Skip to content

How many English words do you know?

Two quick grids. A world of words. Get your vocabulary size, an honest range and player comparisons. About 3 minutes, free, no sign-up.

Go to the test

By the Helibrain teamReviewed by Robert, biophysics PhD candidatePublished Last reviewed

A few minutes. A bigger picture.

Your vocabulary is larger than the words you use every day. This vocabulary size test samples familiar and less familiar English words to estimate the dictionary headwords you understand. You get a rounded number, a range, and comparisons with other players when the sample is large enough.

Explore vocabulary in numbers: age, native speakers and learners, reading habits, learning years and learners by country. Published research is clearly separated from our community's results.

How to test your vocabulary

  1. Explore: tick the words you know in the first grid. “Know” means you could explain at least one meaning to someone. Leave uncertain words unchecked.
  2. Refine: the second grid concentrates near the edge of your vocabulary, with some words from every frequency band.
  3. Discover: see your estimate immediately. The questions about age group, English background and reading come afterwards; all are optional.

Both grids contain made-up words. Guessing does not help: false acceptance lowers the estimate. There is no timer, and you can change a tick before submitting each step. No dictionary needed.

How the estimate works

We count lemmas, roughly dictionary headwords. Inflected forms are excluded by the source filter; words with the same spelling count once. Related derivatives may count separately. A word-family test groups more related words together and therefore produces a different number [1][3].

Our existing English list uses Wiktionary lemmas ranked by frequency in large collections of text. There are seven bands: ranks 1–2,000, 2,001–4,000, 4,001–8,000, 8,001–16,000, 16,001–32,000, 32,001–64,000 and 64,001–74,781. A reviewed sample of 70 candidates per band gives an estimated usable frame of 51,508 lemmas. Band sizes reflect the share retained during that review; they are not an exact census of English.

The first grid locates your approximate edge. To avoid bias from that adaptive choice, only fresh real words in the second grid contribute to the estimate. Within each band, we subtract the made-up-word acceptance rate from the real-word acceptance rate, divide by the share of made-up words rejected, and keep the answer between zero and one. We multiply that corrected share by the band's estimated size and add the bands together.

The 90% model range reflects uncertainty from the small word samples, the shared guessing correction and the original band review. It uses repeated draws from simple probability models, including at the extremes where every real word is ticked or left blank. It does not include every possible error from self-report or the choice of words. A result is marked shaky when false acceptance is too high; selecting everything produces an uninformative range across the whole list.

This is our estimator. LexTALE supports the use of words and pseudowords in a quick lexical test, but its validated percentage score is different from our vocabulary-size estimate [2]. This new two-step test has not yet been independently calibrated.

What your number means

A large study estimated about 42,000 lemmas at age 20 for native American English speakers, with a wide spread: roughly 27,000–52,000 between the 5th and 95th percentiles. Its age findings suggest about 6,000 additional lemmas by age 60 [1]. These are published research references, not results from our players.

The study used a different dictionary and allowed shallower recognition than “could explain a meaning.” Treat its figures as context, not a target, a score conversion or your personal percentile. Learners and native speakers also have different language histories.

Vocabulary is one part of language knowledge. This test does not measure intelligence, health, speaking fluency or writing skill. It asks what you believe you know; it cannot check your definitions. Specialist interests and familiar word parts can change the result.

How player percentiles work

A percentile line tells you the share of eligible players whose estimate was strictly lower than yours. Equal scores do not count as lower. We show a comparison only when its group has enough results.

We compare the current English test version, exclude likely bots and shaky results, and count the first eligible result per known player. Missing bot signals alone are not evidence of automation. Visitors who cannot be recognized may contribute more than one attempt. These are self-selected players, not population norms.

If you add an age group, you can see an age comparison when enough people have answered. Native-speaker comparisons and comparisons with learners from your country follow the same rule. Smaller groups show the labelled research reference instead. Reading and learning-year charts show associations, not proof of cause and effect.

Word sources and licence

Our own reviewed sample uses Wiktionary contributors' English lemmas, via kaikki.org / wiktextract, ranked with wordfreq 3.1.1 by Robyn Speer. We filtered and sampled the data; the resulting word list is shared under Creative Commons Attribution-ShareAlike 4.0, which permits commercial use with attribution and the same licence for adapted data [4][5]. Made-up words were created for this test. No competitor's list, LexTALE items or non-commercial research word norms were used.

Frequently asked questions

How many English words does the average person know?

It depends on how words and knowledge are counted. Brysbaert and colleagues estimated about 42,000 lemmas for a 20-year-old native American English speaker. Word-family counts are smaller because related words are grouped together [1][3]. Our number refers to our own word list and definition of knowing.

Is this vocabulary size test free?

Yes. The test, full result and public statistics are free, with no email or sign-up required.

Can English learners take it?

Yes. The same two-step test adapts to your answers. It estimates words understood, not a language level or qualification. Choose “I learned English as another language” after your result if you want to contribute to learner comparisons.

Why include made-up words?

A plausible-looking invented word can feel familiar. Ticking one helps estimate how freely you are accepting words. We use that rate to correct the real-word responses; too many false positives make the result shaky.

Why does my result change when I try again?

Each run samples different words. With a short test, chance matters, especially near the edge of your vocabulary. The range helps show that uncertainty. Remembering test words can also change later results.

See all 43 free brain tests

Want to work on this daily? The Helibrain app has a free plan and paid plans.

Cite this page: How many English words do you know?. Helibrain. 2026. https://helibrain.com/vocabulary-test. CC BY 4.0.

Common questions

Understand the ability behind this activity

Explore the measurements, age comparisons and evidence for what affects performance.

All brain abilities

How this test works

This test estimates how many words of a language you know. You see real words, from common to rare, mixed with made-up words, and you mark the ones you know. The made-up words correct for guessing: marking them lowers your estimate. In English there are two short grids, and the second one focuses on the edge of your vocabulary. Your result is a rounded word count with a range. Word lists with made-up words have been used to test vocabulary since the 1980s.

Version en-adaptive-1 · Cite this test:

Helibrain (2026). Vocabulary test, ven-adaptive-1. https://helibrain.com/vocabulary-test/methods

Full method →

Sources

  1. Brysbaert M, Stevens M, Mandera P, Keuleers E (2016). How many words do we know? Practical estimates of vocabulary size dependent on word definition, language input and age. Frontiers in Psychology 7:1116. doi.org/10.3389/fpsyg.2016.01116
  2. Lemhöfer K, Broersma M (2012). Introducing LexTALE: a quick and valid Lexical Test for Advanced Learners of English. Behavior Research Methods 44:325–343. doi.org/10.3758/s13428-011-0146-0
  3. Nation ISP, Beglar D (2007). A vocabulary size test. The Language Teacher 31(7):9–13. www.wgtn.ac.nz/lals/resources/paul-nations-resources/paul-nations-publications/publications/documents/2007-Beglar-TLT.pdf
  4. Robyn Speer. wordfreq: word frequencies and data licence. github.com/rspeer/wordfreq#license
  5. Creative Commons. Attribution-ShareAlike 4.0 International. creativecommons.org/licenses/by-sa/4.0/
  6. TestYourVocab. Original methodology: frequency sampling, dictionary choices and a two-step design. Consulted for method history only; no words, text or visuals reused. testyourvocab.com/details

How we made this page: written by the Helibrain team from the studies listed above, and reviewed by Robert (biophysics PhD candidate) on 5 November 2026. We update it when the sources change.