Your kid announces they want to learn coding, and your first instinct might be to picture a screen full of semicolons and someone old enough to touch-type. Relax. Almost none of that is what early coding actually is.
Ordering steps correctly, running them, watching what happens, and fixing whatever went sideways: that is the entire deal at this stage. Typing shows up much later, once a child is old enough to spell reliably and patient enough to hunt down a missing bracket. Before that, coding is closer to giving directions to a very literal friend.
The Computer Science Teachers Association's own K-2 standards spell this out. At that level, a child is expected to "model daily processes by creating and following algorithms" and to "develop programs with sequences and simple loops," using tools built for exactly that, not a keyboard [1]. Debugging counts too. The same standards ask a five-to-seven-year-old to find and fix an error in a short sequence, which is a different skill from writing one correctly the first time.
This is not just standards writers being optimistic about small children. A 2013 case study by Georgios Fessakis, Evaggelia Gouli, and Elisavet Mavroudi put a group of five and six year olds in front of a Logo-style programming environment with draggable command tiles instead of a keyboard. The children worked through a series of programming problems, and the researchers observed real openings for problem-solving and early math thinking along the way, not just enthusiasm for a fun activity [2]. The kids were not memorizing syntax. They were figuring out which order of moves got their character to the goal, then noticing when it did not and trying something else.
A larger study backs this up with robots instead of screens. Marina Umaschi Bers and colleagues ran the TangibleK Robotics Program in three kindergarten classrooms, where children built and programmed robots using wooden blocks that snapped together to represent commands. No typing anywhere in sight. Across the study, kindergartners learned real programming concepts: choosing the right instruction, putting instructions in the correct sequence, and troubleshooting when a robot did not do what they expected [3]. Some concepts landed more easily than others, which is exactly what you would expect from five year olds tackling something genuinely new, but the pattern held. Sequencing and debugging are learnable well before a child can spell "function."
That last part, the debugging, is doing more work than it looks like. A program that runs wrong on the first try is doing exactly what it is supposed to do. Getting it wrong is the whole point of the exercise. The child has to look at what actually happened, compare it to what they wanted, and change one thing before trying again. That loop of predict, run, compare, adjust is the same productive frustration we build learning games around in general, just applied to a sequence of instructions instead of a puzzle piece.
It also explains why early coding sits closer to reasoning than to memorization. A child who gets a sequence right by copying someone else has not learned much. A child who gets it wrong, notices the character turned left instead of right, and fixes the order themselves has built something that sticks. This is the same territory we have written about before: the games worth building ask a child to work out the answer, not recite one.
We think about this a lot in our own coding game, where a child arranges a short set of instructions, presses go, and watches a character carry them out exactly as written. When the character walks off the edge or opens the wrong door, nothing tells the child what to fix. They watch it happen, figure out which instruction was wrong, move it, and run it again. The whole game is cards to reorder and a button to press.
If your child wants to try this at home before any screen is involved, you already have the tools. Write out the steps for a simple task, like making a sandwich, on separate cards, and have them put the cards in order, then act it out exactly as written, including the parts that go wrong because a step landed in the wrong place. That is a program. Fixing the card order when the sandwich comes out backwards is debugging.
What the research says is that young children can learn to sequence, predict, and debug well before they can type. That is where early coding actually starts, and it is where we chose to start too.
References
- Computer Science Teachers Association. (2017). CSTA K-12 Computer Science Standards, Revised 2017. https://csteachers.org/wp-content/uploads/2025/03/csta-k-12-computer-science-standards-revised.pdf
- Fessakis, G., Gouli, E., & Mavroudi, E. (2013). Problem solving by 5-6 years old kindergarten children in a computer programming environment: A case study. Computers & Education, 63, 87-97. https://doi.org/10.1016/j.compedu.2012.11.016
- Bers, M. U., Flannery, L., Kazakoff, E. R., & Sullivan, A. (2014). Computational thinking and tinkering: Exploration of an early childhood robotics curriculum. Computers & Education, 72, 145-157. https://doi.org/10.1016/j.compedu.2013.10.020