Table of Contents

General 13 min read

PISA Learning in the Digital World: What It Measures Before 2027

PISA learning in the digital world results are embargoed to 2027. Here is what the self-regulated learning tasks measure and how to build the skill at home now.

PISA learning in the digital world is the OECD’s attempt to measure something no international test has tried before: not what a 15-year-old knows, but whether they can teach themselves something new using digital tools while nobody helps them. The definition in the framework is precise: “students’ capacity to engage in an iterative and self-regulated process of knowledge building and problem solving using computational tools and practices.”

The computational problem solving scores from that assessment came out on September 8, 2026, with an OECD average of 500 and the US at 513 per its OECD country note. But the interesting half is still coming. Volume III, Effective Learners in the Digital World, will analyze how students actually managed themselves during those tasks, using process data from their clicks and keystrokes. The OECD lists it as forthcoming.

Key Takeaways

Learning in the Digital World (LDW) measures self-regulated learning with computational tools, not digital literacy or coding knowledge.
Students got two extended units with tutorials, feedback, and examples available, so the design rewards learning during the test.
The computational problem solving scale is the scored part: OECD average 500, Macao (China) 572, Singapore 563, B-S-J-Z (China) 560, US 513.
Volume III will use “student response process data” to examine task engagement, strategy adaptation, and self-evaluation. That analysis is forthcoming.
The construct explicitly includes whether a student stays engaged “even when it becomes challenging or frustrating,” which is a behavior parents can practice at home.

What PISA learning in the digital world actually assesses

LDW proficiency is defined by the OECD as the capacity to build knowledge and solve problems through an iterative, self-regulated process using computational tools. Unpack that and there are three separate things being measured at once.

Computational practice. Can the student use a modeling or block-based programming tool to build something, run an experiment, analyze the result, and debug the artifact?

Iteration. Does the student try, check, and adjust? Or do they commit to one approach and grind?

Self-regulation. Do they maintain motivation, monitor their own progress, adapt strategy, and evaluate their work? The PISA 2025 framework treats this as central rather than incidental: “A student’s ability to succeed in the digital world is greatly influenced by their self-regulated learning competencies.”

That last piece is what makes this different from a coding test. A student who knows Python cold but cannot read a tutorial, form a hypothesis, or notice that their approach is failing will not do well here. Twenty-two countries also participated in an optional English foreign-language assessment in the same cycle, with those results scheduled separately.

Task type versus the skill it targets

The OECD has not released the full item bank, and released items appear in Annex C of Volume I. Based on the framework and the proficiency descriptors, here is how the task families map to skills. The right column is what you can practice without waiting for Volume III.

Task type in LDW	Skill it targets	Home practice equivalent
Read tutorial, then modify a simple program	Extracting procedure from documentation	Follow a written recipe with no verbal help; change one ingredient
Adjust one input in a model, observe output	Controlled experimentation	One-variable-at-a-time testing on anything (paper airplane, plant watering)
Build a program that repeats an action	Recognizing when to loop	”Say that instruction in fewer words” games
Run a systematic experiment inside a model	Inferring relationships from data	Keeping a results notebook with what changed and what happened
Debug a program with a fault	Hypothesis-driven diagnosis	Fix something broken by predicting the cause first, then testing
Combine commands into a functional program	Composition and planning	Plan out loud before starting a build
Optimize a working program	Removing redundancy	”Can you do that with fewer steps?”
Persist through a frustrating stage	Motivation regulation	Waiting 90 seconds before helping

That last row is the one I would bet on mattering most, and it is the one no curriculum sells.

Why the results are staggered, and what that means

In the United States the sample is run by NCES. PISA 2025 is releasing in stages. Volume I, with science, reading, mathematics, and computational problem solving, came out September 8, 2026. Volume II covers the English foreign-language assessment for the 22 participating systems. Volume III, the full self-regulated learning analysis from the LDW process data, is listed as forthcoming; the OECD’s publication schedule points to spring 2027 for both, with the foreign-language volume ahead of the digital-world one.

There is a practical reason the staggering matters to a parent. The scored part of LDW, computational problem solving, tells you where students landed. The forthcoming part tells you how they got there: which strategies worked, how students managed frustration, whether self-evaluation predicted success. That is the actionable half.

Volume I gives one preview of what to expect. In computational problem solving, 26% of OECD students reach Level 5 or 6, 21% reach Level 4, 17% reach Level 3, 13% Level 2, 17% Level 1, and 5% below Level 1. The OECD notes there is no formal baseline threshold but flags Level 3 as “a useful benchmark, as it marks the point at which students begin to demonstrate sufficiently solid knowledge and skills to progress more rapidly.” Full detail in our guide to computational problem solving and how to build it at home.

How to Teach Your Kid About Self-Directed Digital Learning

Everything below trains one of the three components: computational practice, iteration, or self-regulation. None of it requires special software.

Ages 5–8: Instructions Only

Pick something your child can physically do: building a block tower to a picture, setting a table, folding a paper boat. Give them written or picture instructions and then stop talking. No hints, no corrections, until they finish or ask a specific question.

The goal is not the boat. It is the experience of getting information from a source rather than from a person. That is the exact muscle the LDW tutorials test. When they ask “is this right?”, answer with “what would tell you?”

Ages 9–12: The Change Log

Give your child something with adjustable parts: a paper airplane, a marble run, a plant, a simple Scratch project. The rule is that every change gets written down in two columns: what I changed, what happened.

After five entries, ask them what the data say. Most kids have never generated their own data set, and the act of reading it back is what PISA calls inferring “simple relationships between different factors in a model,” which sits at Level 3 on the scale.

Ages 13+: Learn a Tool in 45 Minutes

Pick a tool your teen has never used: a spreadsheet’s pivot tables, Tinkercad, a Git tutorial, a synth in GarageBand. Give them 45 minutes, the official documentation, and one concrete goal (“make a chart that shows which month you spent the most”).

Two rules: they cannot ask you, and they have to narrate one thing that went wrong and how they figured it out. That narration is self-evaluation, the component Volume III will measure through process data.

The question to ask: “What did you try that didn’t work, and how did you know it wasn’t working?”

The uncomfortable part of the construct

There is a tension in measuring self-regulated learning on a standardized test, and it is worth naming.

A test score conflates two different failures. A student might score low because they lack the computational skill, or because they would not read the tutorial. The OECD’s own description of below-Level-1 performance hints at this: those students are “largely exploring the software interface in a rudimentary way, rather than working towards a specific goal.” That is a description of behavior, not of missing knowledge.

For a parent, that distinction is the whole ballgame. A skills gap needs practice. A persistence gap needs a different intervention entirely, usually a reason to care and permission to fail. The PISA 2025 engagement data suggest persistence is the one eroding: only 52% of OECD students say they finish tasks that have become boring, and perseverance items fell in 32 to 40 systems since 2022. We cover that data in our piece on PISA 2025 curiosity and perseverance.

What to actually do at home

Default to documentation before conversation

When your child hits something unfamiliar, the sequence should be: look it up, try the smallest version, then ask. Say it out loud enough times that it becomes automatic.

Let the first failure stand

The construct includes maintaining engagement through frustration. If you intervene at the first error, you remove the thing being measured. Ninety seconds is a reasonable default.

Ask for the process, not the product

“Show me what you made” gets you a screenshot. “Tell me what you tried first” gets you the self-regulation narrative that Volume III is built to analyze.

Pick tools with visible failure

Scratch, spreadsheets, and CAD tools all show you when something is wrong. Apps that auto-correct or auto-complete hide the feedback loop, which is where the learning lives.

What not to do

Do not treat this as screen time to be minimized or maximized. The PISA finding on device use is consistent and specific: “limited or moderate use of digital devices for learning at school is often associated with better outcomes,” while leisure use during school “reverses sharply.” The relevant variable is what the screen is for, not how many minutes it is on.

What to Watch For Over the Next 3 Months

Week 4: Does your child look something up before asking you? Count it for a week; most parents are surprised by the ratio.
Month 2 red flags: Abandoning a task within a minute of the first error, or clicking randomly instead of working toward a goal. The second is the OECD’s description of below-Level-1 behavior.
Month 3 self-check: Ask your child to teach you something they learned from documentation this month. If they can, the self-directed loop is working.

Frequently Asked Questions

When do the Learning in the Digital World results come out?

The scored computational problem solving results were published in Volume I on September 8, 2026. The full self-regulated learning analysis appears in Volume III, Effective Learners in the Digital World, which the OECD lists as forthcoming. The English foreign-language results appear in Volume II.

Is this a coding test?

No. Programming is one tool available in the tasks, alongside computer modeling, but the assessment measures whether students can build knowledge and solve problems iteratively while regulating their own learning. A student can know a programming language and still score at Level 1 if they cannot plan, test, or persist.

What does “self-regulated learning” mean in practice?

Four behaviors: maintaining motivation and task engagement, monitoring your own progress, adapting strategy when something is not working, and evaluating your own performance. Volume III will analyze these using process data from how students actually interacted with the tasks.

Which countries did best on computational problem solving?

Macao (China) 572, Singapore 563, B-S-J-Z (China) 560, Japan 557, Chinese Taipei 551, Hong Kong (China) 545. Among OECD countries, Australia, Estonia, Japan, Korea, and New Zealand all scored above 530. The US scored 513 against an OECD average of 500.

How is this different from digital literacy?

Digital literacy usually means knowing how to use tools and evaluate online information. LDW measures whether a student can use computational tools to learn something new on their own, including building and debugging artifacts. It is closer to “can you teach yourself with a computer” than “can you use a computer.”

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

Sources

OECD. (2026). “PISA 2025 Results (Volume I): Future-Ready Students,” Chapter 2 and Reader’s Guide on forthcoming volumes. OECD Publishing. https://www.oecd.org/en/publications/pisa-2025-results-volume-i_73451bc5-en.html
OECD. (2026). “PISA 2025 Assessment and Analytical Framework.” OECD Publishing. https://doi.org/10.1787/86c36975-en
OECD. (2026). “PISA 2025: Students’ reading and mathematics performance declined sharply across the OECD.” OECD press release, September 8, 2026. https://www.oecd.org/en/about/news/press-releases/2026/09/pisa-2025-students-reading-and-mathematics-performance-declined-sharply-across-the-oecd.html
OECD. (2026). “PISA 2025 Results (Volume I): Country Note, United States.” OECD Publishing. https://www.oecd.org/en/publications/pisa-2025-results-volume-i-country-notes_2d4ff9ea-en/united-states_0c8cbc7c-en.html
National Center for Education Statistics. (2026). “Program for International Student Assessment (PISA).” NCES. https://nces.ed.gov/surveys/pisa/
OECD. (2021). “21st-Century Readers: Developing Literacy Skills in a Digital World.” PISA, OECD Publishing. https://doi.org/10.1787/a83d84cb-en

Written by Ricky Flores

Founder of HiWave Makers and electrical engineer with 15+ years working on projects with Apple, Samsung, Texas Instruments, and other Fortune 500 companies. He writes about how kids learn to build, think, and create in a tech-driven world.