Vocabulary size
How many words does the average person know?
A typical 20-year-old native speaker of American English knows about 42,000 dictionary words (lemmas), which come from about 11,100 word families. That is the estimate of Brysbaert and colleagues (2016), based on tests taken by 221,268 people; studies that count words differently arrive at anything from about 10,000 to over 70,000.
Key points
- The median 20-year-old native speaker of American English recognises 42,000 lemmas, from 11,100 word families. The lowest 5% know about 27,000 lemmas and the highest 5% about 52,000 (Brysbaert et al., 2016).
- The same vocabulary is 11,100 word families, 42,000 lemmas or roughly 71,400 word forms, depending on what counts as a word. That is the main reason published estimates run from under 10,000 to over 200,000.
- Vocabulary keeps growing in adulthood: the median 60-year-old in the same study knew 48,200 lemmas, about one new lemma every two days after the age of 20.
- Smaller counting units give smaller numbers: 17,200 base words on average for 20 native speakers in an earlier test (Goulden, Nation & Read, 1990), and about 10,000 word families for students entering three British universities (Milton & Treffers-Daller, 2013).
- Word counts of books are higher still because they count forms and names: War and Peace in English uses 17,996 different word forms, by our count.
How many words does the average native English speaker know?
About 42,000 lemmas at the age of 20. Marc Brysbaert, Michaël Stevens, Paweł Mandera and Emmanuel Keuleers (2016) put an online test in front of English speakers and analysed 265,346 sessions from 221,268 people; participants were asked whether English was their native language, and the estimates refer to native speakers of American English. Each session showed 67 real words, drawn at random from a list of 61,800 lemmas, mixed with 33 invented non-words. Participants marked the words they knew, and each score was corrected for guessing by subtracting the share of non-words they claimed to know.
The median 20-year-old knew 68.0% of the list, which is 42,000 lemmas. Grouped into word families, the same knowledge comes to 11,100 families. The authors add about 4,200 multiword expressions whose meaning cannot be worked out from their parts, such as idioms, which would raise the total by roughly 10%.
Two limits come with the figure. The test asks whether you know a word, not whether you can define or use it, so the authors describe their numbers as upper estimates that measure breadth rather than depth. And it describes American English speakers who chose to take an online vocabulary test, not a random sample of the population.
Why do estimates range from 10,000 to over 200,000?
Mostly because studies count different things. Brysbaert and colleagues note that estimates in the literature go from less than 10,000 to over 200,000 words, and that the highest ones come from counting every dictionary entry, including proper nouns, inflected forms, derived words and compounds. The same person's vocabulary looks very different in each unit:
| Unit | What counts as one word | Words known |
|---|---|---|
| Word families | A base word with its related forms: happy, happier, happiness, unhappy | 11,100 |
| Lemmas | A dictionary headword with its inflections: happy, happier, happiest; happiness is a separate lemma | 42,000 |
| Word forms | Every spelling separately: happy and happier are two words | about 71,400 |
Two more choices move the totals. Names of people and places are excluded from most research counts; Brysbaert and colleagues argue they should be left out of the words known in a language. And multiword expressions such as kick the bucket are left out of all three units above.
So a headline figure means little until you know its unit. A claim of 11,000 words, 42,000 words and 71,000 words can all describe the same person.
What do the main studies find?
Each study below tested native speakers of English or another language, but with a different method and counting unit, so the numbers in the last column should not be set against each other directly.
| Study | Who was tested | Unit | Estimate |
|---|---|---|---|
| Goulden, Nation & Read, 1990 | 20 native speakers of English, tested on a sample of base words from Webster's Third New International Dictionary | Base words | 17,200 on average (13,200–20,700) |
| Nation & Waring, 1997 | Summary of earlier research; a rough rule of thumb | Word families | about 20,000 for a university graduate |
| Milton & Treffers-Daller, 2013 | Undergraduates at three British universities, with the test of Goulden et al. | Word families | about 10,000 on arrival at university |
| Brysbaert et al., 2016 | 221,268 participants, online; estimates for native speakers of American English | Lemmas; word families | 42,000 lemmas; 11,100 families (age 20) |
| Segbers & Schroeder, 2017 | German children in grades 1–8 and young adults aged 18–25 | Lemmas (German) | 73,000 for adults |
The English studies agree more than the numbers suggest. Goulden, Nation and Read's 17,200 base words and Brysbaert's 11,100 are both word-family counts; Brysbaert and colleagues put the difference down to Goulden et al.'s looser criteria for what belongs to one family. Milton and Treffers-Daller's 10,000 families for first-year students is close to Brysbaert's 11,100 for 20-year-olds.
How does vocabulary size change with age?
It rises quickly in childhood and keeps rising slowly through adult life. Nation and Waring (1997) summarised the research as a child starting school with around 4,000–5,000 word families and adding roughly 1,000 families a year, which they call a conservative and very rough rule of thumb. Adult growth is slower:
- 42,000
- lemmas known by the median 20-year-old native speaker of American English (Brysbaert et al., 2016)
- 48,200
- lemmas known by the median 60-year-old in the same study, or 13,400 word families
- 6,000
- extra lemmas between 20 and 60, about one new lemma every two days
- 5,900 to 73,000
- German lemmas, from the average first-grader to the average young adult (Segbers & Schroeder, 2017)
In Brysbaert's data the median kept increasing up to the age of 70 among the people who took the test. Counted as word forms rather than lemmas, the growth looks much faster: by the authors' calculation, reaching about 71,400 forms by the age of 20 amounts to learning nearly 11 new words a day.
How much do people differ?
A lot. Among 20-year-olds in Brysbaert's study, the lowest 5% knew about 27,100 lemmas and the highest 5% about 51,700; in word families, from 6,100 to 14,900. That is nearly two to one in lemmas and more than two to one in word families between people of the same age and language.
Education makes a substantial difference, and the authors found that most of the gap between education levels seems to open up during the years of study, which they link to reading and studying. In a Dutch version of the test taken by nearly 300,000 people, age, education and knowing other languages were the factors most strongly related to vocabulary size (Keuleers et al., 2015).
How many words does the average person actively use?
Fewer than they recognise. All the figures on this page measure receptive vocabulary: words a person understands when reading or hearing them. Brysbaert and colleagues note that productive vocabulary, the words a person uses in speech and writing, is estimated at less than half of receptive vocabulary. Precise figures for everyday use are hard to pin down, because the words a person happens to use in a recording or a set of texts depend on its length and topic.
Do speakers of other languages know more words?
Not in any way current studies can show. The German estimate of 73,000 lemmas for young adults (Segbers & Schroeder, 2017) is higher than the English 42,000, but it comes from a different method: the authors built virtual vocabularies of known sizes from a German corpus, had them "take" the vocabulary test, and used the link between test score and vocabulary size to estimate the size of real participants' vocabularies. German also writes compounds as one word, so a German lemma list contains many items that English would write as two words. The two numbers measure different things in different ways.
For learners the same caution applies to word counts: in languages with many endings, such as Spanish, Italian or German, one lemma has many more spellings than in English, so counts of word forms rise faster. Our counts of 117 books show German, Spanish and Italian books using roughly a quarter more different forms per 5,000 running words than English ones.
How do word counts in books compare?
Book counts usually measure word forms with names included, so they come out high next to research figures in families or lemmas. By our counts:
| Text | Different word forms |
|---|---|
| War and Peace, Maude translation | 17,996 |
| Pride and Prejudice | 6,407 |
| The Great Gatsby | 5,940 |
| Median English book in our set | 7,226 |
War and Peace alone uses more different forms than the median 20-year-old knows word families, yet native readers manage it without a dictionary. Two things explain the gap: every inflection and every name is a separate form, and 6,105 of the forms appear only once in its 566,513 running words, so a reader who skips them loses little. The comparison only works within one unit, which is why we report our book data as forms and do not convert them to families.
How does this compare with what a learner needs?
A learner can read and follow speech well with much less than a native speaker's vocabulary. Paul Nation estimated that knowing 98% of the running words in a novel takes about 8,000–9,000 word families, and following unscripted speech about 6,000–7,000 (Nation, 2006). Nation's families are defined differently from Brysbaert's, so the figures are only roughly comparable, but both are below the 17,000–20,000 base words or families usually quoted for educated adult native speakers. What these thresholds mean and how they were measured is explained on how many words you need in a foreign language.
How can you estimate your own vocabulary size?
With a test that samples words from frequency bands and scales up your score. For English, the Vocabulary Size Test by Nation and Beglar reports a size in word families; free versions and an online test are listed on our guide to vocabulary size. Whatever test you use, check which unit it reports before comparing your result with a figure from this page: a score in families and a score in lemmas for the same person can differ by a factor of three or four, as in the table above.
References
- Brysbaert, M., Stevens, M., Mandera, P., & Keuleers, E. (2016). How many words do we know? Practical estimates of vocabulary size dependent on word definition, the degree of language input and the participant's age. Frontiers in Psychology, 7, 1116. doi:10.3389/fpsyg.2016.01116
- Goulden, R., Nation, P., & Read, J. (1990). How large can a receptive vocabulary be? Applied Linguistics, 11(4), 341–363. doi:10.1093/applin/11.4.341
- Keuleers, E., Stevens, M., Mandera, P., & Brysbaert, M. (2015). Word knowledge in the crowd: Measuring vocabulary size and word prevalence in a massive online experiment. Quarterly Journal of Experimental Psychology, 68(8), 1665–1692. doi:10.1080/17470218.2015.1022560
- Milton, J., & Treffers-Daller, J. (2013). Vocabulary size revisited: The link between vocabulary size and academic achievement. Applied Linguistics Review, 4(1), 151–172. doi:10.1515/applirev-2013-0007
- Nation, I. S. P. (2006). How large a vocabulary is needed for reading and listening? Canadian Modern Language Review, 63(1), 59–82. doi:10.3138/cmlr.63.1.59
- Nation, P., & Waring, R. (1997). Vocabulary size, text coverage and word lists. In N. Schmitt & M. McCarthy (Eds.), Vocabulary: Description, acquisition and pedagogy (pp. 6–19). Cambridge University Press. Text at lextutor.ca
- Segbers, J., & Schroeder, S. (2017). How many words do children know? A corpus-based estimation of children's total vocabulary size. Language Testing, 34(3), 297–320. doi:10.1177/0265532216641152