Hey, I just noticed that the Russian level boundaries have changed. The number of words needed for advanced 2 level is now much lower. Just curious as to the reasoning behind the change, that’s all 
Really? The avatar help page still lists 37 000 words as the criterion
Yeah just checked that - strange! I only noticed it when I was reading on my tablet - I finished a lesson and it said I had around 3000 words (can’t remember the exact figure) until I reached advanced 2, whereas before it was targeting 37000. Maybe just a bug!
Yes, learning targets were changed for all languages recently. @josh you need 24550 known words to move on to advanced 2 level now in Russian.
The Avatar help page will be updated soon.
Thanks for sharing. So it was changed a few weeks after I reached advanced 2 according to the old criterion?
Oh my!
Just curious how you investigate the boundaries.
Also very curious, since I still don’t feel like I know the language at an advanced level.
I always thought the levels were screwed up. The word counts, LingQ levels, LingQ certificates, and CEFR levels don’t correspond to one another. For example, previously in Spanish: Based on the Avatar, 22,000 words should have indicated you reached Advanced Level 2, get a certificate for Advanced Level 2, indicate you have a potential to reach C2 on the CEFR, and tell you that you need 32,500 to reach LingQ’s Advanced Level 3.
Not so.
Instead, when you reached 22K, you were told you were AT Advanced Level 2, got a certificate for reaching Advanced Level 1, told to strive for 32,500 to reach Advanced Level 2, and are likely more in the realm of B2 on the CEFR.
A more detailed explanation of how these thresholds are calculated and what they are supposed to represent in terms of ability, is in order.
The Chinese level boundaries are complete nonsense. You cannot define the same level boundaries for English and Chinese.
According to LingQ I am Advanced 2 (or is it 3?) in Russian. This is nonsense. I would have difficulty passing a B2 test at the moment. I’m just quite good at hitting the k hotkey on Lingq.
I am planning to follow your progress when I can, as my next language project will likely be French or Russian. How is your Russian overall? Is your French awesome (in any way) at 44K? thanks.
1 Like
For the most part we were just standardizing our targets across languages into a few different groups. We are trying to have targets that are achievable in 3 months of dedicated activity on LingQ. We also recognize that there are additional levels after Advanced 2 where you will continue to grow your vocabulary. We felt the previous targets might have been a little overwhelming for people, even though they will achieve them if they follow the process on LingQ.
The avatar page will be updated to reflect the changes.
why not to use a stemmer and set level boundaries just using the average vocabulary size required for passing a proficiency test?
I get what Mark is saying, and I think it’s a good idea. However, the word counts are meant, or ought to be 1) the “Gordian Knot”/ benchmark to show that you are, in fact, making progress, especially when you reach that “long road” along the intermediate part of the inverted hockey stick; and 2) to get a sense of WHERE you are along that path. In other words, how much “potential fluency” you have stored up in the bank.
To me, it would seem that the new more easily attainable achievement levels would be better incorporated into a sub category of levels built into the 90 Day Challenge, with the regular weekly, daily, and avatar targets measuring the real achievement at various levels. Or vice-versa, where there is at least a way to know this.
I realize this baseline info is hard to come by, especially in languages other than English. However, if we take a lesson like Who is She and other sample texts across multiple languages, we might be able to calculate a “conversion” rate. Combine this with user experience, especially Steve’s, and some real targets might emerge. For example, even if we don’t know it for Russian, we know that potential English fluency is between 10-15,000 words. If we find out that Russian has a 4.5 word rate to English, then potential Russian fluency would be 68,000.
This conversion rate is very approximate and using stems is the way to avoid this approximation. The number of ‘pure’ words you need to know to get a particular level is basically the same for all languages.
Also, from the practical point of view, it would be more useful not just counting words, but to show how your words are distributed in a frequency dictionary, i.e. how many fall within first thousand of most common words, second thousand and so on, which is also easier with stems.
1 Like
But maybe it’s not particularly useful for that 3-month challenge thing.
Hey, sorry for replying to you so late. I’m travelling in Thailand atm. My French word count is irrelevant since I speak French with my French mother. I’ve studied Russian for a year now (including 7 months at Saint Petersburg State University). After those 7 months I went to China on a work contract and will be going back to live in Russia in June. When I left Russia last June I could function pretty OK, could get through transactions with service people and my landlady. Now that I’ve been away I have really only used Russian to read literature (and a little bit of using it as a secret language with my fiance and in talking with her mom). Now starting to be able to read 19th century stuff, albeit slowly and with heavy dictionary use. At nearly 40k words on lingq I still get tons of unknown words on every page. At my French word count there are basically no blue words. I would not be surprised if a truly advanced level for Russian on here would be at a 100k+ word count.
That sounds about right. Hence Steve’s 92K words in Russian. Are the tons of blue words you referenced turning up when you read the 19th century stuff, or the regular stuff? Regardless, despite it being only a year, it’s been in country and I imagine you probably have a relatively high level of use relative to your known words/passive vocabulary count. Thanks for replying and I understand the delays.
We have in Russian even less ‘root’-words than in English or German. But we can derive more words from one root - and to learn ‘to guess’ such related words is the main and the most difficult task in the advanced level.
After that even the classic Russian literature of the 19th century won’t be a big problem.
Marginals can use just about 10-30 main roots to express everything they need. Most of them are not used in books.
That’s the real power of prefixes and suffixes