1.1 Official JLPT N4 Structure, Timing & Scaled Scoring

Official JLPT N4 Structure, Timing & Scaled Scoring

Core Principle: The Japanese-Language Proficiency Test (JLPT) Level N4 is an internationally standardized examination designed to evaluate and certify the Japanese language proficiency of non-native speakers at an elementary level. Administered worldwide by the Japan Foundation and inside Japan by Japan Educational Exchanges and Services (JEES), the N4 verifies that a learner can understand basic Japanese encountered in daily life, read passages on familiar topics written in basic vocabulary and kanji, and comprehend everyday conversations spoken at a moderate, natural pace.

Achieving N4 certification serves as a recognized benchmark for academic institutions, international study programs, and practical immigration visa categories—most notably qualifying applicants for the Japanese government's Specified Skilled Worker (SSW-1 / 特定技能1号) visa tracks in sectors such as nursing care, hospitality, and agriculture.

---

1. Exam Architecture & The Three Timed Sessions

The JLPT N4 is administered as a paper-and-pencil examination consisting entirely of multiple-choice questions marked on Optical Mark Recognition (OMR) answer sheets. There are no speaking, essay writing, or interactive oral components at any level of the JLPT.

The examination is structured into three separate physical test sessions totaling 115 minutes of active testing time. Each session has its own dedicated question booklet and separate OMR answer sheet:

Check the edition date of any book you study from. The item counts above are the figures the Japan Foundation published when it shortened N4 and N5 with the December 2020 test. Before that change, Vocabulary carried 35 items and Grammar & Reading carried 35. A large amount of prep material still in circulation prints the old numbers, so a practice test with 35 vocabulary items is testing you on a format that no longer exists. The organisers also state plainly that "the actual number of test items may vary slightly from session to session" — treat every count in this guide as an official approximation, not a guarantee.

Critical Administrative Timing Rules

• Strict Physical Segregation: Each session is conducted independently. Once the 25 minutes for Session 1 expire, the proctors collect both the question booklets and answer sheets for Vocabulary before Session 2 materials are distributed. You cannot borrow time from Vocabulary to use during Grammar & Reading, nor can you return to earlier sections.
• Supervised Intermissions (Kyūkei / 休憩): Unscored administrative breaks separate Session 1 from Session 2 and Session 2 from Session 3, and proctors count, verify, and seal all examination materials during them. Only the three testing times (25 / 55 / 35 minutes) are fixed worldwide — the room-entry time and the length of each break are set by the body running your test site (JEES in Japan, your local host institution overseas) and are printed on your test voucher. Read that timetable rather than assuming a standard break length.
• Audio Synchronization: Session 3 (Listening) is paced entirely by a centralized, uninterrupted audio recording. The recording is played exactly once; audio tracks are never repeated or paused for individual test rooms.

---

2. Comprehensive Problem Set (Mondai) Taxonomy

Within each of the three test sessions, questions are categorized into standardized, numbered problem sets called 問題 (Mondai). Understanding the specific task required by each Mondai prevents disorientation and saves valuable test time.

Session 1: Language Knowledge — Vocabulary (25 Minutes)

1. 問題1 — Kanji Reading (漢字読み, ~7 items): A target word written in kanji is underlined in a sentence. You must select the correct reading in hiragana from four choices.
   • Example focus: Distinguishing voiced vs. unvoiced consonants, long vowels, and contracted sounds (e.g., 案内 → あんない vs. あんないい vs. あんない).
2. 問題2 — Orthography / Kanji Selection (表記, ~5 items): A target word written in hiragana is underlined. You must select the correct kanji representation.
   • Example focus: Differentiating visually similar radicals and homophones (e.g., つくる → 作る vs. 使う).
3. 問題3 — Contextual Vocabulary (文脈規定, ~8 items): A sentence contains a blank slot. You must select the semantically and collocationally appropriate word to complete the sentence.
   • Example focus: Natural verb-noun collocations, transitive vs. intransitive verb distinctions, and thematic daily-life vocabulary.
4. 問題4 — Paraphrases / Synonyms (言い換え類義, ~4 items): A sentence is presented with an underlined phrase. You must choose the option closest in meaning to the underlined statement.
   • Example focus: Vocabulary equivalence (e.g., お金をおろす → お金を出す).
5. 問題5 — Contextual Usage (用法, ~4 items): A single target word is given. You must select the sentence among four choices that uses the word in a natural, grammatically correct context.
   • Example focus: Eliminating sentences with unnatural semantic contexts or incorrect grammatical particles.

Session 2: Language Knowledge — Grammar & Reading (55 Minutes)

Part A: Grammar Subsections (~21 items, ~22 minutes budgeted)

• 問題1 — Form Selection (文法形式の判断, ~13 items): Selecting the correct particle, conjugational ending, auxiliary verb, or conditional form to complete a sentence.
• 問題2 — Sentence Composition / Star Questions (文の組み立て ★, ~4 items): Four scrambled word fragments must be mentally reordered to form a coherent sentence. You indicate the fragment that falls into the star-marked (★) position.
• 問題3 — Text-Level Grammar (文章の文法, ~4 items): A cohesive short passage (e.g., a diary entry, personal letter, or informational note) contains numbered blank slots. You must select connecting adverbs, conjunctions, and discourse markers that fit the overall textual logic.

Part B: Reading Subsections (~8 items, ~30 minutes budgeted)

• 問題4 — Short Passages (短文読解, ~3 items): Passages the Japan Foundation describes as "easy original text of approximately 100–200 characters" on study, everyday-life and work topics (notices, emails, brief daily narratives), with one question each. Questions test direct factual comprehension and core communicative intent.
• 問題5 — Mid-Length Passages (中文読解, ~3 items): "Easy original text of approximately 450 characters" on everyday-life topics (expository essays, descriptive stories), typically one passage carrying all the questions. Questions evaluate causal relationships, main arguments, and chronological sequences.
• 問題6 — Information Retrieval (情報検索, 1 document, ~2 items): Practical real-world materials the blueprint describes as "notices (approximately 400 characters)" — bulletin board announcements, transport schedules, promotional coupons, library guidelines. Questions test the ability to locate specific target constraints (e.g., eligibility, deadlines, pricing conditions) without reading the entire document linearly.

Session 3: Listening (35 Minutes)

1. 問題1 — Task-Based Comprehension (課題理解, ~8 items): The question and multiple-choice options (text or illustrations) are printed in the booklet. You listen to a dialogue and determine the immediate next action a speaker will take.
2. 問題2 — Point Comprehension (ポイント理解, ~7 items): The question and options are printed in the booklet. You listen for specific underlying details such as reasons, motivations, times, locations, or conditions.
3. 問題3 — Verbal Expressions (発話表現, ~5 items): Illustrations are printed showing a specific conversational scenario with an arrow pointing to a speaker. You select the most appropriate phrase or greeting for that speaker to say from three spoken choices.
4. 問題4 — Quick Response (即時応答, ~8 items): Nothing is printed in the booklet. You listen to a short spoken prompt (a question, request, or greeting) followed by three short spoken replies, selecting the most natural conversational rejoinder in real time.

3. The Scaled Scoring Model & The Double-Hurdle Rule

Unlike traditional classroom exams where raw scores are calculated as simple percentages, the JLPT uses an equated scaled scoring model. Raw points are converted into standard scores ranging from 0 to 180 total points.

Official Score Reporting Bands

On your official JLPT Score Report (合否結果通知書), your results are grouped into two scoring sections:

1. Language Knowledge (Vocabulary/Grammar) & Reading: Scored from 0 to 120 points.
   • Note: Although Vocabulary and Grammar/Reading are tested across two separate test sittings, their scores are combined into a single unified 120-point score band on the official certificate.
2. Listening: Scored from 0 to 60 points.

The Double-Hurdle Requirement

To earn an official N4 pass certificate, an examinee must clear three simultaneous numerical thresholds:

The Sectional Minimum Rule: Failing to achieve the sectional cutoff in either band results in an automatic overall failing grade (不合格), even if your total scaled score exceeds the 90-point overall cutoff.

The CEFR Level Printed on Your Score Report (from the December 2025 Test)

Beginning with the December 2025 administration, the JLPT score report carries a reference indication of the CEFR level (Common European Framework of Reference for Languages). Read the official wording carefully, because it is narrower than the "N4 = A2" shorthand you will see on study blogs:

The same "Reference Information" band of the report also shows a percentage-of-correct-responses grade (A / B / C) for Vocabulary, Grammar and Reading separately, and — for examinees outside Japan — a percentile rank. None of these feed into pass/fail either; they exist so you can see which part of Language Knowledge dragged your combined 120-point band down, which the single combined score deliberately hides.

Study consequence. The A/B/C reference grades are the only official diagnostic you will ever receive. If you fail, the split between the Vocabulary, Grammar and Reading letters tells you where to aim the next six months far better than the raw 0–120 number.

Illustrative Performance Profiles (Pass vs. Fail Case Studies)

To understand how the double hurdle works in practice, examine these three real-world candidate scenarios:

• Candidate A (Balanced Profile — PASS):
  • Language Knowledge & Reading: 52 / 120 (Clears 38-point minimum)
  • Listening: 39 / 60 (Clears 19-point minimum)
  • Total Score: 91 / 180 (Clears 90-point total cutoff)
  • Result: PASS (合格). Although near the boundary, all three hurdles are met.
• Candidate B (Lopsided Profile — FAIL due to Listening Cutoff):
  • Language Knowledge & Reading: 102 / 120 (Exceptional performance)
  • Listening: 18 / 60 (Below the 19-point minimum by 1 point)
  • Total Score: 120 / 180 (Far above the 90-point total cutoff)
  • Result: FAIL (不合格). Despite an impressive total score of 120/180, missing the sectional minimum in Listening produces an automatic failure.
• Candidate C (Balanced but Sub-Threshold — FAIL due to Total Score):
  • Language Knowledge & Reading: 54 / 120 (Clears 38-point minimum)
  • Listening: 34 / 60 (Clears 19-point minimum)
  • Total Score: 88 / 180 (Below the 90-point total cutoff by 2 points)
  • Result: FAIL (不合格). Both sectional cutoffs were cleared, but the composite total failed to reach 90 points.

---

4. Item Response Theory (IRT) Scaling Mechanics

The Japan Foundation calculates JLPT scores using Item Response Theory (IRT). Under IRT psychometric modeling, a candidate's score is not a direct sum of correct answers. Instead, it is a statistical estimate of the candidate's latent Japanese language ability.

Why IRT is Used Over Raw Percentage Scoring

1. Equating Across Test Administrations: Because the JLPT is offered twice a year (July and December) across thousands of global sites with newly authored test items every session, minor differences in booklet difficulty inevitably occur. IRT statistical equating adjusts for these variances so that a scaled score of 95 on a July exam represents the exact same level of Japanese ability as a scaled score of 95 on a December exam.
2. Differential Item Weighting: Under IRT mathematical models, items possess varying statistical parameters (including difficulty and discrimination power). Answering a highly discriminating item correctly provides more information about your latent ability than answering a low-discrimination item.
3. Elimination of Guessing Distortion: IRT models account for probabilistic guessing patterns across the entire pool of global examinees, dampening the statistical inflation caused by random guesses on extreme outlier questions.

Strategic Implications for Candidates

• No Negative Marking: The JLPT imposes no penalty for incorrect answers. A wrong answer scores zero; it never deducts points. Therefore, you should never leave any bubble blank on the OMR sheet.
• Do Not Get Trapped on Anomalous Items: If you encounter an unfamiliar kanji or complex sentence, do not spend three minutes agonizing over it. Under IRT, standard intermediate items that form the core syllabus carry the decisive statistical weight. Make an educated guess, mark the bubble, and protect your pacing for the rest of the test.