Nothing wrong with 20-a-day vocab target

[imron]

Discussion moved from this thread.

 

Consistently studying my Anki cards has always been a struggle for me,

 

I hate to say it, but you're probably not going to do much better this time, given that you have said this

 

 

I am working on an Anki deck based on the SUBTLEX-CH word frequency list

coupled with this

 

 

I intend to learn them at a rate of 20 words a day

 

20 words a day is a huge workload once you've been going for a while, and when you add the fact that you're just learning what to you are effectively random words without context it will make difficult to maintain motivation and desire to keep going.

 

The important thing with learning is consistency.  You need to set goals you can consistently and easily meet every single day.

 

I think you also have a priority inversion.  With the above goals you're going to be spending a lot of time on flashcarding and comparatively little time on things like reading.  If you're at the point where you think you can read 30 mins a week, I would consider instead dropping a large amount of the flashcards and try instead to read for 30 mins a day.

 

If you can't do 30 mins a day, pick some other comfortable amount that you can do every day.  Daily is the important part (use a tool like Don't break the chain or my own 100% to keep track of progress.  You are better off spending 10 mins a day doing a task, than 2 hours once a week.

 

Then each day pick between 5-10 of the new words you come across and stick them in your flashcard program (don't worry about the other unknown words, if they're important they'll show up on another day).  Keep it up and eventually you'll get all the 6000 words in that premade deck anyway.  I've written more about developing a reading habit here.

 

You're probably going to think 5 or 10 words a day is too little, because at 20 words you'll have the remaining 6000 words done in under a year, but it will take 1.6 years at 10 a day and 3.3 years at 5 a day.  The thing is though, 5 words a day is easy, and you'll be able to do it day in and day out for 3.3 years and in the long term, you're better off doing 5 words a day for 3.3 years than doing 20 words a day for a couple of months and then giving up because of the high workload/burnout.

[Lu]

Imron is right. On all counts: 20 words a day is a lot; learning vocab is good but you need to be using it; daily habits are important; and it's better to do a little each day for the long term than to do a lot and burn out. If you're feeling studious on a given day you can always do a little extra on that day.

[Guest realmayo]

大块头: a 20-a-day habit isn't necessarily unhealthy but I think you've got to read more, and if it's a choice between 20-a-day and minimal reading, versus 10-a-day and lots of reading, then I would suggest the second option.

 

Don't worry about reading novels if they have too much unknown vocab at the moment. Instead get some reading comprehension books that are appropriate for your level, if you're aiming for HSK6 then they should be HSK5 and HSK6 books. These texts will provide a mix of vocab that HSK5 and HSK6 students should be learning, as well as vocab that is 'difficult' and that you can, if you wish, cross-reference with your 6000 list to see if you want to learn it.

 

Do the same with listening comprehension books.

 

Then you've got a multi-pronged attack on vocab: these words are not just encountered in Anki, but through reading and re-reading, and through listening. You will remember them far far better this way. Plus, you are improving your reading and listening skills, which will obviously improve your Chinese loads. And you are getting better at HSK exercises.

[StChris]

Sorry 大块头, but I'm going to have to join the others in advising against your 20 new words a day target. I completely understand the thinking behind such a target. Actually, I initially wrote 20 words myself, then changed it to 10, before finally swallowing my pride and setting the goal at just 5. Even now I feel a little guilty, as 5 a day seems like a pathetically low number to aim for, but I know from bitter experience that adding 20 a day very quickly results in a monster stack which ends up swallowing up all your Chinese study time, eventually ending in burn out a few weeks down the line. 

 

Just because you're only adding 5 a day to your flashcard stack doesn't mean that you're really only learning 5 new words a day. Assuming the material you are reading/watching/listening to isn't too difficult (let's say that your level of comprehension is at least 75%) , then you'll begin to recognize lots of new vocab due to the fact that they keep on popping up in whatever you're reading or watching. I've also found that when I restrict myself to 5 new flashcard entries a day, I tend to really cherish the vocab that I do pick. I might perform as many as 50 dictionary look-ups on some really Chinese-heavy days, so before I add a word I have to think, "Has it really earned the right to be included in the majestic 5? Why would it be more useful to memorize than one of those other 49?" When you're adding 20 a day they just become a burden.

[imron]

5 a day is a low number, and that's the whole point! You want to be able to hit that target every single day.

The important number, the number that you should be happy/disappointed with, is not how many words you learn in a day, rather it's how many days in a row you keep at it.

[大块头]

To: imron, realmayo, Lu, and StChris
 

Thank you for your thoughtful comments and suggestions. I will definitely back off if 20 words/day proves to be too much, but that rate is based on my previous performance with the material in this deck. I am at the halfway point after about two years of study (a period of time that also included two job changes and three moves). I should have an hour and a half each morning to study with Anki before I head off to my new (and hopefully more long-term!) job. I'll adjust my learning rate so that Anki doesn't swallow up my entire day.

 

As for whether I would be better off if I spent that time reading or listening instead, I subscribe to the school of thought that emphasizes the importance of building a large passive vocabulary. I think strong arguments can be made for and against this approach, but maybe we should move that debate into another thread?

[Guest realmayo]

Learning 20 words a day is not too many.

 

If by learning you mean that you know each word inside out, you can use it freely in conversation and so on, well, five-a-day may well be a very reasonable number.

 

But, as you mention passive vocabulary, then I agree, 20-a-day is certainly doable. Especially if you're prepared to forget and relearn some of them.

 

Obviously reviewing vocab doesn't improves your real Chinese ability: the other skills (reading/listening/speaking etc etc) are required. If only 20% of your study time is on those skills, because you have set yourself a high vocab target, then when you reach that target and turn your attention to reading etc, your progress will be slower than you hope because despite the reservoir of memorised vocab, your brain isn't yet trained to using the language as a language. It would probably have been more efficient to spend some of that vocab time on skills instead. So it's a question of balance.

 

As for warning about overload, that's a valid warning to give to someone who is starting out and has divided a notional target by 365*2 or whatever. But not to someone who has been studying for a while.

 

There's also the issue about how many words you already know. I think there's a lot to be said for splurging on vocab at a certain point in order to get it up to, say, 5000 words. That will get you able to read plenty of material so may be worth sacrificing some skills time for.

 

But above maybe 8000 words or so, flashcarding new vocab becomes different, or at least it did for me. You end up with words that you not only get wrong, but you don't even remember ever having learned in the first place. How to deal with that is separate topic (I now automatically delete words that I get wrong twice) but I'm just pointing out that your experience of learning your first 6000 words might be different to your second 6000.

[大块头]

I agree, 20-a-day is certainly doable. Especially if you're prepared to forget and relearn some of them.

 

Yes, it's important that your leech threshold isn't set too high. If you're not careful a minority of words will take up the majority of your time.

 

 

So it's a question of balance.

 

I've laid out what I think is a good mix of vocab, reading, writing, listening, and speaking here. My listening ability is still slow so I am focusing on that along with building my vocabulary.

[simc]

I thought I'd give my ten cents as I've tried to memorize words from the SUBTLEX-CH list as well.

 

Using a frequency list in a good idea as it ensures you are studying useful words rather than lots of relatively useless ones. But, the problem I found is that the further I went with SUBTLEX-CH, the less well the frequency list matched my personal experience of how common certain words are. I found this to be true after the first 3000 words. Ultimately, I created my own "frequency list" by creating different categories in Pleco for every book and TV show. I considered a word worth studying using flashcards if I had come across it in 4 or more different books or shows (this required writing a small computer program). This created frequency list tailored for me.

 

I agree with realmayo:  "It's a question of balance". You can learn words off a list but to really know them you have to see them in action (by reading/listening/speaking).

[imron]

Ok, now that I've got some time, I've merged all the posts from the other thread (they really belong here anyway), and will post a few more thoughts.

 

20 words a day needs to be taken in context.  By itself it's not bad, however we also know that:

 

* it's 20 words from an essentially random word list - it's not random in that it is based on word frequency in a given corpus, but the words are essentially going to be random to the learner and as simc mentioned, the more you learn the more the frequencies will diverge from the stuff you're reading, making it even more meaningless.  If you're not getting words from context - be that a textbook, graded readers, podcasts, or native material you're missing out on a major boost to comprehension and learning.  Word frequency lists can be a useful tool, but it needs to be in the context of what you are likely to come across and not just stuff that someone else has compiled.  That's one of the reasons I made Chinese Text Analyser - to help people get word frequency lists from the material they were currently reading.

 

* the OP is starting full-time work.  Not a problem in and of itself, but 20 words a day on top of a full-time job definitely becomes much harder, especially in the context of maintaining it *every* day over a prolonged period of time.

 

* OP is neglecting other skills in order to achieve this.  He's mentioned he set aside 30 minutes a day for *passive* listening, 30 mins a *week* for reading, but then 1.5 hours a day for flashcards.  To me that's skewed far to much in favour of flashcards at the expense of reading and *active* listening, both of which would have significantly more benefit for his Chinese level than learning random cards from a list.

 

I can understand the importance of building a passive vocabulary, but it shouldn't come at the expense of your active vocabulary or your other skills.  Unlike that article OP linked to, I don't think that vocabulary is king, and it's certainly not 'god-emperor'.  If anything, it's closer to 'false-prophet'.  For general material, the difference in reading skills between someone with a 10,000 word vocabulary and a 20,000 word vocabulary will be minor - maybe an extra 1-2% in comprehension, and the more you learn the more this becomes true.

 

The OP is at HSK 5 and has almost 14,000 cards in his Anki deck (based on the stats he linked to).  He's definitely at the point where he could start branching out in to native content, and at his current level it would also be very easy to get 20 new words a day from context rather than from a random list.  At the same time he'd be improving his active skills - either in listening, reading or both.  Although given everything else, I still think 20 new words a day is too much and he should cut down on that to focus on other skills because a mostly passive vocabulary is essentially useless by itself.  You'll start reading a book/newspaper and think - I've learnt all the words in this sentence but I have no idea what it means, or you'll be listening to TV/radio/movie and won't be able to keep up because even though you might know every word you are unable to process them at speed in a continuous stream.

 

You can also build passive vocabulary through reading/listening.  This comes in the words you can learn the meaning of through context and also in the words outside your 5-10 daily learning quota that you can still look up but that you won't bother to learn/add to flashcards until some later date.  This keeps daily revisions to a manageable level (in the context of also being able to do other activities) and also makes it easier to learn the word once you've seen it appear several times.

 

I would also caution against starting at 20 a day, and backing off it gets to much - at the point you realise this, you'll be approaching burnout and it will be more and more tempting to skip a day, which turns in to a week, which turns in to a month.  You're better off starting out small (5 words a day), and then every month or so ramping up another 5 words if you find it easy enough to handle.  That way it's much easier to find a level that you can comfortably sustain over a prolonged period.

 

Anyway, if I had the schedule/free-time mentioned by the OP (1.5 hours to study in the morning, followed by a 30min walk to work), I'd be looking to do a max of 30 mins of revision followed by 30-40 mins of reading or *active* listening e.g. studying the material I'm about to listen to on the way to work, using a method like this (note however that 20 words a day mentioned in that post is based on the fact that the OP is studying full-time, and so is in a different situation from you).  Then 20-30 mins on learning the new vocab that I came across in reading/listening (limiting it to 5-10 new words a day), and then 30 mins walk to work doing passive listening on the material I'd previously done active listening for.

[Wurstmann]

I learn 20 new words a day (recognition only). Anki says it takes me 16 minutes on average, including reviews. 

So it really doesn't take a lot of time. 

But I only learn words I encounter while reading books, so by the time they show up in Anki, in the most cases I've seen them a few times already,

which makes it a lot easier to remember them.

[imron]

I learn 20 new words a day (recognition only). Anki says it takes me 16 minutes on average, including reviews.

Does that include time spent creating the cards, or do you have an automated process for that?

[Guest realmayo]

Putting 20 cards into Anki might take me 10 minutes and is the main part of my 'learning' process because it's where I add example sentences and see how the word might behave in different contexts.

Or it might take me just a few seconds via a spreadsheet+import and then I'll edit the card with context if necessary when it first comes up for review.

 

Perhaps we can think of a heavy focus on vocabulary the same was as someone climbing a mountain might get helicoptered from low altitude up to base camp, rather than walking that first part. It is efficient and time-saving, but when he actually sets out to climb from base camp he'll have to go slower than those climbers around him who climbed all the way from the bottom; this is because he's not acclimatised and his body isn't as used to climbing as theirs is.

 

Or in a language sense, his reading/listening skills are relatively weak. But if our helicoptered climber spends time strengthening those skills before setting on the next stage, and knows he'll have to take it slow to start with, then my guess is that the helicoptering was definitely worth it: three steps forward, one step back: he eventually gets to the top the of the mountain in less time than if he'd walked the entire way.

 

However the danger is that the vocab focus goes on too long and as I said earlier, you not only waste time by delaying when you start actually start reading lots, but also what was an efficient learning process up to maybe 6000 words becomes a lot less efficient when you get to higher numbers. Maybe helicopters can't cope with too-high altitude and too-thin air.

 

On a more down to earth note, the OP, who may well be feeling a bit picked on at the moment, says he's aiming for HSK6. In which case I wonder why he needs 12000 words. The last time I saw an HSK vocab list it was around 8000. But good luck to the OP and I'm sure his approach can be successful. My advice is just to be really alert to whether at any stage the vocab-focus seems to have reached its natural limit and a change to the approach might be required. If something's been working well for two or three years, it can be difficult to change it. But maybe he'll get to his target without needing to change and will then be in a great position to start the really hard job of reading lots.

[imron]

I don't think we're in much disagreement, it's just that I'm arguing from the view that he's reached the point now where he needs to start strengthening the other skills and 'helicoptering' isn't going to return as much value going forward.

 

His deck is about 14,000 cards, and even accounting for reversed cards that's still 7,000-8,000 items.  He's reached (or is fast reaching) the point where learning new vocab is going to provide a much lower return on effort than if he'd invested the time improving other skills and picking up some amount of new vocab as a result doing that.

 

I don't mean to pick on the OP, just trying to provide some (hopefully) constructive advice for learning.

 

I agree that being really alert about your learning process/progress, and making changes as necessary is important.

[Guest realmayo]

Imron I might even go so far as to say we largely agree

 

I also think that no one would be having this conversation if either (a) we were learning an easier language or (b) there was abundant reading and listening material for learners of all levels. But because Chinese is hard and there isn't that material, the step from beginner to extensive reading is a huge step and calls for difficult measures: that might be ploughing painfully through too-difficult texts for a while where you're looking up every fifth word, or it might mean splurging on vocab for a while. I do understand people benefit from, say, 30 mins reading a novel every day, but that didn't work for me, 30 mins was too short and too slow and too boring, it wasn't 'reading' but was 'study' and I found preferable ways to study. It wasn't until I could happily read interesting texts for an hour or more each day that I made real progress reading.

[imron]

I do understand people benefit from, say, 30 mins reading a novel every day

It doesn't have to be a novel - graded readers, graded news articles such as Du Chinese and Chairmans Bao, normal newspapers, blogs etc.  There is a growing list of character only material available for people at the intermediate and upper intermediate levels that didn't exist in a convenient way several years back.

[大块头]

imron, realmayo, and everyone else:

 

Thank you for taking the time and effort to share your advice, I really appreciate it. Sorry it's taken me a while to respond.

 

As I said from the start, I'll reduce the number of new cards I learn every day if it starts taking more than an hour and a half to clear my Anki queue.

 

I put this deck together by taking the 16,000 most common words from the SUBTLEX-CH frequency list and then adding English translations from CDICT. There are about 14,000 cards in the deck right now because a certain proportion of those 16,000 words didn't have a CDICT definition and a certain proportion of those cards only seem to be there because of some error in the SUBTLEX-CH statistical analysis. All these cards are passive recognition; there are no sibling cards.

 

It takes me about two hours every week to look through the new cards I'm going to study and refine the answers with my other dictionaries in Pleco. It takes me a little more than an hour each day to complete my Anki reviews and learn 20 new cards.

 

With regards to diminishing returns: I picked 16,000 cards as my target because those words comprise 98% of the SUBTLEX-CH corpus.

 

Spending less time with Anki and more time listening/reading may very well improve my Chinese more quickly, but I have my own reasons for making this deck a priority.

[imron]

Fair enough.  I look forward to hearing how it all went down one year from now in the 2017 Aims and Objectives thread

[eddyf]

So I'm curious, how are you testing yourself on these cards? You say it's passive recognition. Does that mean you get shown the characters and then you judge yourself on whether or not you knew the pronunciation and meaning? Or something else? How exact do you have to be with the meaning? What if a word has multiple meanings, do you have to remember all of them?

[大块头]

@eddyf

 

Yes, the front of the Anki card has the hanzi, and I need to supply all pronunciations with the associated meanings. I am flexible when it comes to part of speech, but not when it comes to the meaning. When I make the cards I attempt to simplify the meaning as much as possible, however, so I don't usually find myself splitting hairs.