Computational modeling is not coding — a distinction that matters
One of the most persistent misunderstandings in K-12 computing education is the conflation of computational modeling with coding. They are not the same thing, and treating them as interchangeable leads to professional development that misses the point.
Coding is a skill — the ability to write instructions in a formal language that a computer can execute. Computational modeling is a practice — the ability to build, test, and refine simplified representations of complex systems in order to understand how those systems behave. Coding can be a tool within computational modeling, but the modeling is the thinking. The code is just the medium.
This distinction matters because it changes who computational modeling is for. Coding, as typically taught, appeals to a narrow band of students and teachers who are already comfortable with programming syntax. Computational modeling, framed correctly, is relevant to every science teacher, every math teacher, and every teacher whose subject involves systems that change over time — which is most of them.
How we teach it
We have used NetLogo as the primary platform for computational modeling in our professional development workshops for nearly a decade. NetLogo is a multi-agent simulator designed for education — originally developed at Northwestern — that allows users to create models of complex phenomena (ant foraging, disease spread, traffic flow, predator-prey dynamics) by defining simple rules for individual agents and observing the emergent behavior of the system.
Our approach, documented in a peer-reviewed study published in Education Sciences, took 22 K-12 teachers through a multi-day immersion in which they first experienced computational modeling as learners — running existing NetLogo simulations, modifying parameters, observing results — and then designed their own classroom applications.
The published findings were instructive. Teachers adopted the tools and concepts successfully during the PD itself. The challenge was in transfer: classroom implementation tended to replicate the specific activities from the PD rather than adapt the underlying modeling approach to new contexts. That finding shaped everything we have done since. It told us that the PD itself needed to spend more time on the design logic — the “why” behind the model, not just the “how” — and that follow-up support during implementation was essential.
The full study is open access: Borowczak, M. & Burrows, A.C. (2019). Ants Go Marching — Integrating Computer Science into Teacher Professional Development with NetLogo. Education Sciences, 9(1), 66.
Why NetLogo works for non-CS teachers
Most PD platforms for computing assume participants have some prior programming experience — or at least a willingness to acquire it. NetLogo does not make that assumption. Its visual interface lets teachers see the model running in real time: dots moving across a landscape, populations growing and declining, resources being consumed and replenished. The feedback is immediate and visual, not textual.
For a biology teacher watching an ant colony simulation emerge from three simple rules, the moment of insight is not “I learned to code.” The moment of insight is “I can build a model of a system I teach, change the parameters, and show my students what happens.” That is a qualitatively different outcome from a coding workshop, and it is the outcome we design for.
We also chose NetLogo because the models are transparent — teachers can inspect and modify the underlying rules without deep programming knowledge. The Logo-based language is readable, and the interface separates the model logic from the simulation display. Teachers do not need to be software engineers to understand what the model is doing or to customize it for their content area.
Where it fits in the curriculum
Computational modeling maps directly to existing science standards — particularly the NGSS practice of “developing and using models.” When a teacher builds a NetLogo simulation of energy transfer or population dynamics, she is not teaching computer science as an add-on. She is teaching her existing science content through a more powerful representation.
This is the integration model that we advocate across all of our work: computing as a lens for existing content, not computing as a separate subject crowding an already full schedule. The 411+ lesson plans on CxEdHub reflect this philosophy, with data collection and computational modeling activities embedded in science, math, social studies, and language arts lessons.
What this means for PD design
If your district is considering computational modeling as part of its STEM strategy, the PD design matters more than the platform selection. Teachers need time to experience modeling as learners before they design for their students. They need follow-up during implementation. And they need explicit attention to transfer — how the modeling approach adapts to content areas beyond the specific examples used in the workshop.
We have been refining that PD design for a decade. The research tells us what works. The question is whether the PD gives teachers enough time to practice it.