N1 2011-07 — audit report

70 questions (問題1–13, the non-listening half). Generated from the OCR and correction audit trails.

問題1–12 are corroborated by the official 日本語能力試験公式問題集 第一集 (2012): the text matches and every answer is confirmed by the official 正答表. (第一集 substituted 2010-07's 問題13 for this sitting's, so 問題13 here rests on the re-typesets.)

OCR confidence

Source: retypeset-digital — quality 4/5: Clean retypeset digital text throughout, with clear boxed question numbers and intact tables; however several pages are rendered at noticeably lower resolution (small but still legible text), every content page carries '日语轻松考' header and 'Yuuki Bùi' margin watermarks (not overlapping question text), and ad/QR-code material appears on pages 1, 14, and 15.

Every section was transcribed twice independently (a full-page pass and a double-resolution half-page pass) and reconciled. Of 13 sections, 4 agreed on all content between the two passes (reconciled automatically, differing only in transcriber notes) and 9 had at least one content difference resolved by a third adjudication pass that re-read the page images. 54 leaf-level differences were examined in total.

Residual OCR risks (26):

• 問題1: The printed underline under 逃れた in question 2 appears to end flush against the following 。; both transcribers placed the period outside <u>, which matches the close-up reading, but the boundary is within a pixel or two at this resolution.
• 問題2: The printed digits 1-4 in the instruction line (１・２・３・４) and the per-option numerals are normalized to half-width ASCII by convention; the original print uses full-width forms, so any downstream exact-match against the print must account for this.
• 問題2: Q8's printed stem has a small gap between （　） and 。 in the source layout; transcribed as （　　）。 with no intervening space per the blank convention — judged to be layout spacing, not a printed character.
• 問題3: The source is a third-party Chinese retypeset; even with A, B, and the adjudicator agreeing, the retype itself could deviate from the official 2011-07 exam wording (no typos were spotted in this section).
• 問題3: Q17 option 3 すなおには and Q14 option 2 めずらしい are printed in kana (not 素直に／珍しい); confirmed as printed, but worth noting they reflect the retypeset's choices.
• 問題4: Q24 option 3: the printed underline terminates flush with the right edge of て; judged that い is outside the underline based on both half-renders, but a sub-pixel extension under い cannot be fully excluded at the available resolution.
• 問題4: Q20 option 3 為替レール is transcribed as printed per convention; the official original exam almost certainly reads 為替レート, so downstream consumers should treat this as a retypeset typo, not exam content.
• 問題4: Option-number digits and the instruction's 1・2・3・4 are normalized to half-width ASCII per convention; the printed retypeset width of these digits was not preserved (deliberate).
• 問題5: Q35 option 3's missing final ら is plausibly a typo in the third-party retypeset rather than in the original exam; transcribed as printed per convention, but cross-checking against another source of N1 2011-07 would settle it.
• 問題5: The source is a retypeset, so any typo it shares with both transcriptions (and with my image reading) would be invisible to this audit; only comparison with the original booklet could catch that.
• 問題6: The PDF has no usable text layer (only the watermark extracts), so all adjudication rests on visual reading of the retypeset; the original exam booklet was not available for comparison.
• 問題6: Q40 option 3 思いきゃ is almost certainly a retypeset typo for 思いきや; kept as printed, but any downstream answer-keying should treat it as the 思いきや option.
• 問題6: Q40's actor name 上田秋さん reads oddly (single-kanji given name); it is what is printed, but a truncation by the retypesetter cannot be ruled out.
• 問題7: The colons in the 注 glosses are printed with visible surrounding spacing and could conceivably be half-width colons with spaces rather than full-width ：; both transcribers independently chose ：, consistent with the conventions' example, so it was kept.
• 問題7: Whether the absence of 。 between the （４４） blank and ああいうやり方 is a retypeset omission or intentional (the fill reading as a clause modifying ああいうやり方) cannot be settled from this retypeset alone; transcribed as printed.
• 問題7: A small printed gap after モンダイである。 before （４５-a） was judged to be normal post-period typesetting spacing, not a transcribable space.
• 問題8: The two dash glyphs in the passage (2) citation are normalized to ―― (two U+2015) per convention; the exact codepoints of the printed retypeset glyphs cannot be determined from the image, but the convention makes this deterministic.
• 問題8: Passage (1) phrase 「食器のみならず、種類の豊富さにあらわれている」 reads slightly oddly and could itself be a retypeset deviation from the original exam text, but it is confirmed as printed and transcribed faithfully.
• 問題8: Small-glyph kanji such as 滉 in 江口滉 and 脩 in 吉田脩二 were verified at 2x zoom and look correct, but extremely similar variant glyphs cannot be fully excluded at this print size.
• 問題9: no human/agent spot-check was performed for this section (A/B byte-agreement on content was treated as sufficient)
• 問題10: no human/agent spot-check was performed for this section (A/B byte-agreement on content was treated as sufficient)
• 問題11: no human/agent spot-check was performed for this section (A/B byte-agreement on content was treated as sufficient)
• 問題12: no human/agent spot-check was performed for this section (A/B byte-agreement on content was treated as sufficient)
• 問題13: Parentheses around bare digit labels — （1）（2）（3） in rows ④/⑥ and possibly （土）/（残業あり） — print noticeably narrower than the unambiguous full-width （〜22時）; both transcribers normalized all table parens to full-width （）, which is kept, but the source may genuinely use half-width (1)(2)(3).
• 問題13: The bar between times was transcribed as a single U+2015 ―; the printed bar is roughly 1 em wide, but the exact source glyph (―, —, or a CJK rule) cannot be distinguished in the raster.
• 問題13: Several table cells contain likely retypeset corruption (row ⑤ 切る。並べる; row ⑧ 荷物を造る), faithfully transcribed; anyone using this table to verify answer keys should account for the original exam wording possibly differing.

Source-text quirks the OCR transcribed verbatim (printed typos/recall artifacts; the ones judged to be retype errors were corrected in the layer below) — 28 noted at OCR time:

• 問題3: Question 15: the printed underline covers only もくろんで; いる。 is outside the underline (re-verified by the adjudicator at 6x zoom from the source PDF; consistent with the て-form answer options).
• 問題4: Q20 option 3 prints 「為替レール」 — almost certainly a retypeset typo for 「為替レート」; transcribed as printed (final glyph confirmed as ル at high zoom during adjudication).
• 問題5: Q35 option 3 is printed as「つもりじゃなかったのかっていわれた」, with no final ら, unlike option 2's「っていわれたら」; possibly a dropped ら in the retypeset, but transcribed as printed (confirmed against the double-resolution image).
• 問題5: Q27 prints 癒す without okurigana や (not 癒やす); transcribed as printed.
• 問題6: Question 40, option 3 is printed as 「思いきゃ」 with a small ゃ; the standard expression is 「思いきや」, so this looks like a retypeset typo. Transcribed as printed (adjudicated against a 16x render of the source PDF: the kana after き is clearly smaller and lower than the full-size か in option 4).
• 問題6: Question 40: the actor's name is printed 上田秋さん (given name appears to be the single kanji 秋); transcribed as printed, but a retypeset truncation of a longer name cannot be ruled out.
• 問題7: No punctuation is printed between the 【44】 blank and ああいうやり方は; confirmed at high zoom and transcribed as printed. The fill reads as a clause modifying ああいうやり方, so this may be intentional, but it could also be a retype omission of a 。 present in the original exam.
• 問題7: Katakana spellings モンダイ and ハヤリ in the passage are as printed (authorial style of 天野祐吉), not transcription errors.
• 問題7: Paragraph 3 prints わけでなく (not わけではなく); verified at high zoom and transcribed as printed.
• 問題8: Passage (2): 「理性的にあることができるのは強者だからです」 is transcribed as printed; 「理性的にあること」 is a suspected retypeset typo for 「理性的であること」. Confirmed against the double-resolution image during adjudication.
• 問題9: Passage (2) and question 55 option 3 print 経済状勢 (状勢 rather than the more common 情勢); transcribed as printed.
• 問題9: Passage (2) final paragraph prints 計画を立てたうえに想像力を働かせて — うえに where うえで might be expected; transcribed as printed.
• 問題9: 問題9 covers questions 50-58 as printed (boxed numbers 50-58), matching the expected range.
• 問題9: Passage (2) paragraph 4 and Q55 option 3 both print 経済状勢 (状勢), an uncommon variant of 情勢/状況; transcribed as printed since it is consistent within the section.
• 問題10: Passage paragraph 3 reads「もらった苗を、書物の言葉を植えるのである」— the doubled を looks like an appositive but could be a retypeset slip; transcribed as printed.
• 問題10: The third passage paragraph（「与えられたこの土を耕し…」）is printed flush left without the usual paragraph indent in this retypeset; transcribed as its own paragraph.
• 問題11: Passage A's citation reads（社団法人日本小児科医会　2010年6月18日取得による）with a full-width-space-sized gap between 医会 and 2010年; transcribed with one full-width space. The phrasing 取得による (suggesting a web retrieval date) is unusual for an exam citation but is as printed.
• 問題11: Passage B's citation prints the title in corner brackets（小西行郎「早期教育と脳」による）rather than the 『』 usual for book titles; transcribed as printed.
• 問題11: In passage B paragraph 1 there is a visible gap between 思います。and（中略）; it is consistent with the natural whitespace of adjacent full-width punctuation glyphs, so no explicit space was transcribed, but a deliberate full-width space in the retypeset cannot be ruled out.
• 問題12: Question 68 option 2 prints 表わされる (okurigana variant of 表される); transcribed as printed.
• 問題12: Paragraph 3 contains the sentence fragment ending …とは限らないのだから。 (period directly after だから) — as printed.
• 問題13: Although the instruction says 右のページ, in this retypeset the boxed questions 69 and 70 are printed above the アルバイト情報 table on the same page (p12); numbering matches the expected 69–70.
• 問題13: Row ⑤ 内容 prints 「…肉売り場で商品を切る。並べる、清掃」 — the 。 after 切る mid-list is suspicious (likely 、 in the original); the leading 「生鮮食品・鮮魚・野菜果物・肉売り場」 also reads oddly (the original plausibly had 生鮮食品（鮮魚・野菜果物・肉売り場）). Verified against the double-resolution image and transcribed as printed.
• 問題13: Row ⑧ 内容 prints 「荷物を造る、荷物の出し入れ」 — 造る is a suspected retypeset typo (expect 作る／荷造り or 運ぶ); verified as 造る in the image and transcribed as printed.
• 問題13: Question 69 reads 「同じ時間仕事をした場合」 as printed.
• 問題13: Row ④ 曜日・時間 prints the （1）（2）（3） labels at the ends of lines (毎日（1）／8：00―13：00（2）／15：00―21：00（3）／24：00―6：00), though each label apparently belongs to the following time range; line breaks kept exactly as printed. Row ⑥ instead prints each label on its own line.
• 問題13: Time ranges are printed with a single 1-em horizontal bar between times, transcribed as one U+2015 ―. The colons in times are printed full-width and kept as ： (e.g. 10：00―14：00); verified at high magnification.
• 問題13: Paren widths in the retypeset are visually ambiguous around digits (the （1）（2）（3） labels look narrower than the clearly full-width （〜22時）); all parentheses in the table are transcribed as full-width （）.

Corrections applied (5)

Evidence-based reconstructions of the original exam text. This section is the canonical correction log for the sitting.

• 問4 q20 answers[2]: 「為替レール」 → 「為替レート」
  • 為替レール is a non-word; the intended noun is 為替レート (exchange rate), which fits the context (株価が...と連携して上下している). The 問4 distractor mechanism is the misuse of 連携 (where 連動 would be correct), which is preserved; only the corrupted noun (ト→ル glyph error) is fixed.
  • evidence: clear non-word; 為替レート is standard Japanese for exchange rate; minimal ル→ト correction; distractor (misuse of 連携) untouched [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問6 q40 answers[2]: 「思いきゃ」 → 「思いきや」
  • 思いきゃ (small ゃ) is a non-word; the standard grammar pattern is と思いきや. As a mondai6 sentence-ordering item, the token must be the real form 思いきや. Minimal や→small ゃ retype typo.
  • evidence: clear non-word; と思いきや is a standard N1 grammar pattern; minimal ゃ→や correction [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問8 q47 passage: 「理性的にあることができるのは強者だからです」 → 「理性的であることができるのは強者だからです」
  • 理性的にある is ungrammatical; the adjectival noun 理性的 takes である, not にある. Parallels 正直であること earlier in the same passage. Clear retype artifact (で→に); minimal change.
  • evidence: grammar: na-adjective 理性的 + である (not にある); same passage contains 正直であること as a direct parallel; minimal に→で correction [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問13 q69 passage: 「商品を切る。<br>並べる」 → 「商品を切る、<br>並べる」
  • Row ⑤ 内容 lists job tasks as a 、-separated series (商品を切る、並べる、清掃). The 。 mid-list breaks the enumeration pattern; a 読点 belongs there, not a 句点. The quirks file flags this exact spot as likely 、 in the original.
  • evidence: Category pattern, confidence high. [retype-corrections-pass2 (propose + adversarial verify), 2026-06-20]
• 問13 q69 passage: 「荷物を造る、荷物の出し入れ」 → 「荷物を作る、荷物の出し入れ」
  • Row ⑧ is a moving-assistant (引っ越し) job; 造る (construct/manufacture) is wrong for handling luggage. The intended つくる verb for packing is 作る. 造る/作る is a homophone retype error flagged in the quirks file.
  • evidence: Category homophone, confidence medium. [retype-corrections-pass2 (propose + adversarial verify), 2026-06-20]