Matura English common mistakes
Matura English · Understand the traps and how to avoid them.
In the word-formation gap task, candidates write a semantically plausible but formally wrong derivative, e.g. filling a noun slot with the adjective form (writing "happy" or "happily" where "happiness" is required), or inventing a form that does not exist ("unpossible", "informations").
Candidates translate the Polish base word mentally and pick the first related English word that comes to mind, without checking which part of speech the gap needs (article + ___ + verb = noun; adverb + ___ + noun = adjective) or whether the derived form is real. Polish word formation is far more regular, so the habit of trusting the base word carries over.
What to do instead
Before writing, label the gap's part of speech from its immediate context (determiner, preposition, subject/object position, verb agreement). Then apply the standard affix set for that class (e.g. -ness/-ity/-ance for nouns, -ful/-ous/-ive for adjectives, un-/in-/dis-/ir- for negation) and reject any form you have not actually seen in a text. Build a personal list of high-frequency families (succeed/success/successful/successfully, decide/decision/decisive) and drill the ones where the stem changes.
In the open cloze (one word per gap, no options), candidates insert a word that fits the meaning but not the grammar, e.g. putting "despite" where a conjunction before a clause is needed ("despite it rained"), or "make" instead of "do" in a fixed collocation, or leaving a gap blank rather than attempting it.
Candidates read only the gap and its immediate neighbours, so they miss the structural signal (a following full clause, a following bare infinitive, a preceding modal). They also treat open cloze as a vocabulary test when it is mostly a grammar-and-collocation test, and blanks feel riskier than a guess.
What to do instead
Read the whole sentence, then ask three questions in order: what structure follows the gap (clause / noun phrase / infinitive / -ing), what word class does that structure demand, and which item in that class collocates with the words on both sides. Practise by classifying each gap in past papers as grammar (article, preposition, auxiliary, pronoun, conjunction) or lexis (collocation, phrasal verb) before answering. Never leave a blank: an unattempted gap scores zero with certainty.
In the multiple-choice lexical gap task, candidates choose the option that is closest in Polish translation rather than the one that collocates in English, e.g. selecting "do a decision", "make a photo", "strong rain", or picking "borrow" for "pożyczyć" when the sentence requires "lend".
The distractors are deliberately built as translation equivalents of one Polish word, and Polish makes fewer verb-noun distinctions (one verb for make/do, one for lend/borrow, one for teach/learn). Candidates under time pressure match meaning, not usage.
What to do instead
Learn vocabulary in collocation chunks, never as isolated words: record "make a decision / take a decision", "do homework / make a mistake", "heavy rain / strong wind", "borrow from / lend to". For each confusable pair, memorise the grammatical frame (borrow something from someone; lend something to someone; teach someone something; learn something). When two options seem synonymous, decide on the frame, not the dictionary meaning.
In the transformation task (two to four words, keeping the meaning), candidates change the meaning slightly or break the required structure: they drop a negation, alter the tense or person, use five words where four are allowed, or fail to include the given prompt word unchanged.
Candidates focus on producing a grammatical sentence and stop there, without a final check against the original meaning and the word limit. The prompt word is often a trap (e.g. given "wish", candidates write a conditional instead of the wish structure), and counting words under pressure is skipped.
What to do instead
Use a fixed three-step routine: (1) identify the grammar point signalled by the prompt word ("wish" = regret or unreal present; "used to" = past habit; "so" = result clause or inversion); (2) write the sentence; (3) verify three things - same tense and same subject/object roles as the original, all negations preserved, and word count within the limit with the prompt word unaltered. Count words explicitly, treating contractions as the number of full words they represent in the key.
In listening, candidates answer from the first word or phrase that matches an option, then stop listening, and miss the correction, negation or contrast that follows ("Well, I was going to say Tuesday, but actually Thursday works better").
The distractors are recorded to be said first and then revised, so surface matching feels rewarding. Candidates also lose concentration after a perceived hit and stop tracking the speaker's argument, and they cannot rewind, so the missed correction is gone.
What to do instead
Train the habit of waiting for the speaker's final position: treat the first mention as provisional and listen for discourse markers that signal revision or contrast (but, actually, having said that, on second thoughts, the thing is). Before the recording starts, underline the key word in each question so you know what to listen for; while listening, keep your pen moving and mark options as possible/confirmed. Practise with transcripts: read the script after answering and highlight the exact line that decided each answer.
In listening, candidates mishear minimal pairs and connected speech, writing "thirteen" for "thirty", "can" for "can't", "walked" for "worked", or missing "a" versus "the" and singular versus plural endings.
Polish speakers rely on spelling and on clear vowel/consonant contrasts; English reduces unstressed syllables, uses weak forms, and distinguishes sounds Polish does not (short /ɪ/ vs long /iː/, /æ/ vs /e/, final voiced vs voiceless consonants). Numbers, dates and prices are usually tested precisely because they are easy to confuse.
What to do instead
Do targeted dictation of short audio segments (10–15 seconds) and compare with the transcript, marking every word you missed and why. Drill the specific traps: -teen/-ty stress (thirTEEN vs THIRty), can/can't (weak /kən/ vs /kɑːnt/), -ed endings (/t/, /d/, /ɪd/), and plural/final -s. Learn to predict from context: if the question asks for a price, expect numbers and write digits immediately.
In reading, candidates answer using general knowledge or personal opinion about the topic rather than evidence in the text, or they pick an option that is true in the real world but not stated in the passage; equally common, they mark a statement true because it repeats words from the text while its meaning is actually contradicted.
Multiple-choice and true/false tasks include distractors built from verbatim text words in a wrong relationship, and candidates match words instead of meaning. Prior knowledge about familiar topics (health, technology, environment) makes an unstated answer feel obviously correct.
What to do instead
For every answer, point to the exact line in the text and say aloud what it states. Reject any option you cannot locate in the text, even if you believe it. For true/false/not given items, distinguish "contradicted by the text" (false) from "not mentioned" (not given) - the test rewards this distinction. Underline the sentence you used, and if two options look right, compare them against the sentence word by word to find the altered element (tense, quantity, agent, certainty).
In reading, candidates run out of time because they read the whole passage closely first, then read it again to answer, and they get stuck on one unknown word or one difficult item instead of moving on.
School reading habits reward linear, thorough reading; exam reading rewards locating information. Candidates also believe every word must be understood, so an unknown word stalls them, and they do not budget time across the four or five tasks in the paper.
What to do instead
Use a task-first order: read the questions or headings, underline the key words, then scan the passage for the relevant paragraphs and read only those closely. Never translate every word - infer unknown lexis from word formation and context. Set a per-task time limit (e.g. 12–15 minutes) and enforce it: if an item resists after one minute, mark your best guess and return only if time remains. Practise full papers with a timer to calibrate.
In the formal email or letter, candidates omit or misplace the required content points, or write all points in one unstructured block with no opening, no closing formula and no paragraphing, and they ignore the register demanded by the task (writing "Hi, what's up" to a school principal, or contractions and slang in a formal letter).
Candidates write from the first idea that comes and run out of time before checking the task instructions; Polish letter conventions and the tendency to write one long paragraph carry over. Register is rarely taught explicitly, so candidates do not notice the addressee in the prompt.
What to do instead
Spend two minutes planning: list each required point as a bullet, decide the register from the addressee (formal: Dear Mr/Ms + surname, no contractions, full forms, Yours sincerely; informal: Hi + first name, contractions acceptable), and allocate one paragraph per point. Write a short opening stating the purpose and a closing with a formula. Before handing in, tick each bullet in the task and confirm it appears explicitly in your text.
In the longer written statement (essay/opinion text), candidates produce a list of opinions with no development, no linking devices and no structure, or they write a narrative when the task asks for argumentation; sentences are short and repetitive, and the same basic vocabulary is recycled throughout.
Limited productive range and fear of error push candidates towards safe, simple sentences; they also do not distinguish text types, so an opinion task becomes a story. Linking devices are known passively but not actively, so they never appear in output.
What to do instead
Learn one skeleton per text type and reuse it: opinion text = introduction with thesis, two body paragraphs each with a claim plus justification or example, conclusion restating the position. Memorise a small active set of connectors (however, moreover, whereas, consequently, on the other hand, in conclusion) and force at least one into each paragraph. Build a personal bank of topic vocabulary and collocations for the recurring exam themes (technology, environment, education, health) and deliberately upgrade one basic word per sentence at the editing stage.