After using MCDs for about a month, I am going to stop doing them.
Compared to a card where you remember the reading and meaning, an MCD takes an incredible amount of time. I could probably go through 400 sentence cards in the time it takes to do 100 MCDs and the only possible benefit of them, output, does not seem to gain any benefit. On demand recall seems just as slow as before.

sikieiki Wrote:After using MCDs for about a month, I am going to stop doing them.
Compared to a card where you remember the reading and meaning, an MCD takes an incredible amount of time. I could probably go through 400 sentence cards in the time it takes to do 100 MCDs and the just as slow as before.

My experience with cloze cards is the same. I've become skilled at filling in the blanks for specific cards, but this improvement hasn't noticeably benefited my output at all. In terms of drilling exercises that don't involve speaking, my output has improved the most through the fill-in-the-blank worksheets I do a few times every week. These worksheets have sentences I've never seen before minus certain particles. My job is to insert appropriate particle in the blanks. My scores have been gradually improving, which means that my claims aren't just subjective.

sikieiki Wrote:After using MCDs for about a month, I am going to stop doing them.
Compared to a card where you remember the reading and meaning, an MCD takes an incredible amount of time. I could probably go through 400 sentence cards in the time it takes to do 100 MCDs and the only possible benefit of them, output, does not seem to gain any benefit. On demand recall seems just as slow as before.

After 1(!) month you don't see improvement and you quit? Sorry, but I can't help making a comment that some of you guys have really unrealistic expectations. Look, if you agree on "if you want to improve output, then you need to practice output" then how exactly are you going to go about it? Shadowing has been labeled inefficient, cloze deletions are labeled inefficient - I bet that any output practice is going to be labeled inefficient if you expect improvements in 1 month. Take it to couple of years and then make judgement.

For the record, I must be very inefficient in my studies because I'm doing both, shadowing and cloze deletions.

@InnyJan
I've been using cloze delete cards for almost four months now. I've improved my writing and conversation skills much more through actual conversations and fill-in-the-blank worksheets. Grammar points stick much more when I use them in actual verbal conversation or in my writing as opposed to cloze delete cards. Try it out.

Inny Jan Wrote:After 1(!) month you don't see improvement and you quit? Sorry, but I can't help making a comment that some of you guys have really unrealistic expectations. Look, if you agree on "if you want to improve output, then you need to practice output" then how exactly are you going to go about it? Shadowing has been labeled inefficient, cloze deletions are labeled inefficient - I bet that any output practice is going to be labeled inefficient if you expect improvements in 1 month. Take it to couple of years and then make judgement.

For the record, I must be very inefficient in my studies because I'm doing both, shadowing and cloze deletions.

To be fair, if one is doing two hours of studying/review per day, that equates to 60 hours in "just" one month. Perhaps not seeing a change or noticeable benefits after 60 hours of a technique puts into question the efficacy of that technique for that one person.

If we've found out anything over the years on this forum is that different people click differently with a variety of techniques and paths. Reporting success, failures, adaptions and new experiments has been the key to growth.

For what it's worth, I noticed immediate improvement a month after I started "processing" dramas using subs2srs in conjunction reading the dramanote scripts and always playing audio from those dramas. On the other hand, doing cloze delete in some fashion seemed like a step backward for me. However, I still recommend processing dramas and cloze delete to others both because of my success with the former and other's reported success with the later. I have to trust that it does work with other people even if my mindset is not compliant to the same.

Inny Jan Wrote:I could probably go through 400 sentence cards in the time it takes to do 100 MCDs and the only possible benefit of them, output, does not seem to gain any benefit. On demand recall seems just as slow as before.

And you could probably go through 800 or more pure vocab cards. If you want to learn usage in a rich and contextual way, just read books and webpages and things. If you want to learn grammar read a book about grammar.

sikieiki Wrote:I bet that any output practice is going to be labeled inefficient if you expect improvements in 1 month. Take it to couple of years and then make judgement.

Why would you continue with something for a couple of years if it has no observable effect after a month? One should be able to speak pretty well after 2 years of regular study. Divide that by 24 and the unit of improvement should definitely be observable.

sikieiki Wrote:After using MCDs for about a month, I am going to stop doing them.
Compared to a card where you remember the reading and meaning, an MCD takes an incredible amount of time. I could probably go through 400 sentence cards in the time it takes to do 100 MCDs and the only possible benefit of them, output, does not seem to gain any benefit. On demand recall seems just as slow as before.

This makes no sense to me at all. MCDs are supposed to take about 1/4 the time that a normal sentence card would require, not the other way around. All you have to recall is that one fraction of a word. If you're spending that long doing it you're not doing it right, no wonder it's ineffective.

Personally I don't only use MCDs, I SRS material that is really compelling, regardless of genre, and pick a method based on the nature of the material. MCDs are best for SRSing long blocks of text... but Subs2SRS is great for studying from TV/movies. When I run out of compelling new material of either form I do core10k.

MCDs aren't going to be useful if you dont find them fun and that necessitates a desire to read big blocks of text. Maybe audio & short sentences is more your thing... try Subs2SRS perhaps.

Note: Fixed attributions.

nadiatims Wrote:

sikieiki Wrote:I could probably ...

And you could probably go through 800 or more pure vocab cards. If you want to learn usage in a rich and contextual way, just read books and webpages and things. If you want to learn grammar read a book about grammar.

This. With a disclaimer that you are developing passive skills only.

nadiatims Wrote:

Inny Jan Wrote:I bet that any output practice ...

Why would you continue with something for a couple of years if it has no observable effect after a month? One should be able to speak pretty well after 2 years of regular study. Divide that by 24 and the unit of improvement should definitely be observable.

I don't believe that one doesn't benefit from studying grammar (ie. cloze deletions) because I can see observable effects on myself. The alternative to not studing grammar is a wacky output and I would't recommend that to anybody.

nadiatims > misquote
Also, it may be true that MCDs are effective after 2 years, but so are a whole lot of other methods. The idea is to pick a method which works well and is enjoyable. Having said that, I know that sentences work to an certain extent, and switching to MCDs was found to be much more difficult for seemingly no improvement. I see no reason to continue them.

>Read a grammar book.
If you want to learn english, just read a dictionary.
/sarcasm
I bargain most people hate grammar and a grammar book has more value being a sleep pill than educational.

MCDs take a long time.
Please produce one reason why recalling one word in a large paragraph is easier than skimming over a sentence for meaning. Especially if the word is anything but basic.

sikieiki Wrote:MCDs take a long time.
Please produce one reason why recalling one word in a large paragraph is easier than skimming over a sentence for meaning. Especially if the word is anything but basic.

because you're not supposed to do mcds if you're at a basic level.
well that's my opinion persoanlly. at my level right now i find them to be really helpfulbut i don't know helpful they would've been 3 years ago when i was at the "basic level"

because you're filling in the blank... if you've have some amount of intuition/comfort with that language it makes that much easier and simpler and funner. without that intuition that you build up over time with immersion and whatnot i would imagine it to be very painful and time-consuming.

there's your one reason!

sikieiki Wrote:MCDs take a long time.
Please produce one reason why recalling one word in a large paragraph is easier than skimming over a sentence for meaning. Especially if the word is anything but basic.

this is just confounding logic on so many levels at once.

first of all the 'sentences' method that people talk about presents a sentence and asks for the meanings and pronounciation of all words, as well as the meaning of the sentence... not a skim...

second, maybe you dont want to do the sentences method that way, but if this is your way then what the hell is a simple sentence gloss going to do for you?? moreover how are you going to get that gloss if you dont know the meanings of those words. moreover again how are you going to read anything if you dont bother to learn readings?? i'm at a bit of a loss here...

third, to literally answer your question, lets just do some simple math here. An average sentence contains about 8-10 words perhaps. So if we are playing this odd sentence-glossing game then you at least need to have a half-arsed understanding of about 4-5 actual nouns, verbs and adjectives there...so 4-5 things. An MCD only asks you how to write 1/2 of one thing in most cases, and a reading... as well as the half-arsed meaning you were doing already anyway. So rather that half-arsedly learning all those other words at the same time you just focus in a slightly more useful way on 1/2 of one thing at a time. I average 5seconds/MCD card.

fourth, basic or not basic is up to the card... this applies to the sentence method's material, the MCD's material, and the chosen word in the MCD. don't do it if it's too hard... that's the whole point of the MCD, to allow you to have a huge number of super easy cards that you can fly through like popping bubblewrap... as well as to be able to zero in on your trouble-spots effectively.

vileru Wrote:Try it out.

I would love to, really. The problem is that I don't live in Japan and I can't use Japanese in the way people staying in Japan can. (Even during my recent trip to Japan, after my shy attempts to speak Japanese, I was able to make a note of and improve on some simple grammar points that I was making mistakes with.)

For me, the cloze deletions are kind of substitute for the real use of the language. They do/will require polishing at the stage when real conversations take place.

I never said I was at a basic level.
I can skim sentences and grasp the meaning quite easily.
I use SRS for vocabulary, thus it is quite simple to skim a sentence, notice the word I intended to study and produce the reading and remember the meaning.

Alternatively, in MCD, producing the same word is quite difficult.

For example, an excerpt from a card :
巻き立つ＊が、白亜の神殿のようにそびえ立っている。
Answer being 入道雲

No need to get angry. I am just saying that MCDs didnt work, or work but arent enjoyable enough for me to use

sikieiki Wrote:For example, an excerpt from a card :
巻き立つ＊が、白亜の神殿のようにそびえ立っている。
Answer being 入道雲

Ow, that would be difficult. I think, if you mean to do that MCD style it should be three cards, one for each character in the 入道雲.

If you mean to go with normal cloze deletion quizzing, you should have a hint - a translation of the sentence or the word, or if you're focused on being monolingual, then a synonym or definition of the word, on the front side.

The way it appears it's just not going to be clear in a case like that what goes in the gap (until and unless you have the passage memorized, of course.) Of course, you did say it's an excerpt, so, maybe you had something.

Its not that reading the sentence or understanding the meaning or which word goes in the blank which takes time, its recalling the reading of the hidden word. Its the same problem when speaking when you cant produce the words but you know what it is.

Some word like "equilateral" or "isosceles triangle" can be easily recognized but difficult for to produce. Much like "cumulonimbus" in the example above, I dont think such cards are fit for usage in MCDs. It just takes too long. It seems too much like studying.

- Just realized you mentioned splitting the single word into 3 cards. That might work. I dont know. It might be viable, but spending that much time to learn a single problem word seems inefficient.

sikieiki Wrote:Just realized you mentioned splitting the single word into 3 cards. That might work. I dont know. It might be viable, but spending that much time to learn a single problem word seems inefficient.

The idea is that it takes much less time to learn three separate "easy" cards than one hard card.