NCEA L1 Japanese common mistakes
NCEA L1 Japanese · Understand the traps and how to avoid them.
Hearing a familiar word in a listening passage and writing down that word as the answer, rather than answering what the question actually asks (e.g. hearing がっこう and writing 'school' when the question asks when the person leaves school).
Under time pressure candidates latch onto the first recognisable vocabulary item and stop processing the rest of the sentence; the listening passage is played at natural speed and they cannot rewind, so they grab a life raft word instead of parsing the whole utterance.
What to do instead
Before the audio starts, underline the question word (いつ/どこ/だれ/なに/どうして) and the target person. While listening, listen for the sentence-final predicate and any time/place markers attached to it, not isolated nouns. Practise with short authentic audio where you must answer one specific question, and train yourself to note the whole clause (e.g. あしたのあさ、でます) rather than a single word.
Confusing long vowels, small っ (geminate consonants), and ん before a consonant in listening, so きって (stamp) is heard as きて (come), or おばあさん as おばさん.
NCEA Level 1 candidates often learn vocabulary visually first; the moraic timing distinctions (long vowel, double consonant, moraic ん) are not salient to English-speaking ears and are not reinforced by reading romaji, which hides them.
What to do instead
Do minimal-pair dictation daily: pairs like きて/きって, おばさん/おばあさん, しゅくだい/しゅくだいません, ビル/ビール. Clap or tap the morae while listening so the extra beat becomes physical. When you learn a new word, say it aloud with its mora count written out (e.g. がっこう = が-っ-こ-う, four beats).
Missing the politeness/register and speech level of a listening or reading text and therefore misidentifying who is speaking to whom, or choosing an answer in the wrong register.
Candidates memorise plain and polite forms as separate lists rather than as a system, so when a passage mixes です/ます with plain forms (common in reported speech and in conversations between friends), they lose track of the relationship and the referent.
What to do instead
For every listening or reading text, explicitly note the register markers (です/ます/ございます vs だ/る/た) and what they imply about the relationship. Practise converting the same short message between plain and polite forms so the mapping becomes automatic, and check whether the question asks you to answer as the speaker, the listener, or a third party.
Preparing a memorised script for the speaking interaction and delivering it regardless of what the interlocutor actually asks, or freezing when the question deviates from the rehearsed version.
Candidates rehearse monologues rather than the interaction, and the stress of a live exchange makes them fall back on the rehearsed chunk. NCEA Level 1 speaking tasks reward interaction (asking back, responding appropriately), not recitation.
What to do instead
Rehearse question–answer pairs, not speeches: record yourself answering the same question three different ways, and have a partner ask unpredictable follow-ups (なぜ? いつ? だれと?). Build a small bank of repair phrases (すみません、もういちど おねがいします; ええと、それは…; 〜は どうですか) so an unexpected question becomes a chance to interact rather than a failure.
Producing romaji-style pronunciation - flat pitch, English stress on the first syllable, and vowels reduced to schwa (e.g. です pronounced 'dess', ありがとう with stress on 'ri').
Candidates read Japanese through English orthography and transfer English stress-timing; Japanese is mora-timed with relatively even duration and no stress accent, and vowel devoicing/reduction is an English habit that makes output hard to understand and costs pronunciation marks.
What to do instead
Practise reading kana aloud with even mora timing, keeping every vowel full: あ-り-が-と-う as five clean beats. Use pitch-accent dictionaries or audio shadowing to copy a model speaker's rise and fall rather than inventing stress, and record yourself, then compare against the model for vowel length and final です/ます clarity.
Answering a speaking prompt in single words or fixed phrases (はい、すきです) without development, so the response cannot demonstrate the range of grammar and vocabulary the task requires.
Candidates optimise for accuracy over range and fear making mistakes, so they keep utterances minimal. At Level 1 the criteria reward attempting varied structures (past tense, negatives, reasons, connectives) even with minor errors.
What to do instead
Use a 'sentence-plus' rule: every answer gets a reason or extra detail (〜から, そして, でも, 〜たことがあります). Prepare three reusable patterns per topic (opinion + reason + example) and practise extending a one-word answer into three sentences on the spot, e.g. すきです → すきです。なぜなら おいしいからです。まいにち たべます。
Reading a Japanese passage by translating word-by-word into English and answering from the English gloss, so particles and sentence-final forms that carry the meaning get lost (e.g. は vs が, に vs で, passive or causative endings).
Candidates with limited vocabulary default to lexical decoding and treat particles as decoration; English word order habits also lead them to attach modifiers to the wrong noun, especially with long noun phrases before の or が.
What to do instead
Read kana-only or lightly glossed short texts and mark the particle function in each clause before translating (は = topic, が = subject, に = time/target, で = location/means, を = object). Bracket long modifiers: [きのう かった] 本. Answer the question in Japanese first, then check it against the English options rather than translating the whole passage.
Confusing katakana loanwords with similar-looking kana or with English meanings, and misreading small ャ/ュ/ョ and long-vowel marks (e.g. パーティー, コーヒー, テレビ vs デパート), producing wrong answers in reading tasks about signs and menus.
Katakana is introduced later and practised less, loanwords are assumed to be 'easy English', and the small kana plus ー are visually subtle, so candidates guess from the first two characters.
What to do instead
Drill katakana as its own script: read loanword lists aloud weekly, cover the English and reconstruct meaning from kana alone. Write out minimal pairs (ビル/ビール, パン/パーティー, ショップ/ショッピング) and practise reading authentic signs, menus and packaging where katakana appears in context.
Writing romaji, or mixing scripts inconsistently, and leaving particles as romaji or as spaces (e.g. 'watashi wa gakkou ni ikimasu' or 'わたし は がっこう に いきます'), and using English punctuation and spacing conventions.
Candidates compose in English first and transliterate, and because their kana recall is slow they fall back on romaji to keep the sentence flowing; they have not internalised that particles attach directly to the preceding word with no space.
What to do instead
Write by hand in kana only, with a rule that romaji is banned from drafts. Practise copying and then reconstructing short model sentences so particle attachment (わたしはがっこうにいきます) becomes muscle memory, and check every piece of writing for script consistency, correct ー and っ, and Japanese punctuation (。、) before submitting.
Writing an English sentence and translating it structure-for-structure into Japanese, producing word order, tense and topic-marker errors (e.g. placing the verb early, using です with an action verb, or omitting the past tense because English context implies it).
Composition under exam pressure reverts to the first language; candidates have not yet automatised the SOV frame and Japanese tense/aspect, and they assume one-to-one correspondence between English and Japanese words.
What to do instead
Before writing, sketch the Japanese skeleton in order: time → place → object → verb (あした がっこうで 日本語を べんきょうします). Build sentences from memorised frames rather than translating, check the final verb for tense and politeness, and keep a personal error log of the three mistakes you make most (e.g. です after action verbs, missing を, wrong past form) and proofread specifically for those.