How can I estimate how many words I know in a language?

Sample from a frequency-ranked word list rather than trying to check every word. A good test draws words from across the frequency spectrum - some very common, some mid-frequency, some rare - and asks whether you know each. The proportion you know at each frequency band, multiplied across the size of the list, gives an estimate of your total. Because word frequency follows a steep curve, knowing your cut-off point (the rank beyond which you mostly do not know words) tells you your approximate vocabulary size.

What is a good vocabulary size for a language?

Rough guideposts for an English speaker: around 1,000 word families covers about 80% of everyday speech and supports basic conversation; 3,000 gets you to comfortable intermediate reading and listening; 5,000 approaches upper-intermediate independence; 8,000-10,000 is where you handle most native material with occasional lookups. Educated native speakers know somewhere in the range of 15,000-20,000 word families. You do not need a native-sized vocabulary to be fluent - the high-frequency core does most of the work.

Why should I test my vocabulary at all?

Two reasons. First, the estimate is practically useful: it tells you which graded-reader tier matches your level and which frequency band to drill next, replacing guesswork with a target. Second, and more importantly, the act of testing strengthens memory through the testing effect, so a vocabulary quiz doubles as a study session. Retrieval practice is one of the most effective uses of study time, which means testing your vocabulary is productive even when you do not care about the score.

How accurate are online vocabulary size tests?

They give a reasonable ballpark, not a precise count. The main sources of error are guessing (you may mark a word as known on a vague hunch), the yes-no format (recognising a word is not the same as being able to use it), and sampling noise from testing only a fraction of the list. Treat the result as accurate to within a band rather than an exact figure. Used consistently over time on the same test, it is a good relative measure of progress even if the absolute number is fuzzy.

What is the difference between passive and active vocabulary?

Passive (or receptive) vocabulary is the words you understand when you read or hear them. Active (or productive) vocabulary is the smaller set you can retrieve and use yourself when speaking or writing. Your passive vocabulary is always larger, often two to three times larger. Most vocabulary size tests measure the passive figure, since they ask whether you recognise a word rather than whether you can produce it. Both matter, but the gap between them is why you can follow a conversation you could not have produced.

How to Test Your Vocabulary Size

You can get a solid estimate of how many words you know in a language without checking every word in the dictionary. The trick is sampling from a frequency-ranked list: because words follow a steep frequency curve, you only need to test a spread of words from common to rare, find the point where your knowledge drops off, and extrapolate. The resulting number is an estimate rather than a certificate, but it is a genuinely useful one, and the act of testing yourself to get it is itself one of the best things you can do for your memory. This article explains how vocabulary size tests work, what the number means, and why the test is worth taking even when you do not care about the score.

How do you estimate vocabulary size without counting every word?

Testing every word you might know would take forever, and you do not need to. Vocabulary size tests exploit a statistical property of language: word frequency follows a steep, predictable curve. A handful of words appear constantly, a few thousand appear regularly, and the vast majority appear rarely. Learners acquire words roughly in frequency order, common ones first, so your knowledge tends to be near-complete at the top of the list and to thin out as words get rarer.

That regularity is what makes sampling work. A good test draws words from across the frequency spectrum, some from the top 500, some from the 2,000-3,000 band, some from the rare tail, and asks whether you know each. It then measures the proportion you know at each band and multiplies across the size of the list. If you know 95% of the top-1000 words, 60% of the words around rank 3,000, and almost none past rank 6,000, the test can estimate your total vocabulary from the shape of that decline. Your cut-off point, the rank where you stop knowing words reliably, is roughly your vocabulary size, because everything below that rank you mostly know and everything above it you mostly do not.

The vocabulary quiz on this site does exactly this: it samples words from the frequency lists that power the Kilo Lingo curriculum and gives you an estimate of where your knowledge tails off. Because it draws from the same frequency data as the lessons and graded readers, the number it returns maps straight onto which tier to read and which band to drill next, which is far more useful than a generic score. Take the vocabulary quiz to turn a vague sense of your level into an actual figure and a clear next target.

What does the vocabulary size number actually mean?

A vocabulary count is only meaningful against some rough guideposts. For an English speaker learning a European language, the useful landmarks are:

Around 1,000 word families: covers roughly 80% of everyday speech, enough for basic, slow conversation.
Around 3,000: comfortable intermediate reading and listening; you follow the gist of most non-technical material.
Around 5,000: approaching upper-intermediate independence; native media becomes accessible with manageable lookups.
8,000-10,000: you handle most native content with only occasional unknown words.
15,000-20,000: the range of an educated native speaker.

Two honest caveats. First, these are word families, not raw word forms: a family groups a base word with its obvious inflections and derivations, so "run", "runs", "running" count once. Different tests count differently, so compare like with like. Second, the number you get is almost certainly your passive vocabulary, the words you recognise, which is always larger than your active vocabulary, the words you can produce on demand. Your speaking vocabulary might be a third to a half of your recognition vocabulary, which is why you can follow a film you could never have scripted.

The key insight the number reinforces is that you do not need a native-sized vocabulary to be fluent. The high-frequency core does the overwhelming majority of the work, which is why a well-chosen 3,000 words takes you much further than 3,000 words picked at random.

Why does testing your vocabulary help you learn?

Here is the part most people miss. A vocabulary test is not just a measurement, it is a study session, because retrieving a word from memory strengthens it. This is the testing effect: the well-replicated finding that being tested on material builds retention more effectively than re-reading it. Every word the quiz puts in front of you is a word you have to pull from memory, and that retrieval reinforces the memory whether you get it right or wrong.

So the real pitch for testing your vocabulary is not the score at the end. It is that self-quizzing is among the highest-yield uses of study time, and the estimate is a free by-product. A learner who regularly tests themselves on their vocabulary is both measuring and building it at the same time, which is a rare two-for-one in language learning. Re-reading a word list, by contrast, produces a comforting feeling of familiarity and very little durable memory.

There is a practical payoff too. Knowing roughly where your knowledge tails off tells you exactly which frequency band to work on next. If you reliably know words to about rank 2,000 and then fall off a cliff, your next drilling target is the 2,000-3,000 band, and your reading level is graded readers at the top-1000 to top-2000 tier. That turns a vague "I should learn more words" into a specific plan.

How accurate are these tests, and how should I use the result?

Treat the figure as accurate to within a band, not to the exact word. Three things blur it. Guessing: a yes-no test lets you mark a word as known on a hunch, inflating the score, which is why the better tests mix in plausible non-words to catch over-claiming. Recognition versus use: the test measures whether you know a word when you see it, not whether you could produce it. Sampling noise: testing a fraction of the list introduces variance, so two runs will differ a little.

None of that makes the test useless. Used consistently over time, the same test is a reliable relative measure: if your estimate climbs from 1,500 to 2,500 over three months, that trend is real even if neither absolute number is exact. Use the figure to choose your reading tier and your next drilling band, retake it every month or two to track progress, and do not treat it as a verdict on your ability. It is a map, not a mark.

Frequently asked questions

The full answers to the common questions on vocabulary testing are collected in the FAQ block above. In short: you estimate size by sampling a frequency-ranked list and finding where your knowledge tails off, useful landmarks run from 1,000 words for basic conversation to 5,000 for upper-intermediate independence, testing is worth doing because it strengthens memory as well as measuring it, online tests are ballpark-accurate rather than exact, and your passive vocabulary is always larger than your active one.

Cross-references

The vocabulary quiz estimates your size from the site's frequency lists.
Active recall vocabulary drills explains the testing effect that makes self-quizzing so effective.
Why the first 1,000 words matter covers the frequency curve that vocabulary tests exploit.
The word drill and spaced-repetition flashcards are the tools for closing the gaps a test reveals.