JLPT N5 common mistakes
JLPT N5 · Understand the traps and how to avoid them.
Guessing the meaning of katakana loanwords from the English word they resemble (assuming アルバイト means 'work/arbeit', マンション means 'mansion', or サービス means 'service charge') instead of learning what they actually mean in Japanese.
N5 learners rely on their English or other loan vocabulary and do not check that Japanese has shortened, shifted, or narrowed the meaning; the katakana looks transparent, so the learner never looks it up.
What to do instead
Build a 60–80 item katakana list with the real Japanese meaning written in your own words (アルバイト = part-time job, マンション = concrete apartment block, サービス = something given free or extra). Test yourself from the katakana to the meaning, not from English to katakana, and mark every word whose meaning differs from the source language.
Using 一つ/二つ for everything - saying 人が 一つ, 本が 二つ, or 犬が 三つ - and misreading counters such as 一人, 四本, 三匹 when they appear with sound changes.
The つ counter is taught first and feels universal, and N5 courses often introduce counters as a vocabulary list without drilling the sound changes (ひとり, ふたり, よにん; いっぽん, さんぼん; いっぴき, さんびき).
What to do instead
Make a grid of 人・枚・本・匹・台・回 across 1–10 and say it aloud daily until the irregulars are automatic (一人, 二人, 四人; 一本, 三本, 六本, 八本). When you meet a new noun, write the counter that belongs to it on the same card so noun and counter are learned together.
Confusing hiragana minimal pairs when reading or writing - おばさん/おばあさん, おじさん/おじいさん, きて/きって, ゆき/ゆうき - and dropping or adding small っ and long vowels in answers.
Long vowels and small っ carry meaning in Japanese but are invisible to learners whose first language does not mark length, so the eye and ear skip them; word processors also autocorrect, hiding the error during practice.
What to do instead
Drill contrast pairs aloud and by hand: say both members, then write both from dictation with no autocorrect. When reviewing a wrong answer, copy the whole sentence and circle every っ, ー, and small ゃ/ゅ/ょ, so length becomes something you check deliberately rather than something you hope to hear.
Choosing the wrong particle for location or destination - writing 部屋で猫がいます, 学校を行きます, or 電車に学校へ行きます - because the noun feels like the important part of the sentence.
Learners memorise particles as translations of English prepositions ('in', 'at', 'to') rather than as markers tied to the verb, so they pick the particle before they look at what the verb is doing.
What to do instead
Learn particles as verb patterns, not as word meanings: に with います/あります, に with 行きます/来ます, で with action verbs such as 読みます/食べます, で with means of transport, を with transitive verbs. Before writing a sentence, identify the verb first, then attach the particle, and keep a one-page pattern sheet with three example sentences per pattern.
Forming the て-form and past forms by pattern-guessing - producing 行いて, 帰って from the wrong verb group, 食べて for a godan verb, or 来て read as きて when it should be きて/くて in context - instead of identifying the verb group first.
Textbooks present て-form rules as a table of endings, and learners apply the ending they saw most recently without checking whether the dictionary form is godan, ichidan, or irregular; 帰る and 切る look ichidan but are godan, and 行く breaks the く rule.
What to do instead
Before conjugating, say the dictionary form and label the group out loud (godan / ichidan / irregular). Memorise the exception set explicitly: 行って, 来て, して, and godan verbs ending in -いる/-える such as 帰る, 入る, 切る. Then do 20 mixed conjugation items a day, writing the group label beside each answer so the habit is checking, not guessing.
Selecting a reading answer because it repeats words from the passage, without checking negation and tense - for example, matching 田中さんは行きました to a passage that says 田中さんは行きませんでした, or 高くない to a passage that says 高かった.
The repeated words create a feeling of recognition, and under time pressure the learner compares vocabulary rather than meaning; N5 distractors are built precisely by copying passage words and flipping ない/ません/でした.
What to do instead
Before looking at the options, underline every verb ending in the passage (ます/ました/ません/ませんでした, い-adjective くない/かった) and note the polarity and time next to it. Then check each option against those notes; any option that changes polarity or tense is wrong no matter how many words it shares.
Reading the passage word by word from the first line and re-reading the same sentence when one word is unknown, so time runs out and the last question is answered randomly.
N5 learners treat every unknown word as a stopping point because they have been trained to understand everything, and they read the text before knowing what the question asks, so they cannot tell which sentence matters.
What to do instead
Read the question first and circle the question word (だれ, いつ, どこ, なに). Then scan the short passage for that word or its equivalent, read only the sentence around it carefully, and move on. If a word is unknown, cover it and continue; N5 items are answerable from the surrounding sentence. Time yourself on three-question sets to build the habit of finishing the section.
Answering from the first content word heard rather than the end of the sentence - treating ～ませんか as ～ましょう, 行きません as 行きます, or picking the option that names the place mentioned first without waiting for the verb that says what happened there.
Japanese puts negation, tense, and the speaker's intent at the end, while the listener is still processing the noun that came first; N5 audio is slow enough that learners relax and stop listening after they recognise a familiar word.
What to do instead
Train yourself to hold the sentence until the final mora. Practise with pairs such as 行きます／行きません, 行きましょう／行きませんか, and decide only after the ending. During the exam, jot a symbol (○ for yes, × for no, ? for invitation) as the sentence ends, and use the printed answer choices before the audio starts to predict what kind of ending you are listening for.
Losing numbers, times, prices, and counters in listening tasks - writing 4時 for 7時, しち for なな, or missing 三本/三千 - and having to guess because no note was taken.
Numbers arrive fast, several appear in one item as corrections (「三時じゃなくて、四時です」), and learners try to keep them in memory while still listening to the next sentence.
What to do instead
Do daily dictation of 20 numbers, times, and counted items read aloud, writing digits immediately. Practise items where the speaker corrects themselves, and always write the first number, cross it out, and write the correction. In the exam, have the pencil down and write every number as you hear it rather than after the item ends.
Being distracted by a word that appears in both the audio and an option, and choosing that option without checking whether it answers the question asked (the audio mentions 駅, so the option containing 駅 is selected even though the question was about the time).
Keyword matching feels like comprehension, and N5 listening items deliberately plant mentioned-but-irrelevant words in the wrong choices; learners also skip the pause where the answer choices are read and first seen.
What to do instead
Use the pause before the audio to read the choices and underline what each one actually claims (a time, a place, a person). While listening, listen for the question word and the sentence ending, and eliminate any choice whose key element was never confirmed. After each practice item, write one line saying which word in the audio was the trap, to make the pattern recognisable.