Abstract
The programming course is a vital public requirement for undergraduates, primarily aimed at cultivating the ability to analyze and solve professional problems through Computational Thinking (CT). Using C language programming as a case study, this research proposes a teaching method oriented toward CT across four dimensions: classroom instruction, experimental teaching, course assessment, and stratified teaching. Results indicate that incorporating CT significantly enhances student learning outcomes compared to traditional pedagogical modes.
Introduction
Traditional programming education often overemphasizes the grammar and syntax of specific languages. While students may memorize how to write a for loop, they frequently struggle to apply these tools to complex, real-world problems. This research addresses the disconnect between "knowing the code" and "thinking like a computer scientist" by shifting the focus toward abstraction, automation, and decomposition.
Methodology
The authors developed a four-pronged strategy to reform the undergraduate C language curriculum at the Space Engineering University in Beijing.
1. Classroom Teaching Integration
Specific C language features were mapped directly to CT concepts to help students build mental models of computation:
- Abstraction: Variables are taught as abstractions of system characteristics, while functions represent abstractions of functional modules. This helps students view code as a model of reality.
- Automation: By explaining how hardware components like EIP registers and stacks automatically execute code, the "magic" of software is grounded in physical computer architecture.
- Generalization: Students compare multiple algorithmic approaches (e.g., different branching structures) to find the most efficient solution for a given problem scale.
2. Experimental Teaching: The 7-Step Method
To prevent students from feeling overwhelmed by a blank IDE, the research introduced a structured pipeline where coding is the penultimate step rather than the first:
- Analyze the problem
- Assign initial values
- Choose basic structures
- Determine output
- Draw a flowchart
- Write code
- Debug
3. Assessment and Stratification
The final exam was overhauled to prioritize system modeling and algorithm design over rote grammar. The weight of programming-heavy questions increased from 30% to 50%. Additionally, the curriculum was adjusted to ensure that different levels of learners could grasp the core logic of computation through stratified instruction.
Key Results
The implementation of this CT-oriented method yielded several quantitative and qualitative improvements:
- Improved Exam Performance: Final exam scores saw a significant increase across the student body.
- Lower Failure Rates: The make-up exam rate was reduced to just 5%, despite the increased difficulty and higher proportion of programming-intensive questions.
- Competitive Success: Students achieved excellent results in national-level professional competitions, such as the National Green Computing Contest.
- Increased Engagement: Faculty observed a general improvement in student enthusiasm and interest in programming as a problem-solving tool.
Conclusion
By treating code as an abstraction and emphasizing the logic of automation before the syntax of the language, this pedagogical shift effectively bridges the gap between theoretical knowledge and practical application. Future work will focus on scaling these results through an online and offline hybrid teaching model to further refine the integration of computational thinking in modern engineering education.
Summary updated