How to Strengthen Computing Instruction in the Physics Classroom

Better computational training is needed to prepare physics students for the workforce, says researcher Danny Caballero.

By Rachel Crowell | October 13, 2023

Modern physics research and computing are so deeply intertwined that many students' future careers will depend on strong computational skills — from knowing programming languages and developing data visualizations, to creating computational mathematical models and "mining [datasets] for patterns and structure," says Danny Caballero, a physics education researcher at Michigan State University.

Those skills need to be deep and broad enough to serve students regardless of whether they plan to work in academia, at government laboratories, or in private industry. "One of our recent graduates is modeling WiFi attenuation for a company," Caballero says. "Another is employed by a cereal company, drawing meaning from their vast sales data.”

Yet many students are receiving minimal, if any, dedicated computing instruction in their physics courses. "Something like 50% of [physics] departments report that they're teaching computing," Caballero says. Based on discussions he’s had with faculty across the country, he thinks this is an overshoot.

Caballero and five team members received the 2023 Excellence in Physics Education Award for their work for the Partnership for Integration of Computation into Undergraduate Physics (PICUP), a community of physics educators dedicated to adding computation into instruction. In 2022, Caballero was elected as an APS Fellow in recognition of his research.

Even in physics departments where students are receiving computing instruction, it may not be optimized for their success as future physicists. Caballero hopes to change that: Much of his work focuses on identifying ways that faculty can "organize the curriculum and the instruction so that people are learning computational things in physics," he says.

For instance, he is examining how students effectively learn computing in groups, and he’s working with collaborators to tackle qualitative questions about how students’ see themselves as learners in physics. “What are their feelings towards this?” he says. “How do these environments that we create for them cause or interact with those outcomes?"

Taken together, those questions add up to one of Caballero's big ones: "How can we change the way that we do things holistically" to help students learn and build "a stronger positive sentiment towards computing in physics"?

One key takeaway from Caballero's research is that the best computing instruction in physics classrooms is consistent and intentional, not haphazard — no matter the level of instruction.

"If you're going to integrate computing into the classroom, [then] it needs to appear in every part of the class," Caballero says. Simply assigning projects that incorporate computing won’t be effective if the instructor doesn't systematically equip students through classroom instruction and experiences to tackle those assignments. "If your course is structured in a way that has a presentation of material to students, time where they're practicing with that material, assessments where they are being told how well they do on that material — all of those things have to have computing," Caballero says.

However, constraints on time, technology, and budgets may limit what professors can add to their courses — especially if a department is starting from scratch, with no previous computing instruction.

In those situations, “starting small and growing the project over time is better than trying to do something large that is complicated and might not net you anything," Caballero says. For instance, on the PICUP website, instructors can access peer-reviewed exercise sets, like "Binary Stars with Equivalent One Body Problem," Snowboard Jumping and Newton’s Second Law," and more. Caballero recommends that instructors try one of them, or pick out two or three to use in one semester. "That's starting small," he says.

Faculty can expect that projects will evolve over time to meet goals, expand, or morph in response to changing technology and students' needs. For instance, early in the COVID-19 pandemic, the shift to online instruction “was very, very challenging,” Caballero says. “There was a lot of adaptation of the group-based dynamics, from classes where students were working together in classrooms with a computer and so forth, to students in virtual groups with a screen being shared."

Maximizing available resources can also make classrooms more equitable. "Embrace the use of open-source software," he says. "Students can continue to use it forever," and these free tools often can "do many of the cool things that some of the packages we pay for can do.”

In some instances, not every student will have their own computer, or the university won't have enough for everyone. (That used to be the case at Caballero's university before a laptop policy was implemented.) It's still possible for students to be taught equitably. In the past, Caballero's department purchased enough laptops to provide one for every four students. "So that was the group size," he says. Students who had their own computers could still bring them, but this policy ensured that students who didn't own one still had access to computers.

While keeping up with changes in computing technology can be challenging, it's also key to equipping physics students with the skills they need for their future careers. "Preparation with computing is just part of being a participant in physics," Caballero says. "If we aren't trying to keep up with the world that our students are encountering, we're not providing the best possible education that we can."

Rachel Crowell is a math and science writer based in Iowa.

To future-proof your physics classroom with computational training, get free resources from PICUP’s website and the APS Effective Practices for Physics Programs (EP3) guide.