EF SET common mistakes
EF SET · Understand the traps and how to avoid them.
Spending a fixed amount of time on every reading item instead of moving on once the passage and question have been read twice without an answer forming.
Candidates treat each question as equally weighted and fear leaving a blank, so a single hard inference item consumes minutes that belong to several easier items later in the section.
What to do instead
Adopt a two-pass rule: on the first pass answer only items you can resolve in under about 60 seconds, flag the rest, and return to them only after the section end is in sight. Practise with a timer so the cutoff feels routine.
Answering detail questions from memory of the passage's general topic rather than locating and re-reading the exact sentence that decides the answer.
Fluent readers over-trust gist comprehension, and distractors are deliberately written to match the passage's topic while contradicting one specific clause.
What to do instead
For every detail or negative-fact item, physically point to the line that contains the deciding word before choosing. If no single sentence supports the option, the option is wrong.
Treating 'NOT mentioned' and 'EXCEPT' items like ordinary questions, picking the first true-sounding option.
Under time pressure the brain latches onto a statement it recognises from the passage and stops scanning, inverting the task.
What to do instead
Before reading the options, underline the negative word in the stem. Then evaluate each option as true or not true against the passage and select the odd one out; verify by confirming the other three are explicitly supported.
Writing a short answer that restates the prompt or gives an opinion with no supporting example, then stopping.
Candidates treat the task as a comprehension check and underestimate how much development the rater expects within the time limit.
What to do instead
Use a fixed skeleton: direct answer sentence, one reason, one concrete example or detail, one consequence or contrast. Rehearse the skeleton on timed prompts until it fills the word count without planning overhead.
Producing accurate but very short, simple sentences because the candidate self-edits complex grammar out to avoid errors.
Error avoidance feels safe, but the band descriptors reward range as well as accuracy, so safe writing caps the score.
What to do instead
Deliberately include at least two complex structures per response (a subordinate clause, a conditional, a relative clause) and one less common connector, then proofread only for agreement and tense consistency.
Missing the second speaker's turn because attention drops after forming a reply to the first speaker.
Candidates mentally rehearse their answer while the audio continues, and the decisive information in conversational items is usually in the response, not the prompt.
What to do instead
Train a 'listen to the end' habit: during practice, hold a mental placeholder instead of composing, and only summarise after the audio stops. Do exercises where the correct answer is only available in the final turn.
Choosing an option because it contains words heard in the audio rather than because it matches the meaning.
Word-matching is fast and feels confirming, and distractors are built by recycling vocabulary from the recording in a different relationship.
What to do instead
Cover the options while listening, note the gist and key numbers or names, then match meaning to option. Reject any option whose only support is shared vocabulary.
Losing the thread after one unknown word and abandoning the rest of the recording.
Candidates treat unknown vocabulary as a comprehension failure rather than noise, so they stop processing the ongoing stream.
What to do instead
Practise tolerating gaps: on each listen, deliberately let unknown words pass and track only who, what, and the speaker's attitude. Review transcripts afterwards to confirm the gist was still recoverable.
Answering the speaking prompt with a memorised generic script that does not address the specific question asked.
Prepared templates feel fluent and reduce anxiety, but raters penalise responses that are off-topic or clearly rehearsed.
What to do instead
Prepare flexible frames with slots (position, reason, example) rather than full sentences, and practise responding to unfamiliar prompts cold, recording yourself and checking that the first sentence answers the actual question.
Filling pauses with long hesitation or restarting sentences repeatedly instead of using short neutral fillers and self-repair.
Candidates equate fluency with speed and silence with failure, so they restart rather than bridge, which fragments delivery.
What to do instead
Rehearse bridging phrases ('what I mean is', 'let me put that another way') and practise a three-second rule: if a word will not come, rephrase the idea immediately rather than stopping.