🧠Cerebite

Vocabulary Size Test

Check every word whose meaning you know. Be honest — a few made-up words are hidden in the list to keep everyone truthful.

Your vocabulary is about
0
words

Tap a word to mark it as known. When you're done, press "See my result".

How the estimate works

The Korean lexicon is sampled in frequency bands — from the 1,000 most common words down to extremely rare ones. You see a few random words from each band; the share you know in a band estimates how much of that whole band you know, and adding the bands up gives your total. Checking made-up words lowers the estimate, so honesty pays. Every retake draws a fresh random sample, so your number may vary by a couple of thousand — that's normal for a 2-minute estimate.

The shape of the Korean lexicon

Korean vocabulary comes from three sources, and knowing which one a word belongs to changes how you learn it. Native Korean words cover the body, family, everyday actions and basic qualities. Sino-Korean words, written with syllables that correspond to Chinese characters, dominate abstract, academic, legal and technical language and make up more than half of the entries in a standard dictionary. Loanwords, mostly from English, fill in technology, food and fashion.

The Sino-Korean layer is the one that rewards study, because it is built from recombining morphemes. Once you know that 학 carries the sense of learning, 대학, 학생, 학교, 과학 and 학기 stop being five separate words. This is why Korean vocabulary growth accelerates sharply once a learner passes the beginner stage.

How the estimate is built

The test samples the Korean lexicon in frequency bands rather than showing you everything. You see a small random selection from each band, from the thousand most common words down to genuinely rare ones, and the share you recognise in a band estimates your coverage of that whole band. Adding the bands together produces the total.

The invented words are the correction. Korean is particularly good at producing plausible fakes, because Sino-Korean morphemes recombine so freely that a two-syllable word which does not exist can look entirely reasonable. Checking one of those lowers your estimate, so answering honestly gives a more useful figure than answering generously.

Reading your number

Frequently asked questions

How accurate is this vocabulary test?
It's a statistical estimate from a 2-minute sample, not a full exam — expect a margin of a few thousand words. Taking it several times and averaging gives a steadier number. Research puts adult native speakers at roughly 20,000–35,000 words, which matches what this method reports.
Why are there made-up words in the list?
They're an honesty check. Self-reported tests are easy to inflate by checking everything — but fake words catch that, and each one you check lowers your estimate. If you avoid them all, your result is trustworthy.
What counts as "knowing" a word?
Check it if you could give at least one meaning of the word when you see it. You don't need to spell it from memory or use it actively — recognizing the meaning is the standard used by vocabulary researchers.
Is this test useful for Korean learners, not just native speakers?
Yes. The frequency-band sampling works at any level: beginners will recognize mostly the common bands and get a small but meaningful estimate, while advanced learners reach deep into the rare bands. Retaking it every few months turns the number into a progress tracker for your Korean studies.
How many words do I need for the TOPIK exam?
Study guides commonly cite roughly 1,500–2,000 words for TOPIK I (levels 1–2) and several thousand more — often 5,000 or beyond — for the upper TOPIK II levels. Compare those figures with your estimate here to gauge which level range is realistic before you register.
What's the fastest way to grow my Korean vocabulary?
Learn by frequency: the most common words give the biggest payoff per minute of study, which is exactly how this test's bands are ordered. Combine a daily habit of a handful of new words with lots of input — dramas, webtoons, songs — so each word is met again in context, and review at spaced intervals.