Language learning and retention rate

[Tom Cato Amundsen]

Hello mnemosyne-users

How high retention rate (as reported by mnemosyne statistics) do you get on average when reviewing cards? I tend to get around 80% and review between 100 and 150 cards a day. Most of my cards are used to learn languages.

I would expect that spending more time reviewing 0 and 1-grade cards before grading them 2 or higher will improve the retention rate. But this also take time, and maybe it would be smarter using that time to read and speak the language I'm trying to learn.

Do anyone have thoughts to share about this topic?

[Vit]

On Friday, April 19, 2013  , Tom Cato Amundsen wrote:

        Hello mnemosyne-users.

How high retention rate (as reported by mnemosyne statistics) do you get on average when reviewing cards? I tend to get around 80% and review between 100 and 150 cards a day. Most of my cards are used to learn languages.

I would expect that spending more time reviewing 0 and 1-grade cards before grading them 2 or higher will improve the retention rate. But this also take time, and maybe it would be smarter using that time to read and speak the language I'm trying to learn.

Do anyone have thoughts to share about this topic?

       --------------
        Hi, Tom.
   On Anki forum,  I found this statement by Damien Elmes:
   "A retention rate of 90% is optimal - much higher and you're doing a lot of work for little gain."   

    My retention % { vocabulary, English } for 6 mo as follows:
1.  ~35% of the columns are at 100% -- in-spite of the fact  that I use mostly 4 and  5 Buttons.
1. Had   4 outliers ( the score was less  than 60%);
2. The median line is about 85%.
3.  The columns swing from 60 to 100%;  the graph  is trendless, erratic.

    I don't see much value in the Retention score - as it is designed; some reasons are:
1. it is calculated on Daily bases. 
2. It includes "Learn Ahead" cards ( increases the score; should not be included ).
3. Stubborn cards (too many Lapses) also included.

     I wish the Score was cumulative,  on weekly and monthly bases.
I stopped looking at it - it is not reliable and not  actionable.

   I wish we had a Retention % charts for the following 7  intervals ( weekly and monthly ):
1.One day Interval .
2.Two, three, four, five, seven and 10 days Intervals.
    These stats would be actionable;   if the score was 90% or more for, let's say, One day interval - that would be a signal to REDUCE or stop the use of 1 day interval and use 2 day  interval instead. The benefit is obvious.
    For now, I track the statistics { Retention for 1 2 3 4 5 7 10  days on paper .
    My hope is that someone like yourself  will  write a plugin ;-).
In the past, you did express an interest in it; my hope is high.

    Vit

[Michael Campbell]

I average about 85% to 92%.

> --
> You received this message because you are subscribed to the Google Groups
> "mnemosyne-proj-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to mnemosyne-proj-u...@googlegroups.com.
> To post to this group, send email to mnemosyne-...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/mnemosyne-proj-users/-/jblaiTUl_gMJ.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[Henrik in Oslo]

Hi Tom Cato

Below is my short answer, however you inspired me to reflect on my learning approach... See separate posting!

I aim at a retention rate no lower than 95%.

If too many items are forgotten, I know that it is likely they will require a lot of time also in the future. I ask myself: 

·       Are they badly formulated? If yes, rephrase/ redo

·       Do I have insufficient  "surrounding" and "supporting" knowledge? If yes: provide this knowledge

·       Some “hard-to-learn” items I simply give a "Priority-2"-tag and put on hold until I am better prepared for this piece of knowledge

·       Other items I re-formulate (would be nice if Mnemosyne allowed "copy-item" and/ or "Reset learning")

·       I may supplement the trouble-item with a different approach (e.g. use picture/ sound).

·       Some of these "hard to learn" items I write down (yes, good old handwriting) and look through every now and then. ( I have even made paper-based flashcards) . I find that changing the learning-context can do wonders... An idea I got just now: Snapshot of the handwritten item, insert into.. etc ☺

/ Henrik in Oslo

[Gnome]

I'm between 70-90% usually, mixed material.

[Oisín]

When I'm not falling behind on reviews, average about 85% in Chinese.

When I'm working under a backlog (often), it's way down, between 70-80% maybe. As you'd expect, this gets worse depending on how long I've been backlogged!

--

You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-u...@googlegroups.com.
To post to this group, send email to mnemosyne-...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msg/mnemosyne-proj-users/-/1DZM6-XbEQYJ.