How to design experiments in 3 months time?

Hi there,

I am currently writing a bachelor thesis and building a jypter notebook correlation tool on this topic.

Just wanted to ask your opinion on the best way of going about it: I would love to do a self experimentation together with 2 other friends tracking the following in 3 months time:
Sleep, Activity - Oura
Glucose - FreeStyleLibre
BreathingRate - WHOOP

Interventions before going to sleep:
Meditation, Fasting, IceBath, Magnesium, Chamomile Tea, RedLightTherapy, Honey, and so on…(around 25 possible interventions)

Any suggestions on what the best strategy is? I’ve read that with SingleSubject-Research the ABA Method is quite promising. So it would look something like this:

  1. Week: Normal
  2. Week: Meditation
  3. Week: Normal
  4. Week: Magnesium
  5. Week: Normal

Or do you think I need more data points to have enough significance?
Looking forward to all opinions!

Hello Hermann,

Interesting project :slight_smile:
Although it is unclear for me from your post how you define “improvement” in the diferent categories you are measuring ? A good definition is probably worth > 50% of success for your project, and the answer might not be obvious.
Another advice I’d have is to keep it simple. It can take time for a change to take place, 1 week of action / 1 week off seems very short to me (it depends on what you measure)

Keep us posted !