Hokkaido Entrance English common mistakes
Hokkaido Entrance English · Understand the traps and how to avoid them.
When translating a paragraph, the candidate replaces each English sentence with a Japanese sentence of the same shape, keeping English word order and rendering every noun as a literal dictionary equivalent, so the Japanese reads like a calque.
Under time pressure the candidate treats translation as a word-substitution exercise and never re-plans the sentence around Japanese predicate-final structure; school vocabulary lists reinforce one-to-one mapping.
What to do instead
Before writing, read the whole paragraph and underline the main verb of each clause; draft each Japanese sentence by deciding the final predicate first, then attach the subject and modifiers to it. After drafting, reread only the Japanese and rewrite any clause whose particle use or word order would not occur naturally in a Japanese expository text.
The candidate translates a pronoun or a deictic expression (it, this, they, the former) by its default Japanese equivalent without checking what it refers to in the preceding text, producing a translation where the referent is wrong or ambiguous.
Pronoun resolution requires holding the previous sentence in working memory while producing output; candidates translate sentence by sentence and do not look back, and English pronouns have no single Japanese equivalent.
What to do instead
On the first reading, draw arrows from every pronoun and demonstrative to its antecedent and mark whether the antecedent is a person, thing, or whole clause. In the translation, replace the pronoun with the concrete noun or with 前者／後者／そのこと as appropriate rather than leaving a bare それ.
In summary writing, the candidate copies or lightly rewords the first and last sentences of the passage and stitches them together, so the summary omits the argument's middle steps and contains no statement of the author's main claim.
Topic sentences are the most visually salient part of an English paragraph, and copying feels safe because the wording is guaranteed correct; the candidate has no procedure for extracting claims.
What to do instead
Before writing, produce a three-line skeleton in note form: the topic, the author's claim about it, and the two or three reasons given. Then write the summary from the skeleton only, with the source face down, and check afterwards that each skeleton line appears as one sentence in the output.
The candidate's English composition uses correct short sentences but no explicit linking between them, so each sentence starts a new topic and the paragraph has no visible line of reasoning.
Japanese academic writing often relies on implicit connection through repeated nouns and context, so the candidate does not feel the absence of connectors; explicit English discourse markers are also weakly taught.
What to do instead
Build a connector inventory by function (addition, contrast, cause, concession, example, conclusion) and force at least one contrast and one causal link into every opinion paragraph. Read the draft backwards sentence by sentence, asking what logical relation each sentence bears to the previous one, and insert or correct the marker accordingly.
The candidate answers reading comprehension questions by matching words in the question with words in the passage, choosing the option that repeats the most vocabulary rather than the one that states the correct relation.
Lexical overlap is a fast, low-effort heuristic that works on easier tests; under time pressure the candidate stops at the first option that looks familiar.
What to do instead
For each question, locate the relevant sentence and paraphrase it in the candidate's own words before looking at the options. Then eliminate any option whose content is not licensed by that sentence, and treat heavy word overlap with the passage as a warning sign rather than evidence.
When asked about the author's purpose or tone, the candidate selects an extreme option (完全に否定している, 強く非難している) or an option describing the topic rather than the stance, because the passage contains strong examples.
Candidates conflate the intensity of the examples with the author's stance and do not distinguish hedged claims (may, tends to, in part) from categorical ones.
What to do instead
Circle every hedging or boosting expression in the passage and note whether the author commits fully or partially. Match the option's strength to that evidence: if the author writes tends to, discard options containing always or never, and discard options that merely name the topic instead of stating what the author does with it.
In paragraph-ordering tasks the candidate orders paragraphs by the interest or importance of their content rather than by discourse signals, ignoring pronouns, definite articles, and connector openers that point backwards.
The candidate reads for content and treats cohesive devices as decoration; Japanese paragraph order is also often taught by content logic, so the signals are not noticed.
What to do instead
Transcribe the first and last sentence of every paragraph onto one sheet and mark every anaphoric element: definite article plus noun, this/these plus noun, such, the former, pronouns, and connectors. A paragraph opening with such a reference cannot come first; build the chain from the paragraph that introduces the most entities.
The candidate writes an English composition with Japanese-style organization: a long build-up, the thesis revealed only at the end, and no topic sentence, so the grader cannot find the argument.
The ki-shō-ten-ketsu pattern places the conclusion late and values indirection; the candidate transfers this rhetorical habit into English and is unaware that English academic prose expects the claim up front.
What to do instead
Use a fixed template for every practice essay: sentence one states the position, sentences two to four give one reason each with an example, the final sentence restates the position with a qualification. Practise writing the first sentence as a complete claim that could be disagreed with.
The candidate's composition and translation contain article, number, and tense errors that do not block meaning but accumulate, such as a thesis without an article, information treated as countable, or a general truth in the past tense.
Japanese has no articles and no obligatory number marking, so these features are invisible during self-monitoring; the candidate proofreads for meaning, not for form.
What to do instead
Keep a personal error list from graded work and proofread in targeted passes: one pass reading only for articles before every noun, one for singular versus plural, one for tense consistency across the paragraph. Mark every general statement and confirm it uses the present tense.
The candidate runs out of time on the reading section because they read every passage at the same speed and answer questions in the order printed, leaving translation and composition rushed or blank.
The candidate has never timed a full paper and does not know which sections are score-dense; unfamiliar vocabulary triggers repeated rereading of the same sentence.
What to do instead
Take two timed mock papers and record minutes per section. Allocate time in advance, answer vocabulary and detail questions before inference questions, and set a hard rule to abandon any single question after 90 seconds. During practice, skip unknown words that are not needed for the question and read for clause structure instead.