Duolingo English Test flashcards
Duolingo English Test · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Read and Complete
- A literacy task presenting a short passage with letters missing from some words; you type each full word, using grammar, collocation, and context to infer it.
- Read and Select
- A vocabulary task that lists real English words alongside invented ones; you mark only the genuine words, judging by recognized usage rather than plausible spelling.
- Duolingo score range
- Scores run from 10 to 160 and are reported in 5-point steps, together with four subscores: Literacy, Comprehension, Conversation, and Production.
- Read then Write
- A literacy task in which you read a short passage or prompt and then type a response under time pressure, scored on relevance, grammar, and vocabulary range.
- Adaptive item difficulty
- Item difficulty shifts during the test: correct answers trigger harder items and errors trigger easier ones, so the score reflects the hardest level reached.
- Score validity period
- Duolingo English Test results remain valid for two years from the test date, and scores are usually released within about two days of testing.
- CEFR score mapping
- Approximate conversion of Duolingo bands to CEFR levels: 10-55 is A1-A2, 60-95 is B1, 100-125 is B2, 130-150 is C1, 155-160 is C2.
- Literacy subscore sources
- The Literacy subscore is built only from reading-and-writing tasks such as Read and Complete, Read and Select, and Read then Write, not from the entire test.
- Listen and Select
- A listening task that plays a series of English words once each; you tick only the real words and reject invented ones shaped like English.
- Listen and Type
- A dictation task in which you hear a sentence and type it exactly; scoring compares your transcription with the original for spelling, capitalization, and order.
- Partial credit transcription
- In dictation tasks, correctly typed words earn partial credit even when the sentence is incomplete, so writing the certain words beats an empty box.
- Accent variation
- Listening items use several native accents at natural conversational speed, so recognition must work across unfamiliar pronunciation rather than one textbook voice.
- Comprehension subscore sources
- The Comprehension subscore draws on listening-and-reading tasks, chiefly Listen and Select and Listen and Type, together with reading items such as Read and Select.
- Limited audio replays
- Most listening prompts play a fixed, small number of times, commonly once, with no replay control, so answers must be formed in real time.
- Nonword distractors
- Invented answer options are constructed to obey English sound and spelling rules, so plausibility is not evidence; only recognized meaning and usage count.
- Implicit reading comprehension
- Reading is measured through short cloze and word-selection items instead of long passages with questions, so inference happens in brief bursts.
- Listen, Then Speak
- A conversation task where a spoken prompt is heard a limited number of times, then answered aloud within a fixed window; scoring weighs relevance, fluency, grammar, and vocabulary.
- Interactive Listening
- A simulated dialogue task in which you choose the contextually appropriate reply and may record a spoken follow-up, testing turn-taking and social appropriateness.
- Fluency scoring
- Automated speech scoring penalizes long silent pauses, filled pauses, and restarts, so a steady pace with simpler wording can outscore halting complex speech.
- Register choice
- Conversation items test whether a reply fits the situation: informal wording suits a friend, while polite forms suit a professor or official.
- Speaking sample
- A recorded video interview at the end of the test; it is not scored numerically but is sent to institutions and can still influence decisions.
- Conversation subscore sources
- The Conversation subscore comes from listening-and-speaking tasks, mainly Listen, Then Speak and Interactive Listening, which pair comprehension with a spoken response.
- Off-topic speaking penalty
- A fluent answer that ignores the question scores low, because relevance to the prompt is rated before length, sophistication, or delivery.
- No human interviewer
- All speaking is recorded and machine-scored, so there is no interlocutor to repeat the question, clarify wording, or ask follow-up questions.
- Write About the Photo
- A one-minute writing task: describe what is visible using concrete nouns, spatial prepositions, and present-tense verbs rather than guesses about feelings.
- Speak About the Photo
- A short preparation window of roughly thirty seconds followed by a recorded description of an image, rewarding clear naming of objects, positions, and actions.
- Read Aloud
- You read a printed sentence aloud; omitting, adding, or substituting words and mispronouncing them lowers the score, so exact reproduction of the text matters.
- Production subscore sources
- The Production subscore reflects writing-and-speaking output tasks, including Write About the Photo, Speak About the Photo, Read Aloud, and the writing sample.
- Writing sample
- A short unscored essay at the end of the test that travels with the score report; it must be your own work and is screened for plagiarism.
- Lexical range scoring
- Repeating the same common words caps the vocabulary score, while varied and precise word choice across sentences raises it.
- Coherence markers
- Explicit connectors such as 'however', 'because', and 'as a result' signal logical structure and are rewarded in both written and spoken production.
- Response length tradeoff
- In timed production tasks, a few complete accurate sentences outscore a longer fragmentary answer, because grammatical accuracy is weighted alongside length.