Japanese Learning Answer Dataset — August 2026
In one line: aggregate statistics from 56,613 anonymous answer events on Nihongo to Japan, contributed by 1,053 anonymous learner identifiers across 11,987 questions.
⚠️ This is an August 2026 snapshot, not a study of the whole of 2026. The data covers 2–13 August 2026, twelve days in total (the aggregation pipeline went live on 2 August; the first day is partial). No longitudinal trend can be read from it. Version 1.1.0, snapshot frozen 2026-08-14.
On "learner identifiers": these are anonymous identifiers generated at random in the browser and cannot be merged across devices. One person using a phone and a laptop produces two identifiers, so 1,053 is not a verified count of distinct people — it is an upper bound. This page says "learner identifiers" throughout rather than "people".
What this is not. This is not a representative survey of Japanese learners in Taiwan or anywhere else. The sample consists of people who chose to practise on this site — a self-selected, observational sample. Every figure describes what happened in this set of answers, and cannot be extrapolated to any population. See Limitations.
1. Five main findings
1. Grammar patterns are the only axis below 60%, 38 points behind vocabulary
Among quiz-context first attempts (see Methodology for the set definition), vocabulary items were answered correctly 92.1% of the time (n=22,869, 627 learner identifiers, 95% CI 91.8%–92.5%), while grammar-pattern items reached only 53.7% (n=8,678, 532 identifiers, 95% CI 52.6%–54.7%).
The confidence intervals are far apart, so this is not sampling noise. The data shows a large gap between the two axes; it says nothing about why.
2. Grammar-pattern "accuracy" runs from 53.7% to 74.1% depending on which answers are included
Under different analysis-set definitions, observed accuracy on the grammar-pattern axis ranges from 53.7% (quiz-context first attempts, n=8,678) to 74.1% (all answer events, n=24,293). Because the events included and the learner-identifier composition both differ, the whole difference cannot be read as the product of a single counting rule.
Step by step:
| Step | n | Raw accuracy | Change vs previous |
|---|---|---|---|
| All answer events | 24,293 | 74.1% | — |
| → keep only first attempts | 22,638 | 73.2% | -0.87 pp |
| → keep only quiz contexts (drop lesson/review) | 8,678 | 53.7% | -19.54 pp |
| (comparison) lesson-context first attempts | 13,937 | 85.4% | — |
Removing repeat attempts moves the figure by only 0.9 percentage points. Almost all the remainder comes from the difference between lesson-context and quiz-context answers.
Paired comparison on the same learner identifiers: 75 identifiers have at least 5 grammar-pattern first attempts in each context. They reach 86.6% in the lesson context (n=5,858) and 55.5% in the quiz context (n=2,055), a paired mean difference of +27.5 pp (95% CI 22.6–32.4; median +26.9). So the gap is not simply a matter of who is answering.
⚠️ A same-question comparison is not possible, so no conclusion can be drawn. Of 5,772 distinct grammar-pattern questions, only 6 appear in both contexts, and only 1 has at least 3 answers on each side. The two contexts draw on almost disjoint question sets, so the context effect cannot be separated from differences in question difficulty. The lesson-context questions may simply be easier.
What can honestly be said is this: when citing an accuracy figure from a learning platform, always ask which analysis set it came from. All four sets are in the downloadable data.
3. N3 and N2 have the lowest raw accuracy — but this data cannot separate them
Quiz-context first-attempt raw accuracy by level: N5 90.2%, N4 73.4%, N3 67.8%, N2 69.4%, N1 76.0%.
The intervals for N3 (95% CI 66.6%–69.0%) and N2 (95% CI 67.6%–71.1%) overlap, so it is not correct to say "N3 is harder than N2". What can be said is that in this sample, N2 and N3 both sit below N4, N1 and N5 on raw accuracy.
⚠️ More importantly, each level draws on a completely different question set. This ordering is not a difficulty comparison and is not evidence about which JLPT level is objectively harder.
4. The spread within particles is wider than the spread between levels
Quiz-context first-attempt accuracy on individual particles: が 60.9% (n=358, CI 55.8%–65.8%), に 62.4% (n=370), で 63.3% (n=330), versus を 78.7% (n=385, CI 74.3%–82.5%) and の 83.3% (n=120).
The intervals for が and を do not overlap; the gap is 18 percentage points.
5. One-vote-per-identifier and one-vote-per-answer differ by 7.04 points
Computed over exactly the same 372 learner identifiers, the same analysis set and the same 34,392 eligible events:
| Weighting | Accuracy |
|---|---|
| Event-weighted (one vote per answer) | 82.5% |
| Identifier-weighted (one vote per identifier) | 75.5% (median 76.2%) |
| WEIGHTING_EFFECT_PP | 7.04 pp |
Cohort: Learner identifiers with at least 20 answers in the primary analysis set. Both figures share the same numerator population and denominator, so the gap reflects the weighting choice alone and is not contaminated by differences in sample composition.
The cause is repeated measures: the single most active identifier accounts for 3.7% of all answers and the top ten for 19.4%. Any study using platform data hits this, so we publish both numbers.
2. Dataset overview
| Item | Value |
|---|---|
| Valid answer events | 56,613 |
| Anonymous learner identifiers (upper bound on people) | 1,053 |
| Distinct questions | 11,987 |
| Knowledge points | 39 |
| Temporal coverage | 2 to 13 August 2026 (12 days); the first day is partial |
| Snapshot cutoff | 2026-08-14T00:00:00Z |
| First-attempt events | 52,934 |
| Quiz-context first attempts (primary set) | 37,245 |
| Learner identifiers in the primary set | 688 |
| Lesson-context first attempts (comparison set) | 15,741 |
Answers by source: challenge 23,369 (41.3%), self_study 16,782 (29.6%), placement_test 11,797 (20.8%), special_training 4,157 (7.3%), review 508 (0.9%).
Distribution of the database evidence_class column (a raw table field, not the name of an analysis set in this report): A 37,245 (65.8%), B 17,048 (30.1%), C 2,320 (4.1%). The primary analysis set is exactly evidence_class = A.
3. Accuracy by learning axis
| Skill axis | n | Identifiers | Raw accuracy | 95% CI | Flag |
|---|---|---|---|---|---|
| Grammar patterns | 8,678 | 532 | 53.7% | 52.6%–54.7% | OK |
| Particles | 1,986 | 137 | 69.1% | 67.1%–71.1% | OK |
| Word forms / conjugation | 2,322 | 386 | 83.2% | 81.7%–84.7% | OK |
| Idiomatic expressions | 1,341 | 261 | 86.4% | 84.5%–88.2% | OK |
| Vocabulary | 22,869 | 627 | 92.1% | 91.8%–92.5% | OK |
The following axes fall below the minimum cell size (n<30 or fewer than 5 learner identifiers), so no accuracy figure is published — only the event count: Mixed / all-round (n=27, 8 identifiers), Reading comprehension (n=22, 4 identifiers).
4. Accuracy by JLPT level
⚠️ This is the section most easily misread. Each level draws on a completely different question set, and no IRT difficulty equating was performed. The figures below are the proportion of these particular questions answered correctly — not a measure of which level is harder.
| JLPT level | n | Identifiers | Raw accuracy | 95% CI |
|---|---|---|---|---|
| N5 | 19,395 | 380 | 90.2% | 89.8%–90.6% |
| N4 | 6,099 | 313 | 73.4% | 72.3%–74.5% |
| N3 | 5,985 | 432 | 67.8% | 66.6%–69.0% |
| N2 | 2,726 | 212 | 69.4% | 67.6%–71.1% |
| N1 | 3,040 | 139 | 76.0% | 74.4%–77.5% |
The same levels across all three analysis sets:
| Level | Quiz-context first (primary) | Lesson-context first | First attempts (all) | All events |
|---|---|---|---|---|
| N5 | 90.2% (n=19,395) | 83.7% (n=11,154) | 87.8% (n=30,495) | 88.2% (n=33,286) |
| N4 | 73.4% (n=6,099) | 83.1% (n=2,694) | 76.5% (n=8,758) | 76.6% (n=9,262) |
| N3 | 67.8% (n=5,985) | 82.1% (n=1,507) | 70.8% (n=7,522) | 71.1% (n=7,727) |
| N2 | 69.4% (n=2,726) | 79.2% (n=260) | 70.2% (n=2,985) | 70.5% (n=3,072) |
| N1 | 76.0% (n=3,040) | 86.5% (n=126) | 76.4% (n=3,174) | 76.5% (n=3,266) |
5. Lowest-accuracy knowledge points
Only knowledge points with n≥100 and at least 5 learner identifiers are listed (23 qualify). n, level and category are shown alongside the percentage — small cells look extreme for reasons that are usually just noise.
| # | Knowledge point | Category | Level | n | Identifiers | Raw accuracy | 95% CI |
|---|---|---|---|---|---|---|---|
| 1 | Grammar patterns (N4) | Grammar patterns | N4 | 1,746 | 234 | 48.2% | 45.9%–50.6% |
| 2 | Grammar patterns (N3) | Grammar patterns | N3 | 2,536 | 390 | 49.2% | 47.3%–51.2% |
| 3 | Grammar patterns (N2) | Grammar patterns | N2 | 1,183 | 191 | 51.3% | 48.5%–54.1% |
| 4 | Grammar patterns (N1) | Grammar patterns | N1 | 1,299 | 118 | 57.0% | 54.3%–59.7% |
| 5 | Particle が | Particles | — | 358 | 81 | 60.9% | 55.8%–65.8% |
| 6 | Particle に | Particles | — | 370 | 75 | 62.4% | 57.4%–67.2% |
| 7 | Particle で | Particles | — | 330 | 85 | 63.3% | 58.0%–68.3% |
| 8 | Grammar patterns (N5) | Grammar patterns | N5 | 1,963 | 209 | 64.4% | 62.3%–66.5% |
| 9 | Word forms (N3) | Word forms | N3 | 913 | 294 | 73.2% | 70.2%–75.9% |
| 10 | Idioms (N2) | Idioms | N2 | 167 | 62 | 76.6% | 69.7%–82.4% |
| 11 | Particle を | Particles | — | 385 | 61 | 78.7% | 74.3%–82.5% |
| 12 | Word forms (N2) | Word forms | N2 | 278 | 107 | 80.2% | 75.1%–84.5% |
6. Most consistent strengths
| # | Knowledge point | Category | Level | n | Identifiers | Raw accuracy | 95% CI |
|---|---|---|---|---|---|---|---|
| 1 | Word forms (N1) | Word forms | N1 | 296 | 79 | 95.6% | 92.6%–97.4% |
| 2 | Vocabulary (N5) | Vocabulary | N5 | 16,029 | 311 | 94.4% | 94.0%–94.8% |
| 3 | Vocabulary (N1) | Vocabulary | N1 | 703 | 94 | 89.2% | 86.7%–91.3% |
| 4 | Vocabulary (N4) | Vocabulary | N4 | 2,828 | 243 | 89.0% | 87.8%–90.1% |
| 5 | Idioms (N1) | Idioms | N1 | 742 | 111 | 88.8% | 86.3%–90.9% |
| 6 | Word forms (N4) | Word forms | N4 | 461 | 149 | 87.0% | 83.6%–89.8% |
| 7 | Idioms (N4) | Idioms | N4 | 140 | 91 | 86.4% | 79.8%–91.1% |
| 8 | Idioms (N3) | Idioms | N3 | 290 | 142 | 85.9% | 81.4%–89.4% |
7. Observed patterns worth noting
Observation A: N1 raw accuracy sits above N2 and N3
N1 76.0% (n=3,040, 139 identifiers) is above N2 69.4% and N3 67.8%.
This is an observed pattern, not evidence that N1 is easier. At least three explanations cannot be ruled out with this data: only 139 identifiers answered N1 items versus 432 for N3, so the groups differ in composition; the question sets differ; and no ability estimation was performed. Answering "which level is genuinely harder" needs IRT equating and an anchor-item design, and this dataset has neither.
Observation B: the grammar-pattern axis is the lowest at every level
| Axis × level | n | Identifiers | Raw accuracy | 95% CI |
|---|---|---|---|---|
| Grammar patterns × N4 | 1,734 | 233 | 48.0% | 45.6%–50.3% |
| Grammar patterns × N3 | 2,531 | 389 | 49.2% | 47.2%–51.1% |
| Grammar patterns × N2 | 1,179 | 191 | 51.1% | 48.3%–54.0% |
| Grammar patterns × N1 | 1,295 | 118 | 57.0% | 54.3%–59.7% |
| Particles × N4 | 766 | 92 | 58.5% | 55.0%–61.9% |
| Grammar patterns × N5 | 1,939 | 208 | 64.0% | 61.8%–66.1% |
| Word forms / conjugation × N3 | 948 | 302 | 74.2% | 71.3%–76.8% |
| Particles × N5 | 1,220 | 111 | 75.8% | 73.3%–78.1% |
| Idiomatic expressions × N2 | 167 | 62 | 76.6% | 69.7%–82.4% |
| Word forms / conjugation × N2 | 278 | 107 | 80.2% | 75.1%–84.5% |
Crossing axis with level, almost every one of the lowest cells is a grammar-pattern cell. That is more informative than the level breakdown on its own.
8. What teachers and materials designers can use this for
This section deliberately separates data from interpretation. The data is citable; the interpretation is our judgement, and you are free to disagree with it.
| Data (citable) | Interpretation (our judgement, not data) |
|---|---|
| Quiz-context first attempts: grammar-pattern axis 53.7% vs vocabulary 92.1% (n=8,678 and 22,869) | On this platform there is a clear gap between recognising words and assembling sentences. If practice time has to be allocated, grammar patterns may have the higher marginal return. |
| が 60.9%, に 62.4%, で 63.3% sit well below を 78.7% and の 83.3% | を and の have comparatively narrow functions, while が, に and で are heavily polysemous. Teaching the distinct functions of a single particle separately may work better than introducing particles one at a time. |
| N2 and N3 raw accuracy (69.4% / 67.8%) sit below N4, N1 and N5 | ⚠️ Do not use this to rank difficulty. A more plausible reading is that the learners answering at N2/N3 differ in composition from those at other levels, rather than anything about the items. |
| Grammar patterns: 85.4% in the lesson context (n=13,937) vs 53.7% in the quiz context (n=8,678); paired difference +27.5 pp across 75 identifiers | In-lesson and quiz answering produce very different results — worth keeping in mind when platform practice scores are used to judge learning outcomes. ⚠️ But the two contexts use almost disjoint question sets, so we cannot say whether this is the context or simply easier questions. |
9. Methodology
| Item | Detail |
|---|---|
| Population | Answer records produced by learners using Nihongo to Japan. |
| Sampling | Self-selected, observational sample. Not random sampling; not population-representative. |
| Unit of analysis | answer event (one learner identifier answering one question once). |
| User count | learner identifiers = distinct anonymous identifiers. They cannot be merged across devices, so this is an upper bound on the number of people, not a verified headcount. |
| Metric | raw accuracy = correct ÷ valid answer events. Not an ability estimate. |
| Confidence intervals | Wilson 95% CI. Reflects sampling error only; does not correct for selection bias. |
Three analysis sets
| Set | Definition | n |
|---|---|---|
| Quiz-context first attempts (primary analysis set) | Answers from challenge / placement_test / special_training where this is the learner identifier’s first attempt at that question, not a repeat within 24 hours, with complete question metadata. | 37,245 |
| Lesson-context first attempts (comparison set) | First attempts from self_study / review — answers given while working through the lesson material. | 15,741 |
| First attempts (all contexts) | Each learner identifier's first answer to each question, regardless of source. | 52,934 |
| All answer events | Every valid answer event, including repeats. | 56,613 |
⚠️ The primary set makes no claim about hints or explanations. The fieldsused_hint/viewed_explanation/used_answer_keyare false throughout the database, but only because no producer ever writes them — none of the answer-capture call sites passes these parameters, and the product currently has no pre-answer hint or answer-reveal feature. The constant false is a property of the instrumentation, not evidence about learner behaviour. These fields are therefore not used as analysis conditions, and no claim is made about hint use.
Independent verification of the primary set: recomputing the earliest event per (identifier, question) on the server, 99.59% of primary-set events (37,094 of 37,245) pass the first-attempt test. The 151 that do not are most likely cross-device — the same person answering on two devices, each recording a first attempt. Every event in the set has attempt_number = 1.
What is excluded
- Known degraded questions: 520 question ids are excluded from every aggregate. Two causes: (1) non-grammar cloze items from the lesson materials (kana conversion, orthography, numeric answers) mis-tagged into the grammar-pattern axis; (2) an older "question of the day" build that used the answer token as the question identifier, collapsing different questions onto one id. The front end has been fixed; historical data is excluded via a denylist. No raw event was deleted — the table is append-only.
- Events after the snapshot cutoff: this report is frozen at 2026-08-14T00:00:00Z. Data keeps accumulating, but every figure on this page and in the downloadable files comes from that single frozen point.
Minimum cell size
Cells with n<30 or fewer than 5 learner identifiers are SUPPRESSED: the event count is published but the accuracy is not. Cells with n≥30 and ≥5 identifiers but n<100 are flagged LOW_SAMPLE and should always be cited with n. Cells with n≥100 and ≥5 identifiers are flagged OK.
In this snapshot 23 knowledge points are OK, 12 are LOW_SAMPLE and 3 are SUPPRESSED.
Anonymisation
- Each event carries only an anonymous device UUID generated in the browser. No name, email, IP address, device fingerprint, or free-text input of any kind.
- The receiving Cloudflare Pages Function writes whitelisted columns only and explicitly never writes IP addresses.
- The public dataset contains aggregate statistics only — not a single raw answer event.
- Learners can opt out of aggregate collection with one click on the privacy policy page.
10. Limitations (please cite these alongside the figures)
If you read only one paragraph, read this one. This data describes how people who came to Nihongo to Japan to practise performed on Nihongo to Japan questions. It is not a picture of Japanese learners in Taiwan, it is not a measure of JLPT difficulty, and it cannot be used to compare the objective difficulty of different levels.
- Selection bias: the sample is people who chose to practise here. They are likely more motivated than learners in general, and possibly younger and more comfortable with online tools.
- Repeated measures: the single most active identifier accounts for 3.7% of all answers and the top ten for 19.4%. Across the same 372 identifiers, event weighting and identifier weighting differ by 7.04 percentage points.
- Analysis sets are not directly subtractable: the lesson and quiz contexts draw on almost disjoint question sets (of 5,772 grammar-pattern questions, only 6 appear in both), so the accuracy gap between them cannot be attributed to a single cause.
- Hint and explanation behaviour is unobservable: the relevant fields are never written, so this dataset makes no claim at all about them.
- Unequal question difficulty: each level and axis draws on a different question set, so raw accuracy is not a normalised difficulty measure.
- No IRT equating: there is no item-response-theory equating and no anchor-item design across levels, so this data cannot answer which level is harder.
- Unequal sample sizes: there are more than six times as many N5 answers as N1 answers.
- Short time window: 12 days only (the aggregation pipeline went live on 2026-08-02). No longitudinal trend can be read from this data.
- Automated traffic cannot be identified: there is no mechanism to distinguish humans from scripts. Response time is null for 94% of events, so even a basic "answered impossibly fast" filter is unavailable. This is a known data-quality gap.
- Not population-representative: it does not represent all Japanese learners in Taiwan or in any other country.
11. How to cite
Citation in papers, research reports, teaching materials and journalism is welcome. Please credit the source with a link back to this page.
APA (7th)
Nihongo to Japan. (2026). Japanese learning answer dataset — August 2026 (Version 1.1.0) [Data set]. https://www.nihongotojapan.com/en/research/japanese-learning-data-2026MLA (9th)
Nihongo to Japan. Japanese Learning Answer Dataset — August 2026. Version 1.1.0, 2026, https://www.nihongotojapan.com/en/research/japanese-learning-data-2026.BibTeX
@misc{ntj_jlad_2026,
title = {Japanese Learning Answer Dataset --- August 2026},
author = {{Nihongo to Japan}},
year = {2026},
version = {1.1.0},
note = {Snapshot 2026-08-14; aggregate statistics from 56,613 anonymous answer events, 2--13 August 2026},
url = {https://www.nihongotojapan.com/en/research/japanese-learning-data-2026}
}Licence status
This is not an open dataset and is not under any standard open licence (CC BY, CC0 or similar). Nihongo to Japan is free to use, but "free to use" is not the same as "openly licensed".
Explicitly permitted:
- Citing the statistics on this page in papers, research reports, teaching materials and journalism, with credit and a link back to this page.
- Downloading the aggregate data, re-analysing it for research, and producing your own charts.
- Classroom and educational use.
Charts: the charts on this page are available to view. To reproduce our charts, please contact us first. The aggregate figures themselves may be cited under the terms above.
Explicitly not permitted: presenting these figures as representative of all Japanese learners in Taiwan or anywhere else; republishing the numbers with the sampling limitations removed; redistributing the dataset wholesale as your own data product.
For anything else, please get in touch via the about page.
12. Download the data
Two formats with identical content, both aggregate statistics only (no raw answer events):
- japanese-learning-2026.json — full metadata, definitions, methodology, limitations and every aggregate cell.
- japanese-learning-2026.csv — one row per cell. Columns:
dimension, key, label_zh, label_en, category, jlpt_level, analysis_set, n, learners, accuracy, ci95_low, ci95_high, flag.
Version 1.1.0 · snapshot 2026-08-14 · methodology version 1.1. Future updates will increment the version and preserve the earlier figures rather than overwriting them in place.
⚠️ For citation, use the versioned frozen dataset rather than the live aggregate endpoint.
A live aggregate endpoint also exists at /api/aggregates, returning current numbers on each call. It uses different definitions: an all-answer-events basis rather than quiz-context first attempts, and its totalUsers counts only the primary-set learner identifiers (688) rather than all 1,053. Its numbers change daily and are not suitable for citation.
For journalists
- Sample: 56,613 anonymous answers from 1,053 anonymous learner identifiers (not a verified headcount) over 12 days, 2–13 August 2026.
- Method in one line: raw accuracy on a self-selected observational sample; headline figures use the quiz-context first-attempt set (first attempts in the challenge, placement-test and drill contexts).
- Limitation in one line: this is not a representative survey of Japanese learners, and it cannot be used to compare the objective difficulty of JLPT levels.
Three figures you can quote directly:
- Grammar-pattern accuracy 53.7% (n=8,678) versus vocabulary 92.1% (n=22,869) — a gap of 38 points.
- The particle が is answered correctly 60.9% of the time (n=358); を reaches 78.7% (n=385).
- Grammar-pattern accuracy runs from 53.7% to 74.1% depending on which answering contexts are included (lesson context 85.4%, quiz context 53.7%). ⚠️ The two contexts use almost disjoint question sets so the cause cannot be pinned down — but that is itself a story about how learning-platform data gets quoted.
Citation formats · Download the data · Contact
For researchers and educators
Structure: one row per aggregate cell, indexed by dimension (axis / jlpt / knowledge_point / axis_x_jlpt) × analysis_set (quiz_first_attempt / lesson_first_attempt / first_attempt / all_events). Each cell reports n, learners, accuracy, Wilson ci95_low/ci95_high and a flag. The JSON additionally carries the analysis-set decomposition, the paired comparison and the weighting effect.
Sampling: self-selected and observational; the unit of analysis is the answer event; a learner identifier is an anonymous UUID, so the identifier count is an upper bound on the number of people (the same person on two devices produces two identifiers, and identifiers cannot be linked).
Known methodological gaps — we think these are worth stating rather than hiding:
- No IRT difficulty equating and no cross-level anchor items, so between-level comparisons do not hold.
- The lesson and quiz contexts draw on almost disjoint question sets (only 6 of 5,772 grammar-pattern questions overlap), so context effects and item difficulty cannot be separated.
- The hint / explanation / answer-key fields are never written, so those behaviours are unobservable. They cannot be used as analysis conditions and their absence cannot be read as "it did not happen".
- Response time is null for 94% of events, so no speed–accuracy trade-off analysis and no filtering of implausibly fast answers.
- No demographic variables (age, first language, years of study, country), so no stratified analysis is possible. The dataset has no first-language field and no geography field, so this report makes no inference about learners' native language or location.
- A 12-day window, so no learning curves and no retention analysis.
- Learner-level data is not published (aggregates only), so multilevel models cannot be refitted externally. If that matters for your research, please get in touch.
Download JSON · Download CSV · Citation formats · Full methodology
← Back to Research & Data | How Taiwan Learns Japanese 2026 | For educators & libraries