes using specific action verbs (e.g., "independently complete," "master," "possess") to make curriculum goals measurable and achievable.
  • The "Closed-Loop" Feedback: A core tenet of their OBE approach is the continuous improvement mechanism. Evaluation results aren't just grades; they are data points used to dynamically optimize the teaching process for the next cohort.
  • Interdisciplinary Integration: Data visualization isn't just coding; the authors argue for the inclusion of "design aesthetics" and "user experience" (UX) modules to ensure students create works that are both functional and effective.
  • Virtual Simulation: To mimic real-world work environments, the authors suggest using virtual simulation platforms where students can conduct practical operations in immersive, high-stakes scenarios.

Key Results

The paper provides a roadmap for curriculum reform rather than a quantitative experimental result. However, it outlines significant structural changes:

  • Increased Practicality: A recommendation to increase in-class practical hours from 8 to 12-16 hours to ensure sufficient "keyboard time."
  • Evaluation Shift: Moving from a standard exam-heavy model to a diversified mix: assignments (20%), quizzes (25%), and a combination of project reports and oral defenses (30%+) to better reflect "Ability Goals."
  • Standardized Scoring: The introduction of refined quantitative indicators for project reports (e.g., scoring based on data cleaning standardization and the rationality of visualization design) to reduce grader subjectivity.

Limitations and Future Work

The authors acknowledge that the current integration of "Quality Goals" (ethics and values) remains somewhat superficial, often relying on classroom discussion rather than immersive practice.

Future directions include: * Deepening University-Enterprise Collaboration: Building stable practical teaching bases with big data companies to provide students with "live" industrial challenges. * Long-term Tracking: Continuously collecting student outcome data over multiple years to refine the "goal-implementation-evaluation" loop. * AI Integration: Further exploring how artificial intelligence can be integrated into the visualization workflow as a teaching tool rather than just a topic of study.

Summary updated

Abstract

OBE (Outcome-Based Education), as the core concept of talent training in emerging engineering disciplines, emphasizes student learning outcomes as the orientation and improves teaching quality through a continuous improvement mechanism. Data Visualization, a core compulsory course for the major of Data Science and Big Data Technology, integrates theoretical and practical aspects. Its teaching quality directly affects the cultivation of students' data analysis and visualization practical abilities. Taking the "Data Visualization" course as the research object, this paper constructs a curriculum goal achievement evaluation system based on the OBE concept, analyzes the current status and existing problems of curriculum goal achievement from three dimensions (quality, knowledge, and ability), and then proposes targeted improvement paths, providing a reference for the teaching reform of similar courses.
BibTeX
@article{hou2026analysis,
  author = {Fei Hou and Zhihui Li and Shaoze Zhang},
  title = {{Analysis and Improvement Paths of Curriculum Goal Achievement in Data Visualization based on the OBE Concept}},
  journal = {International Journal of Education and Humanities},
  year = {2026},
  doi = {10.54097/x5b1bx53},
}
Figure