alt Hacker NewsHow many of the 170k English words do you know?
Comments
Interesting concept, but 100 words is really quite a lot to get through... It's tiresome trudging through the easy words at the start, and I never got to see the interesting words before getting bored.
I've seen other systems like this calibrate far more quickly by assigning a sort of score and confidence behind the scenes. Confidence starts out low and increases over time - correct/incorrect answers rapidly adjust score at the beginning, then things settle down.
In practice this means you get a sequence of increasingly uncommon words initially, until you get one wrong, then you drop back to something easier until you start getting things right again, and eventually circle around words at your level.
Also - too many clicks per word. It's low stakes, just let me click the definition once and I'll live if I misclick (or add an undo button).
In addition to everything everyone else has said: their math is off by half (or 100%, depending on how you count), due to a structural error.
(context: native English speaker, big reader, huge nerd, perfect SAT score)
I got all 100 correct on the first try without looking anything up! Confusingly, that only resulted in a "SCIENTIFIC ESTIMATE" that I know 85,000/~170,000 words?
Their "How is this calculated" page that appears at the end explains their error:
> According to the Oxford English Dictionary (Second Edition), there are approximately 171,476 words in current use.
> We use Stratified Sampling. Instead of testing random words, we divide the language into 5 distinct difficulty bands based on frequency of use:
> 1. Core Basics ~3,000 words > 2. Intermediate ~7,000 words > 3. Advanced ~10,000 words > 4. Expert ~25,000 words > 5. The Obscure ~40,000+ words
> If you answer 2 out of 3 'Intermediate' questions correctly, we estimate you know roughly 66% of the 7,000 words in that band.
> Total Score = Σ (Accuracy in Band × Band Size)
Their strata add up to 85000, not ~170k, making a perfect score still give a 50%.
They're also using a pretty limited and perhaps non-difficulty-representative subset of the language.
Cute, but wrong on many counts.
It is quite easy to cheese the problems: many of them don't look like word definitions ("a sharp pain in the back"), many problem have this "correct answer + opposite meaning + 2 unrelated things" answer structure, and for the second half of the answers, very often the longest answer is the correct one. The wrong options are not well designed here.
The sample of words is also heavily biased towards concepts relating to words, speech, speakers, and/or persuation. They are likely generated by an LLM which is primed on the task of choosing words, and end up choosing words related to "words".
For context, I'm an L2 speaker, linguistic nerd, and I use English mostly in academic/professional settings. I got 75,400 by a combination of the tactics above; in reality it might be closer to 10-15k.
The design is also painfully similar to Duolingo if anyone can spot that.
Pretty fun.
I suggest skipping the submit button and just showing it's correct when pressing and moving on after a sec or so. Having to click on submit twice really breaks the flow.
Also in all the words I tried I noticed out of the 4 options one is the correct one, another is the opposite of the correct one, and the other 2 are random stuff. You can basically skip any option whose antonym isn't present as well.
It should be possible to respond "I don't know". When you really-really don't know, it's unfair to get a 1/4 chance at right anyway, or even better if you use routine multiple-choice tactics.
I got credit for a few that I would have happily just missed.
I have a copy of the shorter Oxford English Dictionary from 1970 which I inherited. It is two massive volumes and is only shorter in comparison to the full dictionary which is 12 volumes (more in more modern editions).
My shorter OED contains 163,000 words (compared to the 600,000 words of the longer).
According to this site I know 71,000 words... Let's test that against the OED. I should have about 43% chance if knowing a word picked at random.
In my totally scientific test (ha) I chose 50 words at random from the OED and discovered I knew 29 of them for a score of 58% which is more than two sigma from 43%, this disproving the hypothesis.
I forgot what that was now, but it was a fun experiment.
It seems like the right answer is usually the longest of the choices, I managed to get a few just by picking the longest. It would also be nice if there was a "I don't know" instead of guessing and skewing the results by getting it right, though maybe thats accounted for
Got 59,800, Performance Breakdown:
Core Basics 19/20
Intermediate 17/20
Advanced 19/20
Expert 14/20
Grandmaster 12/20
I guess, it's not too bad for a non-native speaker.
Minor feedback:
1. The correct answer for "Lethargic" is "Affected by lethargy". I think, definitions should not use words that share common root with the defined word, because:
a. it makes guessing too easy
b. it basically becomes a circular definition which is meaningless
2. Options almost always include 1 correct answer, 1 direct opposite and 2 completely random. Once you learn to recognise it, you can easily rule out 2 random options and have a 50/50 guess.
Not that I want to cheat in such a game, but for many words everything but correct definition is shorter or follow some "dumb rpg text" template.
Like if author used LLM to generate wrong definitions per word instead of actually mixing definitions of words.
Like for me most of more complex words been adjectives with few nouns. And in many cases you can just see 2/4 or 3/4 definitions are not for adjective.
I think it was way too easy to guess corretly based on exluding obviously incorrect choises and then going with vibes.
There were many words I couldn't have explain the meaning of at all, if I wouldn't have had the options, but having the options made it easy. I wouldn't count those correct answers as a part of my vocabulary (even passive), even if I could answer with relative confidence.
English is my third language. My vocabulary has been stuck at an "OK" level because I struggle to actually retain and understand new words.
I built https://segue.app to solve this. It uses illustrations (pictures) and etymology to help with deep understanding and long-term retention so words actually stick. Yeah, it is all AI generated.
If the goal is to actually calculate how many words we know, then you should include an "I don't know" option. Sure, some people will choose to guess to inflate their score, but some of us will be honest because we legitimately want to know our scores.
If you force me to guess, then I'm going to guess. Not only does that give me a 25% chance of getting it right at random, but as others have pointed out, it is very hard to make a multiple choice question that isn't guessable by an astute enough test taker. I think I knew 80 - 85 of those words, but I scored 97, because those questions were very guessable.
Also, reiterating everyone else's comments with respect to the UX needing fewer clicks, and also the definitions not being exact or precise in many cases.
Feature request: fewer clicks. It should be one click per question
That was fun. Bit confused by the result because it says I was "wow are you stephen fry?" Which I assume meant I did decent. (72K).
But then below it said "you are a man of few words".
I take it the latter is just because I've only done the test once? But it's mixed messaging on first attempt I think.
Something is wrong with the estimation method -- I got 100/100 words correct (albeit the second time I did the test -- one word I had gotten wrong the first time occurred again), and it estimated my vocabulary at 85000 words. Given the stated methodology, the correct estimate should be 170000.
Suggestion: Add an "I don't know" button. If I don't know a word, I can admit it - but if I have to guess, then I have a 1/4 chance of getting incorrect credit.
I got 96/100 with minimal guessing. Being a native speaker of a Romance language is a huge advantage here; words like “Quotidian” and “Defenestrate” might be exotic in English, but are almost trivial for an Italian.
I like it a lot, but unfortunately you can cheat a bit: there are always two opposite answers and two unrelated ones. The correct answer is (almost?) always one of the opposites.
78.000 (-2 advanced, -3 grandmaster), pretty good for a second language; the test's maximum appears to be 85.000.
The alternatives to choose between appear to be LLM-generated, you can see several patterns ("now" and "forever" appear a lot).
Years ago, I used to play a similar game that you could keep playing and where you levelled up when you had enough words correct in a row, or down for a single mistake. A fun thing about it was that at very high levels, it got easier for me because they mixed in some old English words which were essentially the same as in Dutch, my native language. There was a charity aspect to it as well, I think it was https://freerice.com/ , but they seem to have simplified the game now.
The university of Ghent (Belgium) also used to have an interesting test which rated your proficiency according to average scores at certain education levels. There I got 41.000 (IIRC), which was rated as average for a university-level native English speaker. An update at the bottom of https://languagehat.com/ghent-vocabulary-test/ discusses where that test went and has a few alternatives. Edit: https://www.myvocab.info/en is pretty similar to this test (found in another comment).
Reading through the comments, I've noticed you can tell the native speakers by their scores in the word categories. A native speaker will score 20/20 in the first two bands and progressively less in the following ones. For those who have learned English as a foreign language, the scores are more evenly distributed.
So it's not uncommon to see a native English speaker totaling 90 as 20,20,19,17,14, and a foreigner reaching the same total as 18,18,18,18,18. Strangely enough, the algorithm favors the latter, because it assigns more weight to the higher-end bands.
Is this of any use? I doubt so, but it was fun.
P.S. of course a more reliable clue of nativeness is the use of "its" and "it's" interchangeably, a mistake EFL learners wouldn't do.
"It's a dead language!" they said, "It's a waste of time!" they said, "It's not like you can talk to dead Romans." they said. WHO IS LAUGHING NOW!?
At first I noticed that for many questions two or three of the answers are obviously wrong. So in many cases the correct answer can be guesses easily. But then I noticed that in 90% of the cases the correct answer is the longest of the four. This makes guessing even easier. The whole thing has a vibe-slopped feel to it.
Interesting choice of words I'd say: as a French person this test is pretty much a test about “how close is the English word to the original French meaning” as the test was almost devoid of obscure words of Germanic origin.
At least I learned a bunch of «faux-amis» in the process.
I got 88 out of 100, but all I learned from that is that I am really good at guessing. For something like 20 of the words I was able to guess by eliminating the options that sounded unlikely and in a few cases just guess from the meaning of parts of the word.
I'd prefer an "I don't know" option just for a more honest assessment of how many words I truly know versus how many words I can guess.
These should maybe be checked through. Many are the second or third definitions, and some even reference the word in the definition e.g Lethargic: exhibiting lethargy
Non-native speaker of English here, got 81k. Mostly with intelligence, not language skills.
Once you figure out the pattern of "one answer sounds like the requested word, two are opposites, one is unrelated", the test suddenly becomes easy. Not all questions follow that pattern, but many of them do.
Sometimes there are two or three answers that sound like the question, sometimes a word that is clearly an adjective relating to a person (ending in -us) has non-adjective definitions. I don't think there's even a single question where more than two of the answers make sense, even if you've never heard the word before. That leaves very little room for mistakes.
A much better test, which dynamically adjusts difficulty level: https://www.myvocab.info/en
A common pattern is the word's true definition and its opposite, plus two mostly unrelated meanings. So, when in doubt, you can improve your changes by picking one of the opposing pair. That's a bit of short-coming.
The 171,476 figure from OED is used inaccurately in a way that shows a gross misunderstanding of dictionaries and language. The number 171,476 refers to the number of full entries for words in “current use” as defined in the 20-volume Second Edition of the Oxford English Dictionary (OED). It does not represent words. It also does not include all the OED's variant spellings, inflected forms, phrases or run-ons (sub-entries derived from the main entries). Additionally, the OED is by no means a complete inventory of English. In fact, it's probably millions of words short, especially as it has an incredibly slow update cycle. Source: I am a dictionary editor and lexicographer, use OED daily, and know the people who make it.
An alternative algorithm which would probably converge faster than 100 questions would be something like Elo or Glicko 2.
A word's "difficulty" would be some function of how rare it is. Once you have a reasonable estimate of the user's "skill" you can infer that a user won't know more difficult words. The benefit of this is you're not spending time asking the user about words they probably know.
Of course it's possible at an individual level, difficulty does not monotonically increase as a function of how rare the word is. A person might be very familiar with a domain-specific subset of English. But the "stratified sampling" approach will also have this problem.
There is a similar problem in chess, where players have ratings which really only change on one dimension. So there can theoretically be a mismatch when puzzles are also scored on a single axis, since a "harder" puzzle that contains a motif a player is familiar with will actually be easier for the player.
I am building in the language learning sector, and this test is almost certainly not accurate (depending on what you want to measure). It's fun and cool though. But basically this is all based on a frequency list, which itself depends on the corpus. I have not been able to find a good corpus of English which is representative of modern spoken English. Spoken english depends on your age range and subculture and and changes every few years. Example: https://observablehq.com/@yurivish/words
Most of the corpuses I've found heavily over-represent newspaper articles and books, obviously. So the frequency ranking is biased towards academic/crime/geopolitics, not spoken english. But even then, it depends what you most commonly speak about!
There's no better way to do it, though. I'm just providing context.
The harder words are trivia questions an educated English native could get. What I mean by this is they're all words that you'd have a chance of knowing for a reason. Things like defenestrate, antidisestablishmentarianism, hippopotomonstrosesquipedaliophobia. I know these words, but these are not words I know because I've ever had cause to use them. Words can get way harder than this and still be actually used, and not strictly only in a scientific sense. I'm thinking things like "Ginnel" (narrow passage between houses) or "Vamp" (a part of a shoe) or "Moraines" (hilly landscape formed by glaciers) or "Lea" (land used to pasture animals)
Usually the longest answer is the correct one.
Also sometimes two options are the opposites of each others. In this case, one of them is correct.
I feel like you can get close to 70/100 with this heuristics, without actually knowing any words.
Nice! Some feedback: The score it shows doesn't really mean anything to me. I think it would be more interesting for the user to know how they rank (perhaps in percentile terms) relative to the overall english-speaking population and/or relative to other users on the site
I just got 72k words, I most definitely do not know that much I just clicked the longest definition for most of them.
One soon discovers that those fancy words are not Ænglish words at all. If you know 6 other languages, you will pass this test 100%.
This app is a great example of what AI does to your brain. No one making their own choices in the app design would make each question need three clicks.
I could actually get almost all of the last third correctly by choosing the option that's the longest, has a semicolon, or a coma.
Aside from that, I didn't like that most of the words only had one or at most two definitions that sounded viable.
A lot of these words have either Latin or Greek origins, for most questions you can deduce the correct answer by asking the question: "Which of these would make sense to develop into a separate word through the mostly non-modern history of the language?".
I would enjoy it way more if all four options sounded equally viable, and I couldn't deduce the correct answer without actually being sure about the meaning of the word. I understand that coming up with choices like that for each question is way harder if you actually validate all of them manually.
I got a score of 76000 best estimate with 85 being correct, even though English is not my native language and I'm not that good at it.
I found a big problem with this - I noticed that the longest answer is very often the correct one, which kinda ruined the game. Even though I didn't want it to, it started affecting my decision-making. Luckily, I only noticed this around question 85, though those are really the tricky ones.
Good news for the project is that I think you can easily tweak the LLM to generate better alternatives.
I got 89/100, which extrapolates to 72,700. As a non-native speaker, I'm quite happy with that.
>Required Reading
>Read the dictionary from A to Z. It's a gripping tale with a terrible plot.
I actually have! I was very bored with the barely-above-"see spot run" books in the classroom at around 8, and we didn't yet have open access to the school library. The dictionary was a better option than all the others I had access to (in class).
Any other dictionary-completionists in here? Regardless of size - I'm fairly sure mine was rather small, though not a pocket-sized one.
Would other people define "complacent" as "Smug satisfaction with oneself"? I'm not so sure.
Regardless, this was fun.
Stuck it out to the end against my better judgement. Got 89/100 due to difficulties at the "Grandmaster" stage (12/20).
I thought it was going to be tougher because the very first word on my run was "Yield" and none of the options seemed convincing to me. I went with something that was at least fairly adjacent to the "something produced by" (as opposed to "submit to") meaning and this did successfully yield (he he) my first point.
there's also https://www.myvocab.info/en
From what I can tell they actually have a bit more robust science behind their algorithm (and a lot less questions to answer)
I think native speakers of Latin derived languages have an advantage given the proposed words in my run. The list was overly biased that way. In fact, many of the advance and grandmaster levels words are basically that. Latin derived words.
At least that was my experience as a native Italian speaker. My English vocabulary is good, but not great by any means and by reading books in English I know that there are plenty of words that are not derived from Latin
Nice one, what I noticed is that out of 4 options 1 wrong is just something looking similar in letters, and 2 options are opposite meaning of each other - so actually the choice is 1 out of 2, not 4.
Also many highest difficulty words are actually combinations of multiple smaller words which makes it easier to guess, I got more right in expert/grandmaster than in advanced.
Even though it said ""Unbelievable. Are you actually Stephen Fry in disguise?" it still estimates I know less than half the English vocabulary. Humbling.
78,500.
The very first one was "Unique". I wondered if "the only one of its kind" was still the correct answer, having seen "very unique" used all too often recently. They accept "only one of its kind".
Missed "hegemony" (wasn't sure a hegemony had a leader), "quotidian" (should have known that, seen it before), "ultracrepedarian" (new word to me), "absquatulate" (19th century slang), and "fartlek" (Swedish interval training).
As others have pointed out, too many clicks per word. I am a sucker for a 'how many words do you know' quiz so I finished anyway. Overall I'm skeptical of the classifications. In broad strokes, the early words are easier and the latter words are more challenging, but the middle is pretty muddied.
Some of the words chosen are rather absurd/inappropriate: breviary (which I got wrong but felt like a vaguely religious word) was characterized as intermediate but I think it's much more obscure and less obvious than that; Hippopotomonstrosesquippedaliophobia was used as a word (I got that wrong as well) - any type of 'phobia' word is really the sort of thing a fourth grader opens up a page in the dictionary and points out, not a word that is used... ever; metamorphosis and kinetic were labeled expert, which I don't agree with (what elementary schooler doesn't learn about the metamorphosis of a caterpillar into a butterfly? what high schooler doesn't learn about kinetic energy?).
Most words were reasonably well defined in a way that most people would understand or recognize. A few words had poor definitions: lethargy ("the state of being lethargic" - obvious); complacent ("smug satisfaction with oneself" - I disagree that complacency is intrinsically smug); magnanimous ("generous toward a rival" - I disagree that a rival must be involved); gauche ("socially awkward" - this is sort of close but the given definition completely misses the idea of being tactless).
They call it scientific and give a hand-wavey formula, but they don't explain how words are stratified in the first place. If stratified sampling is a formally recognized method of doing this, it would be nice to have a link to a real reference. I think I know a lot of words, but I am skeptical of the estimate this app provided (north of 75k).