The correct answers are I bought a umbrella, She found an unique gift, and He is a honest salesperson.
"A umbrella" is wrong—"umbrella" starts with a vowel sound /ʌ/, so use an umbrella. "An unique" is wrong—"unique" starts with /juː/ (a consonant sound), so use a unique. "A honest" is wrong—"honest" has a silent h, starting with /ɒ/, so use an honest.
Article
A/an vs the vs no article: the three-way choice that trips up learners whose first language has no articles (Russian, Japanese, Korean, Mandarin). Each option changes meaning — I saw a dog (any dog) vs I saw the dog (the specific one) vs Dogs are loyal (the species).
Articles are determinatives that mark noun specificity. A/an = indefinite, first mention. The = definite, known referent. Zero article = generic or uncountable.
Diagnostic: ask does the listener already know which one? Yes → the. No, and it's countable singular → a/an. Generic or uncountable → zero article.
A2 | Elementary | Pre-intermediate
A2 vs B1: A2 handles routine transactions and simple past narration. B1 handles connected discourse, explaining reasons, and understanding main points in clear standard speech. If you can tell what happened but not why it matters, you're still A2.
A2 is the elementary level of the CEFR: past simple, present perfect, first conditional, basic modals, and routine communication about familiar topics.
Diagnostic: can you link ideas with because, although, so that and hold a conversation beyond scripted topics? No → A2. Yes → moving into B1.
Medium
Medium vs Easy: Easy has one obviously correct answer and clearly wrong distractors. Medium has one correct answer but plausible distractors — you need to actually know the rule, not just guess from sound.
The Medium tag filters for A2–B1 challenges with realistic difficulty: one rule per question, plausible alternatives, everyday contexts.
Diagnostic: if you're scoring 90%+ on Easy, move here. If you're below 60% on Medium, go back to Easy for that topic. Target 70–80% accuracy for maximum learning.