JLPT N1 common mistakes
JLPT N1 · Understand the traps and how to avoid them.
Works through the reading passages strictly in order, spending 8–10 minutes on each, then reaches the final long passage and the 情報検索 (information-retrieval) question with only a few minutes left and has to guess.
The Language Knowledge + Reading booklet gives roughly 100 minutes for a large amount of text. Candidates treat every question as equally valuable and try to understand every sentence fully, which is impossible at N1. The 情報検索 question looks intimidating (tables, small print) so it gets pushed to the end even though it is the fastest to answer.
What to do instead
Set a per-passage budget before the exam (e.g. 短文 2 min, 中文 5 min, 長文 9 min, 情報検索 3 min). Do the 情報検索 question first: it is pure keyword matching and needs no full reading. If any question runs 1 minute over budget, mark a provisional answer and move on.
On 筆者の主張・主旨 questions, chooses an option that is literally stated in the passage but is only a supporting example, or an option containing absolute wording such as すべて・必ず・絶対に・誰でも.
Those options contain familiar vocabulary copied from the text, so they feel "safe". Candidates match words rather than roles: they do not check whether the sentence is the author's conclusion or an illustration of it, and they under-weight qualifiers like 必ずしも・一概には・～とは限らない that the author used deliberately.
What to do instead
Before looking at the options, underline the sentence containing ～べきだ・～ではないだろうか・要するに・つまり・したがって - that is the claim. Then reject any option whose scope is wider than the claim (absolute wording, or a claim about "everyone/society" when the author spoke about a limited case) and any option that merely restates an example.
Rejects the correct answer because it does not reuse the passage's exact words, and instead picks the option that repeats the most vocabulary from the text.
N1 correct options are almost always paraphrases (言い換え), while distractors are built by quoting prominent nouns and verbs from the passage and attaching a wrong relation. Candidates who read for word-matching are systematically trapped by exactly this design.
What to do instead
For each option, name in Japanese what relation it asserts (原因, 目的, 条件, 逆接, 対比) and check that relation against the passage, not the vocabulary. Train by covering the options, summarising the relevant paragraph in one short phrase of your own, then choosing the option closest to that phrase.
In 文脈規定 vocabulary questions, chooses the word whose kanji suggest the right meaning even when it does not collocate with the surrounding noun or particle (e.g. picking 発表する where the sentence needs 講演する, or 深刻 where the collocation requires 重大).
Many N1 vocabulary items are near-synonyms distinguished only by collocation, transitivity, or register. Candidates memorise a single Chinese-character gloss per word and never see the word in a sentence, so they cannot feel which partner nouns it accepts.
What to do instead
Memorise vocabulary in fixed chunks, not as isolated words: record the noun it usually takes, the particle it takes, and one whole example sentence. When a question has two plausible meanings, decide by collocation (which noun/particle fits) rather than by translation.
In 文章の文法 (text grammar), fills each blank by testing only the sentence containing the blank, ignoring the sentences before and after, and so chooses a connective with the wrong logical direction or a tense that clashes with the rest of the paragraph.
The format looks like ordinary fill-in-the-blank grammar, so candidates apply sentence-level rules. But this item tests discourse: 逆接 vs 原因 vs 並列 across sentences, and subject/tense consistency across the paragraph.
What to do instead
Always read from the sentence before the blank to the sentence after it, and first label the relation between the two clauses (理由? 逆? 追加? 例? 言い換え?). Then eliminate options by relation, and only afterwards check form. Keep a short mental list of connectives grouped by relation.
In 文法形式判断, selects a grammar pattern because the meaning seems roughly right, without checking the required preceding form or the pattern's built-in tense/polarity (e.g. using ～が早いか after a ます-stem, or ～に限って where the sense requires ～に限らず).
Study lists pair each pattern with an English/Chinese gloss, so the learner stores meaning only. N1 patterns are distinguished by rigid 接続 rules and by whether they express a one-off event, a habitual state, or a negative evaluation.
What to do instead
For every pattern, memorise four things together: preceding form, what follows it, polarity/tense it forces, and one full example. When two options both "mean" something similar, decide on 接続 first - often only one option can legally attach to the given form.
During the listening section, silently translates each sentence into their own language before moving to the next, so by the third or fourth sentence they are still processing sentence two and then miss the key information entirely.
Translation feels like comprehension and is the habit built in earlier study. At N1 speed (natural, near-native rate with reduced forms) translation lags several seconds behind the audio, and there is no pause to catch up.
What to do instead
Practise listening without translating: shadow the audio aloud, and train yourself to hold meaning as images/keywords in Japanese. If you notice you are translating, drop the current sentence immediately and rejoin the audio at the next marker word (でも・だから・つまり・ただし) rather than trying to reconstruct what you lost.
Still mentally finishing the previous question when the next item's question statement (質問文) is read, so they hear the dialogue but do not know what they are supposed to be listening for.
Each item is played once with no rewind, and the stem is read before the dialogue. Candidates who review the last item during that window lose the only chance to know the target, and then listen "generally" instead of for a specific fact.
What to do instead
Treat the stem as the most important audio in the item: stop all thinking at the end of the previous item, and write down the target (e.g. 男は何をしますか / 女が最も驚いたことは). Use the option-reading pause to mark differences between options before the dialogue starts.
In ポイント理解, takes notes that are a linear transcript of what was said, then picks the option mentioned first or most often, instead of the one that satisfies the conditions actually asked about.
N1 dialogues deliberately have speakers discuss and reject several options, and the first-mentioned candidate is usually wrong. Without a structured note, the most recently heard or most frequently repeated item feels like the answer.
What to do instead
Use a fixed note grid: 誰が / 何を / いつ / どこ / 条件（～なければ、～の代わりに、～ずに）. Write each option with a ✓ or ✗ and the reason it was rejected. Answer from the grid, and specifically re-check for negation and condition words (ただし、～ないと、～さえ～ば).
In 即時応答, answers by finding the response that shares a word with the prompt, or that is polite in tone, rather than one that performs the required speech act (declining, apologising, correcting, hedging).
The exchange is one line each, with no context build-up, so there is nothing but the prompt to work from. Word-overlap options are written precisely to attract this, and 丁寧な言い方 sounds acceptable even when it is functionally wrong (e.g. a polite refusal where the prompt was an offer the speaker must accept).
What to do instead
Classify the prompt's function first (依頼・勧誘・申し出・確認・苦情・褒め), then judge options by whether they fit that function, discarding any that only repeat a word. Pay attention to the final particle and intonation of the prompt, since ～ない？ and ～ないか vs ～ねえ affect whether agreement or disagreement is required.