Aptis Advanced cheat sheet

Aptis Advanced Reading passages sit at C1–C2: articles, reports, reviews and correspondence written for educated native readers, not simplified learner texts.

Every Reading item is selected-response (multiple choice, matching, ordering or gap-selection); answers are machine-scored against an exact key, so no partial credit exists for a near-miss.

An option that repeats vocabulary verbatim from the passage is usually a trap; the key normally appears as a paraphrase or synonym chain.

Cohesion items (ordering sentences or paragraphs) are solved by tracing reference chains: pronouns, definite articles and linking adverbials must point to something already introduced.

Inference items require locatable evidence: the correct option is implied by clues but never stated word for word, so reject any option you cannot tie to a specific line.

In opinion-matching tasks, attribute a view to a named person only where the text explicitly commits that person to it, not where the writer states it in their own voice.

Gap-fill items must satisfy both meaning and syntax: test the whole clause for collocation, preposition and grammatical form before choosing.

There is no negative marking in Aptis Reading, so an unanswered item is a guaranteed zero and a guess carries no penalty.

Global items (best title, writer's purpose, main idea) are cheapest to answer after the detail questions, once the passage content is already mapped.

Aptis Advanced Listening uses monologues and multi-speaker dialogues delivered at natural speed in a range of accents, including regional British varieties.

Items target gist, specific detail, speaker attitude and inference; the correct option is normally a paraphrase rather than the words actually heard.

Distractors frequently recycle a word from the recording in the wrong context or reassign one speaker's opinion to another speaker.

Listen for signposting: 'however', 'actually' and 'the thing is' typically introduce the correction or qualification that carries the answer.

Numbers, times and prices are common traps: a speaker often states one figure and then revises it, and the revised figure is the keyed answer.

Hedged or counterfactual forms ('I was going to…', 'it might have been…', 'I wish I'd…') signal that the surface meaning is not the intended answer.

Note-completion answers must fit the grammar of the surrounding notes and stay inside the stated word limit, or the item is marked wrong however accurate the idea.

Aptis Advanced Writing is rated by trained examiners on analytic criteria: task fulfilment, coherence and cohesion, lexical range, grammatical range and accuracy.

Each prompt specifies a purpose and a reader; a fluent response that answers a different question scores low on task fulfilment regardless of language quality.

Register is a scored feature: formal correspondence excludes contractions and idioms, while informal messages to friends admit both.

Paragraphing counts: one controlling idea per paragraph, an opening line stating purpose and a closing line that rounds off the message.

Grammatical range is demonstrated by relative, concessive, conditional and passive structures used accurately; errors cost more than simple sentences do.

Falling far below the stated word limit caps task fulfilment because the required points cannot be developed; padding far above it risks relevance penalties.

Fixed frames should be automated: 'I am writing to…', 'I would be grateful if…', 'I look forward to hearing from you'.

A realistic split is roughly five minutes planning, the bulk of the time drafting, and five minutes checking agreement, articles, spelling and punctuation.

Aptis Advanced Speaking responses are recorded and rated by examiners on fluency, pronunciation, grammatical range and accuracy, vocabulary and coherence.

Each task is separately timed; silence scores zero, so an imperfect but continuous answer always outperforms a stalled accurate one.

Pronunciation is judged on intelligibility, not accent: word stress, sentence stress and clear consonant clusters matter more than vowel quality.

Short personal questions are best answered in two or three sentences with a reason or example rather than a single word or bare statement.

Extended turns are structured as point, reason, example, conclusion; this produces the coherence the rating scale rewards inside the time limit.

Range is rewarded, so conditionals, perfect aspect, passives and relative clauses should appear naturally rather than be avoided.

Memorised speeches are penalised: an answer that does not address the specific prompt loses marks for relevance however fluent it sounds.

Discourse markers and brief fillers ('well', 'to be honest', 'let me think') buy planning time and raise fluency ratings compared with hesitation noises.