Student experiences in a computational physics lab through the lens of Physics Computational Literacy

This study utilizes the framework of Physics Computational Literacy to analyze semi-structured interviews with upper-division students, revealing that they navigate tradeoffs in developing various literacy aspects and approach social computational literacy differently based on unspoken assumptions.

The Gist

Imagine learning to drive a car. You have to learn three things at once:

1. The Mechanics: How to actually turn the key, press the gas, and steer (the Material skill).
2. The Logic: Understanding why the car moves, how the engine works, and how to navigate a map to get to your destination (the Cognitive skill).
3. The Traffic: How to talk to other drivers, merge safely, and follow the unspoken rules of the road so everyone gets where they need to go (the Social skill).

This paper is about a group of researchers at Oregon State University who watched students trying to learn "Computational Physics." Think of this as teaching students to drive a very complex, futuristic spaceship using code instead of a steering wheel.

The researchers wanted to know: How do students actually experience this learning process? To find out, they interviewed five students halfway through their training. They used a special "lens" called Physics Computational Literacy to look at what the students said.

Here is what they found, broken down into simple stories:

1. The "Juggling Act" (Trade-offs)

The biggest discovery was that students often feel like they are juggling three balls, and they can't keep all three in the air at the same time. They have to drop one to catch the others.

• The Story of Betty: Betty wanted to understand the physics deeply (the Logic), but she felt that working in a group slowed her down. She decided to work alone so she could focus on the code.
  • The Trade-off: She chose to drop the Social ball (working with others) to keep the Material (coding) and Cognitive (thinking) balls spinning.
• The Story of Ethan: Ethan liked to figure things out by himself first because he felt he thought clearer when alone. He would only ask for help when he was truly stuck.
  • The Trade-off: He prioritized his own Cognitive understanding over the Social aspect of collaboration.
• The Story of Alice: Alice told the researchers, "It's more important to understand why the code works than just to finish the assignment and get the graph."
  • The Trade-off: She chose Cognitive (deep understanding) over Material (just getting the job done).

The Takeaway: Students are smart enough to realize they can't do everything at once. They make conscious choices about what to focus on, often sacrificing social interaction to get the technical work done.

2. The "Unspoken Rules" of the Road (Social Literacy)

The researchers also found that how students handle the "Social" part of coding varies wildly, often based on things nobody ever explicitly taught them.

• The "Twin" Effect: One student, Carl, said he did much better this term because he was paired with someone who knew exactly as much coding as he did. It was like having a co-pilot who speaks your language. When they were on the same level, they could help each other without one person feeling left behind or bored.
• The "Anxiety" Factor: Alice, who has anxiety, said she sits alone. She knows how to ask the teachers (TAs) for help, but she doesn't talk to her classmates. She believes group work works for most people, but it doesn't work for her.
  • The Problem: The class assumes "Social Literacy" means sitting in a circle and chatting. But for Alice, "Social Literacy" means quietly asking an expert for help. If the class only rewards the "chatting" style, they might accidentally leave students like Alice behind.
• The "Etiquette" Lesson: Another student, Derek, noticed that in one class, nobody talked about how to write clean code or add comments (notes for other humans to read). In the next class, they did. He realized that writing code isn't just for computers; it's a way of talking to other humans.

3. What Should Teachers Do?

The researchers concluded that we need to change how we teach this subject.

• Make the "Social" part explicit: Don't just assume students will learn to collaborate by osmosis. Teachers need to explicitly teach students how to work together, how to comment on code, and how to explain their thinking to others.
• Be Flexible: We can't force everyone to learn in the same "social" way. Some students need to work alone to think clearly; others need a partner. The classroom needs to be a place where different styles of "socializing" with code are accepted.
• Balance the Balls: Teachers need to help students stop juggling and start integrating. They need to show students that you can talk about code (Social) while you are figuring out the physics (Cognitive) and writing the program (Material).

The Bottom Line

Learning to do physics with computers is like learning a new language, a new sport, and a new social culture all at once. This paper tells us that students are often struggling to balance these three parts.

The solution isn't to force everyone to be a "team player" in the exact same way. Instead, teachers need to be like a skilled dance instructor: they need to show the steps clearly, acknowledge that some dancers prefer to practice alone before joining the group, and make sure everyone knows that the goal isn't just to finish the dance, but to understand the music and enjoy the company.

Technical Summary

Here is a detailed technical summary of the paper "Student experiences in a computational physics lab through the lens of Physics Computational Literacy" by Nearhood and Hamerski.

1. Problem Statement

Computational physics is a fundamental component of 21st-century physics practice, yet upper-division computational physics education remains a significantly understudied area. While research in this field has expanded, there is a lack of comprehensive theoretical frameworks applied to upper-division curricula. Furthermore, existing curricula are often inconsistent ("spotty"). The authors identify a critical gap in understanding how students experience and learn computational physics in these advanced settings, specifically regarding the trade-offs they make between different facets of learning and the social dynamics of their development.

2. Methodology

• Context: The study was conducted at Oregon State University within a specific junior-level course sequence consisting of three 10-week, single-credit lab courses. These courses run alongside upper-division physics courses in the "Paradigms" program. The curriculum utilizes Jupyter notebooks and follows the structure of "computational essays."
• Participants: Five students were interviewed at the end of their second term in the three-term sequence (Winter 2024).
• Data Collection: The first author (a graduate researcher not involved in teaching) conducted semi-structured interviews averaging 40 minutes. Questions focused on student interactions, advice for future students, coding backgrounds, and the connection between coding and physics problem-solving.
• Analytical Framework: The study utilized the Physics Computational Literacy (PCL) framework (Odden et al.), which categorizes computational literacy into three distinct aspects:
  1. Material Computational Literacy (MCL): Practical use of computing (e.g., coding, syntax).
  2. Cognitive Computational Literacy (CCL): Understanding the computing process and using it to understand physics concepts.
  3. Social Computational Literacy (SCL): Collaborative work, communication, and norms regarding code.
• Analysis: Interviews were transcribed and coded using Joseph Maxwell's system (organizational, substantive, and theoretical coding). The coding process identified emergent themes and mapped them to the PCL framework to reveal tensions and trade-offs.

3. Key Contributions

• Application of PCL to Upper-Division Education: The paper successfully applies the PCL framework to an upper-division context, demonstrating its utility in breaking down complex student learning experiences into manageable, theoretically grounded components.
• Identification of Inter-Literacy Trade-offs: The study reveals that students are not passively learning all aspects of computational literacy simultaneously; rather, they actively negotiate and make trade-offs between Material, Cognitive, and Social literacies.
• Reconceptualization of Social Literacy: The authors challenge the assumption that "social" learning only occurs through direct, real-time group collaboration. They highlight that social literacy can be developed through individual reflection, anxiety management, and adherence to coding "etiquette" (documentation/commenting).

4. Results

The analysis yielded two primary findings:

A. Inter-Literacy Trade-offs
Students expressed conscious tensions between the three aspects of PCL, often prioritizing one at the expense of another:

• MCL vs. CCL: Student "Betty" prioritized getting code to run (MCL) over understanding the underlying physics (CCL), explicitly stating she thought less about physics to ensure the code worked. She resolved this by working alone to focus on the code.
• SCL vs. MCL/CCL: "Betty" also noted that group work (SCL) sometimes hindered her ability to think deeply about the code (CCL/MCL). Conversely, "Ethan" preferred working alone first to clarify his thinking (CCL) before engaging socially, viewing social interaction as a tool to accomplish material tasks rather than a primary learning mode.
• CCL vs. MCL: Student "Alice" explicitly prioritized understanding why code works (CCL) over simply finishing assignments and generating graphs (MCL), viewing the former as more valuable for long-term success.

B. The Variable Nature of Social Computational Literacy (SCL)
The development of SCL is highly individual and driven by unspoken assumptions:

• Group Composition: Student "Carl" found success only when paired with peers of similar coding knowledge, suggesting SCL is facilitated by cognitive and material parity.
• Accessibility and Anxiety: Student "Alice" (who has an anxiety disorder) avoided traditional peer collaboration, opting to work alone and consult Teaching Assistants. She recognized that while group work works for others, it does not work for her. This highlights that rigid definitions of "social" learning can exclude students.
• Coding Etiquette: Student "Derek" identified the "etiquette" of commenting and formatting code as a social norm. This suggests that SCL includes asynchronous social practices (documentation) and not just synchronous collaboration.

5. Significance and Implications

• Curriculum Redesign: The findings are driving a redesign of the Oregon State University curriculum (starting 2025-2026) to consolidate the three 1-credit labs into a single 3-credit course and introduce a sophomore-level physical computing course.
• Explicit Instruction on SCL: The authors argue that Social Computational Literacy is often neglected. Future curricula must explicitly target SCL development, moving beyond just "group work" to include diverse modes of collaboration and communication (e.g., shared resources, documentation).
• Balancing Structure and Agency: The study emphasizes the need for learning environments that provide structure for collaboration while allowing flexibility for students with different learning needs (e.g., those with anxiety) to engage socially in ways that are accessible to them.
• Theoretical Advancement: The paper demonstrates that PCL is a robust framework for identifying specific learning tensions. It encourages other researchers to apply this framework to other contexts to further refine theoretical models of computational physics education.

In conclusion, the paper posits that effective computational physics education requires recognizing the inherent trade-offs students make between coding skills, physics understanding, and social interaction, and designing curricula that explicitly support the development of all three dimensions of literacy.