Research: Create a meta-analysis demonstrating the superiority of SuperMemo's approach

[alexis-]

• Include existing research
• Use SMA as a platform to collect statistics (with user consent) ?
• The three paradigms shifts of SM
  • Spaced repetition
  • Find a way to demonstrate concisely why SRS is so powerful
  • Explain what it entails in every day usage
  • Incremental reading
  • Find a way to demonstrate concisely why IR is so powerful
  • Demonstrates how it materializes
  • Extracting and re-arranging imported/existing text is much faster than typing everything (e.g. compare with Anki or regular note taking)
  • Web imports (wiki, news websites, etc.)
  • Books (pdf, epub, ...)

[candywal]

These are some ideas on how to take these foward.

With regards to SRS and Active recall there is a lot of research in cognitive science already supporting it and displaying it's supremacy so we can refer to that research, I think at a certain point Woz even offered a free copy of SM if you sent the SM team statistics: https://www.supermemo.com/en/archives1990-2015/articles/survey here is some data on the SM website. So this I'm pretty sure we can demonstrate rather easily with secondary Data and some extra research - I mean it's hit the mainstream.

For the second one, it takes a while to get the hang of IR, so maybe we could assign reading of a topic for a group of people with Incremental reading, and a group without which worked for 20minutes a day each on the material. Maybe a seperate collection for SM users and then test on memory and comprehension at the end? The issue is where will we find the non-SM users willing to try this project?

In terms of arranging and aquiring knowledge, maybe a few case studies showing how easy it is for several text imports to be set up in supermemo, and how much retention a normal SM user has of texts they read a while in the past? Also, we can compare two people making notes on a text using classical methods of their choice versus a SM user processing the text with SM then test on comprehension now, and in a random period later around a month two compare the retention of the knowledge and the use of the notes. Maybe give the two groups a time period to process a text and see who was able to process more, and test for comprehension?

[rajlego]

I think the most important thing to test and prove is just how much more efficient the SRS component of SM is than Anki. I've had plenty of Anki users ask for proof and while there are plenty of people with anecdotal proof of SM being better, there's no hard data from a direct experiment that I can show them.

Probably the easiest thing to do for comparison is to get a language learning deck and have the same person do half the deck in Anki and the other half in SuperMemo and compare repetition loads/retention after a few months.

When you posted the survey I had an idea: not ideal now but it would be cool if we could add in a survey to SMA and collect data on people's usage habits/SM usage statistics.