The statistics present an illusory impression of having learnt vocabulary, in so far as I understand the term “leant”.
I have been through several hundred of my 2000 known words to discover that I actually know less than one-third of them when not referencing the sentences in which they occur in the mini-stories.
The illusion is created because the words in the spaced repetition reviews are presented with sentences from the mini stories. Unfortunately, outside a mini story scaffolding, the vocabulary is simply not retained to the same extent. Thus creating an illusion that vocabulary has been learnt when in fact it hasn’t.
The situation is even worse for the verb tenses. While I might recognize all the conjugations as belonging to the infinitive, I couldn’t tell how most of them are to be used, not even simple tenses.
This is exactly as predicted in the book “Make It Stick - the science of successful learning”. Re-reading is not an effective learning strategy and creates an illusion of learning.
Of course, it may well be that I have a poor aptitude for language learning. I would be interested in other users’ experiences.
My approach is I simply don’t study for stats - other than the hours I spend learning which I track myself independent of Lingq. I pay virtually no attention to the Lingq stats.
When doing SRS, LingQ is simply too aggressive in promoting words. I think this is what you’re discovering.
For myself, I stopped using SRS at all after a few weeks in favor of using that time/energy to consume more content from a wider variety of sources.
I manually promote words from 1 to 3 when I first start to recognize the character, approximate sound, and meaning. Then 3 to 4 or straight to 5 once recognition becomes more automatic and accurate. If I forget a word after not seeing it for a while, which is only natural, I’ll mark it back down.
My only use for keeping track of known words is being able to estimate the difficulty of new material. It’s really not much good for anything else that I can tell.
If you’re struggling to recognize words outside of the context in which you first encountered them, it sounds to me like you would benefit from simply viewing a wider variety of sources and not trying to learn every single word you see.
I can promise you from direct experience, you 100% will see every important word over and over and over and over again, and they will stick. As you keep going, you will internalize these common words quickly and your circle will start to naturally expand over time to include less and less common words.
The hardest thing to get used to at first is just letting new words fly right on by until you bump into them a few times.
Learning vocabulary takes time. Not every word sticks after seeing it only a few times. Some words and phrases stick easily, but most don’t. You just have to keep reading new material. Eventually you’ll start to pick up even the stubborn words.
If all you’re learning from is the mini-stories, you’re doing it wrong. You need to be reading from other sources. The more variety, the better.
And LingQ isn’t about “re-reading”. Again, that’s not how you’re supposed to be using it. The point is to find new material all the time. Only by doing that can your brain start to make the connections it needs to make vocabulary stick.
And it might be best to forget about tenses and grammar. It helps some folks, but I suspect most people find the framework of language to be a stumbling block. It’s far easier to just read and let your brain figure out the context organically, as we do with our first language.
I’ve been learning Spanish using only LingQ for a year now. When I started, I could read only with a lot of difficulty. Now I’m reading fairly easily. So it does work. It just takes time.
I agree with what’s been said above. First of all, you don’t learn for statistics, but to acquire a skill. And a skill as complex as a language cannot be represented by a single number.
Secondly, when you are rereading the same text over and over again, you start memorizing it like poem. Why should your brain bother to internalize the meaning of a word or a grammatical structure, if the important information is the sentence, paragraph, whole text?
Going through a text repeatedly that you consider challenging or after some time is not wrong per se, but if you experience it doesn’t serve you well, try moving to new content. Personally, I never read a text twice. Simple because it is boring.
Don’t worry about the statistics, number of hours that you’re spending in reading or listening or the amount of the known words. Just focus on improving the capacity to find materials that are interesting to you. About the activity to re-reading and listening the same content over and over again, in my opinion, it is important in order to catch the meaning of the words and phrases that you don’t know. but again, let your braind and interests make the decision for you. 2000 known words is just the beginning. I’ve studied English for only one month and with the right attitude, I feel like I understand more the contents which I am choosing to learn. Have a good day!. ( And sorry for the wrong words and phrases. My first time writing here. No courage, no goals !!! rsrsrs)
Lingq isn’t about micromanaging vocabulary. It is about reading (with or without audio) as much as you can in your target language. Don’t worry about re-reading material you have already been through. Keep going and things will eventually improve.
I set the status of a word to ‘learned’ or ‘known’ manually and only when I am completely confident that I know it. That way the statistics can be relied upon as a true reflection of learning progress. I don’t use the SRS in LingQ.
I get the sense of a disconnect here between those who are using LingQ for beginning to learn a new language (say the first full year of committed study) and those at an intermediate stage and beyond.
For beginners, what alternative is there to listening to lessons many times over to start to learn how a totally or relatively unfamiliar language sounds, ie to improve listening comprehension? And as part of that then you also learn to read and recognise/use new words.
Also, for many, less popular languages, there just isn’t that much true beginner-level content readily available at present. For anyone who thinks otherwise, I would say just try starting to learn a language that’s not in the top 8-10 of popular LingQ languages and that is in a different language family from your own (eg Asian vs English or vice versa). Content like the mini-stories then becomes invaluable and has to be read many times over.
But I can also see that once you are more advanced in a language and are encountering only a few new words in any new piece of content that you consume then you may have a completely different attitude to eg new word stats and may not think it useful to read content more than once.
I have also read Make It Stick and my experience matches yours. I was learning German the LingQ way for a year, and it just wasn’t working. I could sometimes recognise words in context, but not recall them actively. Even basic stuff was not sinking in. As the book explains, we need testing, we need output, input alone is not enough. It is very frustrating that Steve Kauffmann continues to promote methods that have been discredited e.g Krashen’s input model.
I try and practice output, as a form of testing and experimentation. After all children play with language and although we do not learn like children, I think we still need to practice forming sentences, and discovering what we get wrong, so we can correct it.
One problem with language input alone is that it isn’t enough to understand messages. One can understand a sentence perfectly without understanding a verb conjugation. Thus one may not notice the verb conjugation, and hence not learn it. Explicit learning will solve that problem.
I loathe Duolingo, but to their credit they implement simple testing in their lessons, and do not rely solely on input.
I don’t think you have a poor aptitude. You most likely have average aptitude that most of us has. I think you’re making a judgement way too early in the game. 2000 words in the LingQ world is not a lot and you likely just have not seen these words enough. They will start to stick. As I think Prometheus mention, some will stick easy and some will take a very large number of reps to get them to stick.
See the words in isolation IS much more difficult, but at least in terms of reading and listening, you mostly encounter words in sentences. If you know the word in the sentence, I count that. Doesn’t mean I know it in isolation YET. It doesn’t mean I can produce that word in output, but LingQ isn’t really a tool for output. (It can be…I do have some ways I do this). It does mean I can recognize the word in this particular context. I may read it again in a different context and not remember it. I set it back to 3 (or whatever you like). See it in different contexts will start to cement it. I’ve learned a ton of vocabulary without ever uttering those words. So it does work. (can’t speak to the Make It Stick book or what it suggests).
One thing possibly you have to be careful of, if you are re-reading the same material over and over again. At some point you will know the story by heart. Then you might think you know the word, because you have the story memorized. On beginning material, I wouldn’t repeat it more than a handful of times max before moving on to the next story. I wouldn’t repeat the same story in the same hour. Put some space in between repeat readings. Better yet, find new material. (Yes, this takes effort. Yes, some languages this might be difficult).
I can’t say anything about LingQ’s SRS. I’ve used it a few times. I don’t care for it personally, or Anki. I am doing some spaced repetition in a different manner to fill in gaps of high frequency words I’ve missed along the way, or need extra work to cement them in, but still the majority of my time is input (rightly or wrongly =)…at my stage I probably should be working more on output). I do think there is a place for SRS at the beginning to learn the top 500-1000 words. Maybe a different tool like Memrise, Duolingo or Anki for that. Otherwise plugging away at the mini stories or other beginner material should get you there.
For verb conjugations, perhaps a basic book like Teach Yourself or Assimil if it is available for your language would be a helpful start to give you some of the very basic grammar and conjugation rules. Just to give some structure to what you’re seeing. LingQ usually also has grammar guide for the language under your profile section or in the lesson menu (can’t remember exactly as it’s been awhile).
I wonder what people mean here by “knowing” a word or “learning”, for that matter.
Yes, producing the language is much harder than input, especially input scaffolded by the LingQ method (or similar). I don’t think anyone ever claimed that just by doing LingQ, you will be able to speak the language fluently. (In fact, Steve Kaufmann himself very frequently points out how much time he spends with tutors working on his speaking in various languages.)
But what you are doing is building up passive vocabulary and intuition. If you are ever going to say the word, you’ll need to know the word first, and that’s where LingQ (and other comprehensible input approaches) help.
You’re probably right on that, but I’d argue that LingQ may not be the right platform to learn the very beginnings of a new language. Although many people say Duolingo is a bad language learning app, I’d say it is probably better than LingQ for those just starting out with a new language.
I’ve used LingQ since I was a beginner in Spanish off and on since 2020. After years of playing with it from many angles, I find it so so for beginners and fantastic for intermediate and above. For my experience at least.
Looking back on how I was able to finally become fluent in Spanish and now dabbling in German and French at times, It was was a combination of vocab/sentence study in a limited fashion, such as lingQ, anki/looking up words when reading books… with a ton of comprehensible input.
Although not as fun as watching a show, I loved Dreaming Spanish. It seems to fill in the missing kinetic/visual aspects of the language that just studying text or listening to audio is missing. Thus missing the more embodied aspects of memory: movement, relational, emotional, visual. Of which shows give you but they’re too difficult for most beginners.
I personally don’t use LingQ extensively. I use it for my intensive study. In service of immersion and comprehensible input. That’s the flaw of all apps, if one believes they’ll learn the language by the app. Apps in my opinion are there simply to make immersion comprehensible.
When I got into the Refold/AJATT methods, my shakey A2 level shot up to a solid B2 level within the year and was able to speak. Just wish I would’ve began speaking sooner, it’s extremely difficult. (And doesn’t come intuitively with enough input with most people, despite what Krashen teaches).
This is obviously a complex subject and nuances are easily missed. One variable is just how different your target language is from what you might know already.
But also it seems to me that there are two beginner levels - pre-A1 and A1. I agree 100% that LingQ is not the the app for pre-A1. (Though that’s not to say that that the LingQ platform could not extend its scope by providing more curated pointers to non-LingQ material for complete beginners)
But once you understand the writing system and have some basic words then LingQ can be useful with very simple content, albeit some lessons simpler than the current mini-stories would be valuable.
IDK, I started Chinese from absolute ground zero with LingQ and it was fine.
The first thing I did was start going through Who Is She? series one section per day, reading/listening to each many, many times (and doing the SRS for the first couple weeks because I just didn’t know any better).
It was perfect because it starts very slowly and the vocabulary/story build over time. I was able to pretty quickly internalize the very early basics and branched out to low word count graded readers, which reinforced them.
I don’t think I’d have been better off studying elsewhere first, but who knows for sure?
First, if you mark a word “known,” it shows up white next time, without the aid of a dictionary definition. If you mark a word as not known, you’re provided your selected definition the next time the word. Whether I mark a word known or unknown is based upon the app’s behavior and the simple reflective question: Do I want to see the definition the next time I see the word as I’m reading? First and foremost, LingQ is an assisted reader for individuals reading in a second language.
This gets me to a related “tip” in how I use LingQ. I use LingQ in sentence mode primarily. With a typical sentence, there’s often enough space to show the definitions of two or three yellow unknown words from the sentence. If I’m reading something to easy for my level, many sentences will have all white words. If I’m reading something to hard for my level, all of the assisted definitions won’t fit on the page.
Using these two approaches together, I typically go from sentence to sentence and need to conveniently glance at the right-there definition. If I don’t want to see that definition again, I click it away to known. And on to the next sentence. This is how I use LingQ efficiently as an assisted reader.
Secondarily, LingQ is a spaced-repetition system. I haven’t used that since the first few weeks of using LingQ. Therefore, I don’t bother with the 1-4 detailed levels of not yet known.
Whether or not I can use the word in active vocabulary, etc. is all a separate matter. I use known/unknown within the context of productive use of LingQ.
I remember my teacher was requesting to me if I know a word saying this word. I would say no. Then she used it in a sentence and I did recognize the word. So if you are abble to recognize a word in context and not in isolation that shows the method is working. Personally I sometime mark a known word back to new. Frequency of me doing this is very low. Checking your stats, you have not yet seen the words in many contexts. It will come with time and volume of reading / listening.
Also mini stories are basically 40 seconds stories said twice followed by questions. It’s too short to be involving. Try to find longer video 5 to 10 minutes long.
It’s true that the “illusion” exists, but this applies regardless of the learning method you choose.
Even with your native language, there are countless words you think you know but actually don’t.
The anxiety of “maybe I don’t really understand this” is always present, and it’s something you need to manage on your own.
The status of “known” always comes with the condition: “in the context of texts I’ve encountered so far.”
If you come across a text that you can’t understand with your current knowledge, you can simply adjust the status at that point.
I personally prefer not to use the quizzes provided by LingQ because they forcibly upgrade the status of words.
I prefer to adjust the status myself.
This is also one of the reasons I don’t highly rate Duolingo.
Language books and internet sites. The ability to understand spoken language isn’t only determined by how extensively you have listened to the language, but also by the amount of words you know. So building up some basic vocabulary is crucial. At this very beginner stage you don’t necessarely need audio.
I am a German who knows English (probably low C1 level) and some Spanish (A2->B1). I learn Korean for 1.5-2 years for maybe 30-60 minutes a day and am now at early B1 level. I never had any issues finding learning material. It is true that there is much less available then there is for european languages, and Korean probably isn’t the most exotic choice among the asian languages, but especially the internet is a good learning resource.
I think that you can learn a language using LingQ even without any prior knowledge. However, it isn’t a very efficient approach. And in general one shouldn’t rely on one app solely for learning a language. Especially if the app isn’t designed to learn the targeted language specifically. For example, I am
using LingQ for vocabulary acquisition by extensive reading
watch series so I can hear the language (I usually don’t use audio with LingQ)
study grammar using grammar books or looking it up at the internet
study Hanja from time to time (to get a better feeling for how words are made up)
use ChatGPT for translation practices, both German->Korean and Korean->German with different textstyles at my current language level (newspaper article, short essays, dialoges or stories etc…)
practice things like grammar and sometimes vocabulary using ANKI (not too often, as it is boring )
So LingQ is just one tool in my toolbox. You cannot build a cathredal using only a chisel.
The problem I see is the branding. The name language learning app implies that you can use Duolingo, Memrise and the like to learn a language. But it really only helps you build up the fundaments. For a beginner, this can be very misleading and frustrating, to a point were some might give up. It just raises wrong expectations. Your are right, though, that they can be useful for getting into a new language, and if they would communicate it that way, the criticism would probably be less harsh (if even existent).
)