My own experience says that writing about code has made me a better programmer, but only in specific, identifiable ways. No study I could find tests that exact claim. The closest experimental evidence points to a different lever: writing code itself, ideally with immediate feedback, is what most reliably builds coding skill. Writing about code works best as a companion to that practice, not a substitute for it.
What I mean by “writing about code”
For me the term covers four kinds of work: tutorials, reference documentation, essays like this one, and the comments I leave in code. Explaining a concept to a colleague in a message counts too. These are different genres with different demands. A tutorial has to run. A comment has to stay true as the code changes. An essay has to hold an argument together. What they share is one move: turning something I understand into an order another person can follow.
As an Amazon Associate I earn from qualifying purchases.
The mechanism I have noticed
The effect I notice is not that writing makes my typing faster. It is that drafting an explanation exposes gaps in my understanding. Three patterns recur.
- Hidden assumptions surface. Writing “this function assumes the input is sorted” makes me ask whether it is, and whether anything enforces that. Often nothing does.
- Decomposition gets sharper. A paragraph that wanders across three ideas usually signals a function that is doing three jobs. Rewriting the paragraph tends to suggest splitting the function.
- Examples force working code. A snippet I have not run is one I may get wrong. Once I started executing every example before publishing it, I found errors I would otherwise have shipped.
Consider a small illustration. Say a function reads numbers from a file and returns their average. Before I tried to explain it, I would have said it “handles input.” Explaining it step by step forces the questions: what happens with an empty file, a header row, or a line with a missing value? Each answer becomes either a guard clause or a test. The explanation did not write the code, but it located the places where the code was silently undefined.
This is a plausible account of a reflective practice, not a measured one. The mechanism is that drafting forces ideas into a linear sequence, and linear sequences expose missing steps. Learning research on writing and reflection is compatible with this idea, though none of the studies below tests blog-style writing about programs.
What the studies show, and what they do not
Several studies sit near the title’s claim. None measures it directly. The table summarizes what each one examined and where its evidence stops.
| Study | What it examined | What it supports | What it cannot show |
|---|---|---|---|
| Gold, Tjaden, and Carvalho, 2026 (preregistered experiment, N = 250 participants; preprint abstract) | Practice-based programming instruction compared with video, measured on a novel code-generation test | Practice beat video. Participants who wrote code with immediate feedback performed best among the compared approaches. | Anything about writing prose articles. The result concerns producing code, and the details available are at abstract level. |
| Writing-to-learn case study, 2019 | Short, low-stakes writing during programming, read through the comments novice programmers left | Short writing can make reflection, analysis, synthesis, and metacognition visible while coding. | A causal estimate of skill gains. The focus was how comments reveal thinking. |
| Python study, 2009 (students) | Relationships among writing code, tracing code, and explaining code | Students who did reasonably well at writing code usually also could trace and explain code. | That explaining produces better coding. The finding is an association. |
| Cal Poly thesis, 2018 (CS students) | Whether programming skills transfer to academic prose | Students reported more confidence writing organized papers, and their paragraphs were more often focused on a single topic. | The title’s direction. This evidence runs from programming habits to prose, not from prose to code. |
Why the direction matters
The title claims that writing about code improves writing code. The evidence most often tests the reverse, or tests producing code against watching it. Keeping four distinctions apart avoids most of the confusion:
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →- Generating code versus watching or tracing someone else’s code.
- Writing prose about code versus writing comments or reflections while coding.
- Working with immediate feedback versus working without it.
- A correlation between two skills versus an experiment that changes one of them.
Treating these as one intervention is the common error. A claim that “explaining code helps you learn to code” sounds equivalent to “producing code with feedback helps you learn to code,” but only the second has direct experimental support.
Rank #3
Learner differences matter more than method alone
A 2020 University of Washington report on novice Python learners, led by Chantel Prat and colleagues, found that language aptitude, fluid reasoning, working memory, and resting-state brain activity predicted learning better than numeracy did. Numeracy explained an average of 2% of differences in outcomes in that study. Prat also said that the combined measures explained more than 70% of the variability in how quickly learners picked up Python. That figure is her characterization of the study’s own measures, not a general law.
Prat put the implication this way: “Many barriers to programming, from prerequisite courses to stereotypes of what a good programmer looks like, are centered around the idea that programming relies heavily on math abilities, and that idea is not born out in our data.” That point concerns perceived prerequisites. It says nothing about whether writing helps.
For a writer, this means the same practice may help some learners more than others, and a single essay habit will not close every gap.
How to test the claim on your own work
If you want to know whether writing helps you, rather than assume it, a simple comparison will tell you more than any study summary:
Best Value
- Pick one concept you already use daily and write a short explanation of it for a reader who knows the language but not your codebase.
- Before writing, list the edge cases and assumptions you believe the code handles. Keep the list.
- After writing, compare the list with the draft. Note each assumption that the draft made you state or question.
- Run every code example in the draft in a clean environment, and record each failure.
- For a comparison group, spend the same amount of time on a coding task with no article attached, and count the bugs you find in review.
Over a month of this, you will have a small, honest dataset about your own habits. It is still one person’s data, but it tests the claim directly, which the studies above do not.
What the evidence supports
The defensible version of the title is narrower than the title itself. Writing about code has, in my experience, exposed unstated assumptions, sharpened how I split problems, and made my examples run. Those are observations from one programmer. The controlled evidence we have says that producing code, ideally with feedback, is the stronger lever, and that programming habits can carry over into how people organize prose. The reverse path, from prose to better code, has not been tested directly.
If you write about code to get better at it, treat the writing as a way of checking your understanding, and keep most of your hours for the code itself.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
