Every now and then, I think of how much schooling has changed since I was a kid. This is true across all disciplines, and certainly when it comes to science-related subjects. Honestly, as I write this, only a single science memory from before ninth grade comes to mind: learning about archaeological discoveries from textbook and film. I must admit, though, the fact that I remember this is most likely because my classmate Rusty ran around the wooded part of our schoolyard during recess for the better part of a week pretending he was Dr. Richard Leakey digging up Homo sapiens skulls, not because of any classroom lesson. Clearly, Rusty found the textbook reading to be inspiring; I doubt it worked out that way for most of my classmates. Looking back, I realize that I went into science because I liked solving word problems, not because I wanted to be a scientist or, frankly, really understood what the job of a scientist was like. When I was a graduate student and even an early-career college professor, I struggled with coming up with my own questions to investigate because my prior experience honing this skill was so limited. I think of how much better prepared I would have been if I learned science as it is being taught now. 

The same is true when I think about computer science. My experiences with anything related to this field were learning programming languages in elective courses in high school and college. We learned about structure and syntax and routines, but never wrote code to do anything particularly meaningful—certainly not anything that resembled what one might use programming for in the “real world.” While, of course, I’m sure we were developing computational thinking skills to some degree, we weren’t aware that we were doing so, and those skills weren’t what we were graded on. I didn’t understand at that time why those skills were important, how they could be used, or what a computer scientist really does.  

While learning programming languages certainly has its place, now there’s wider-spread recognition that there is so much more that is important to learn. Today’s K-12 computer science learning goals are firmly grounded in computational thinking—skills that are foundational, broadly applicable, and relevant. There’s consensus that every young person should have opportunities to develop and master computational thinking skills like algorithmic thinking, pattern recognition, decomposition, and abstraction. Kids can begin exploring these ideas even before they know how to read. And opportunities to do so are often “unplugged”—no computer necessary.

It hasn’t always been easy to connect the field of computer science with the natural sciences like biology, chemistry, and physics. However, once we consider computational thinking as an essential foundation, it is easy to see synergies with the NGSS science and engineering practices. For example, abstraction helps you determine which variables must be considered and which can be ignored, an important skill for modeling. Pattern recognition is critical for making claims based on evidence. Algorithmic thinking comes into play when planning investigations, and decomposition when designing solutions to engineering problems. By definition, computational thinking is a process that humans use to solve problems. It allows us to reformulate a seemingly difficult problem into one we can figure out how to solve. Advocates argue that teaching young people to think computationally is essential for closing the gap between education and the workplace. Seems to me that, like the practices of science and engineering, computational thinking skills are also skills for life.

Beth Murphy, PhD (bmurphy@nsta.org), is field editor for Connected Science Learning and an independent STEM education consultant with expertise in fostering collaboration between organizations and schools, providing professional learning experiences for educators, and implementing program evaluation that supports practitioners to do their best work. 

citation: Murphy, B. 2023. Computational thinking and why it matters. Connected Science Learning 5 (1).