Development and illustration of a framework for computational thinking practices in introductory physics

Physics classes with computation integrated into the curriculum are a fitting setting for investigating computational thinking. In this paper, we present a framework for exploring this topic in introductory physics courses. The framework, which was developed by reviewing relevant literature and acquiring video data from high school classrooms, comprises 14 practices that students could engage in when working with Glowscript VPython activities. For every practice, we provide in-class video data to exemplify the practice. In doing this work, we hope to provide ways for teachers to assess their students’ development of computational thinking and give physics education researchers a foundation to study the topic in greater depth.

Figure 1

Timeline of the methodological steps that were taken to develop the CT framework proposed in this report. (a) First, a literature review was conducted to identify works pertaining to computational thinking practices. (b) An initial list of practices was proposed with 17 uncategorized, standalone practices. (c) The initial practices were used to analyze in-class video data and find examples of the practices from students working with computationally integrated physics activities. (d) An initial framework was constructed and shared with external researchers to provide feedback on how the framework could be improved in terms of its relevance and precision of definitions. (e) Classroom videos were analyzed again to find examples that served as evidence for the updated practices. (f) The framework was refined based on findings from in-class data and feedback from researchers. (g) The framework was critiqued by a PER group to discuss its relevance to physics, as well as any confusing language in the framework’s definitions. (h) The finalized framework comprised 14 practices in 6 categories.

Figure 2

Outline of the practices in our computational thinking framework. The framework has 14 practices comprising 6 distinct categories. The categories of extracting computational insight, building computational models, and data practices all contain several practices within them. The categories of debugging, demonstrating dispositions towards computation, and working in groups on computational models exist as standalone practices.

Figure 3

(a) A group of three students (pseudonyms A, B, and C) discuss their computational model with the teacher (T). The students’ model displays (b) first a blue box moving to the right, (c) then the blue box stopping on the right side of the platform and a pink box appearing after the blue box completes its motion, and (d) lastly the pink box continuing its motion infinitely to the left of the platform. Yellow arrows in (b), (c), and (d) have been added to illustrate the direction of motion for each crate at different instances.

Figure 4

Computational model of a boat crossing a river. The large yellow arrow represents the boat, the smaller yellow arrows represent a motion map of the boat’s velocity at different time instances, and the dark brown rectangle represents a log floating in the river (i.e., only affected by the current). Panel (a) displays the boat incorrectly continuing after it reaches the opposite shoreline, and panel (b) displays the boat correctly stopping when it reaches the opposite shoreline.

Figure 5

Computational model of two cubes colliding (a) before and (b) after an elastic collision. The blue cube represents a semi truck and the red cube represents a red Ferrari, with arrows representing each object’s velocity.

Figure 6

(a) Incorrect graph and (b) correct graph of energies in a vertical spring-mass system simulation.

Figure 7

Images of two contrasting groups working on the colliding crates activity from Liam’s classroom. Panel (a) shows students A and C being able to easily see the computer, while student B has to lean in to see the device. Panel (b) shows students A, B, and C all with a line of sight to the computer so that they can easily collaborate.

Figure 8

Image of a group of students experiencing success while working through a computational activity.

Figure 9

Outputs displayed in Michael’s solutions for computational models of (a) projectile, (b) river crossing, and (c) spring energy problems.

Figure 10

Outputs displayed in Liam’s solutions for computational models of (a) colliding crates, (b) momentum conservation (top panel for before the collision and bottom panel for after the collision), and (c) block on a ramp activities.