6.1 Listening Formats & Strategy

The Section You Cannot Skip

The N2 listening section (聴解, chōkai) is worth 60 of the 180 total points — a full third of your score — and it carries its own sectional minimum of 19 out of 60. Because N2 uses three independent minimums, a strong Vocabulary/Grammar and Reading result cannot rescue a failed listening score: score 18/60 on listening and you fail the whole exam even if your total is well above 90. The block runs about 50 minutes, and every clip plays exactly once at natural, unpaused native speed. There are no slow-learner pauses, no repeats, and no printed transcript.

A note on this guide

This is a text study guide with no audio. You cannot literally practise ears-on listening here, so this chapter teaches the skills instead: the five fixed formats, the spoken-Japanese cues that decide answers, and worked transcripts you can read aloud or shadow. After each idea, drill the real thing on the free practice bank at /practice/jlpt-n2 and with the JLPT Official Practice Workbook audio — reading a transcript builds the decision framework, but only listening at speed trains the reflex.

The Five Problem Types

Every N2 listening test follows the same five 問題 (mondai, problem sets) in the same order. The narrator names the type and reads an example before each set, so you always know which game you are playing. The decisive fact is whether the answer choices are printed on your booklet — because that changes where your eyes go and whether you must hold four options in memory.

問題1 課題理解 — what happens NEXT

You hear a situation setup, the question, a conversation, and then the question again. Options (often short printed phrases) sit on the booklet. The trap: a conversation lists several tasks — call the client, print the handout, book a room — but only ONE is the thing this specific person must do right after the talk. Listen for the final instruction and for cancellations: 「あ、それはもういい」「じゃあ、先に〜して」. Read the printed options during the setup so you can map spoken tasks onto them.

問題2 ポイント理解 — one narrow fact

The question is stated first, then you get a few seconds to read the printed options, then the audio plays. Because you know the question in advance (「女の人はどうして遅れましたか」= why was she late?), listen only for that point and let the rest wash past. Distractors are reasons that are mentioned but rejected: 「電車じゃなくて、実は目覚まし時計が…」.

問題3 概要理解 — the big picture

Nothing is printed. You hear a monologue or dialogue, then the question and the four choices are all spoken. This tests the main theme or the speaker's overall opinion (発話意図), never a tiny detail. Do not chase numbers or names — track the direction of the argument (praising? warning? proposing?). Because the choices come after the passage, keep a one-line mental summary ready.

問題4 即時応答 — reflex replies

Also nothing printed, and uniquely it has only three choices, all spoken. You hear one short line — often keigo, a favour request, or a set phrase — and pick the natural reply. These are fast (~11–12 items) and decided by register and function: an apology wants acceptance, an offer wants thanks or polite refusal. Miss one? Let it go instantly and reset for the next; freezing costs you the following item too.

問題5 統合理解 — integrate and compare

A longer talk (e.g. a shop assistant describing four plans, or three people discussing options). Note-taking is essential. The final item usually prints two questions with choices (e.g. which plan each speaker picks). Track who wants what with a quick grid.

Strategy: predict, note, don't freeze

Predict from the stem. For 問題1/2 the question is read first — decide what to hunt for before the conversation starts. Take skeletal notes: initials for speakers (M/F), arrows for actions, ○/× for accept/reject, and jot numbers, times, and 「やっぱり」/「でも」 — the words that flip decisions. Never get stuck. One clip plays once; if you lose an item, commit to your best guess, breathe, and be fully present for the next narrator line. A single freeze can cost three items in a row.

A worked prediction

Suppose 問題2 opens with 「女の人はどうして会社を辞めますか」 (why is the woman quitting?). Before a word of dialogue, you already know your only job is to isolate a reason, and that the four printed choices are candidate reasons you should skim now. As the audio runs, you can safely ignore greetings, weather small-talk, and side jokes, spending all attention on causal markers — から, ので, ため, and the reversal frame 「〜んじゃなくて、実は…」. This is why key-point items reward preparation more than raw listening ability: the stem hands you a filter, and disciplined listeners use it to discard 80% of the audio and lock onto the one clause that answers the question.

Common mistake: answering the FIRST thing you understand

Lower-scoring test-takers grab the first option that matches a word they recognised and stop listening. But N2 clips are built so the early information is usually a decoy and the decisive line comes near the end. Keep listening through the whole clip, hold two candidates if unsure, and choose only after the speaker stops. Comprehending 60% of the words but catching the final decision beats understanding 90% and answering too early.