DIY Experiments: Prototyping
Why join now? We have support this summer for data analysis & visualization for people that do prototyping – join ASAP!
We’d love to have you join us! 😍 To join, you can…
(1) Use our Slackin bot to make an account, then (2) join the #diy-experiments channel in Open Humans Slack to chat.
(3) Add your experiment to this doc and you’ll be upgraded to the core collaborating team. 😉
(4) You can join and talk with others in our weekly videochat meetups:
- US meetup: Thursdays 12pm Pacific / 3pm Eastern https://meet.google.com/wsd-zvci-jjx
(5) Record updates in the Quantified Self forum Project Logs
N-of-1 self-experiment with personal data.
Do you have a question about yourself? Set it up in an A/B format, and collect data to answer your question.
- Question: “Does alcohol impair my sleep?”
- Test: Alternating weeks – zero alcohol vs. a beer every evening.
- Collect data: Sleep tracker + morning sleep satisfaction survey
- Record data about the thing you want to change. Does listening to music affect your mood? Record data about your mood.
- Test an intervention. Our default recommendation is to test between two conditions (A vs. B) and alternate between these weekly, for a total of six weeks (three weeks for each, with a pattern of ABABAB).
- Check your data ASAP. You’re collecting data – but is it good data? Share a copy with our analyst support for feedback. Some data sources end up being flawed! You don’t want to waste 6 weeks collecting bad data.
- Also track “adherence”. In addition to collecting data about the outcome (i.e. about what you think will be affected, like mood or sleep quality), consider collecting data about how well you stick to your plan. This is also called “adherence”. Knowing when you stuck to the plan – and when you didn’t – may help with data analysis.
Real examples so far…
Some folks are logging their project in the Quantified Self forums “Project Logs”:
We aim to ask questions and observe a change in data we’re measuring. (Or no change!) We can use data sources already supported by Open Humans, but don’t feel limited to it. Do what works for you, even if that means tracking it manually!
We’re inspired by how valuable self-experimentation can be as a method for discovery. For more inspiration, see Allen Neuringer’s decades of self-experimentation shared at the 2018 Quantified Self Conference.
We also dream about the scale-up: we want to see community-led research. But research starts with you – answering your own question is exciting! We think that’s the natural starting point for growth. A 1-pager here describes this idea in more detail.
We hope this prototyping will result in…
- Improved data imports in Open Humans
- A set of open source data analysis notebooks that can be re-used in Open Humans
- Most importantly, a much better idea of how to design a self-experimentation site/ecosystem