2 replies to this topic

[AgeVivo]

Hello

you may have seen the LongeCIty project idea : "Replicating Baati Through Community Science"
you may have seen the c60 lifespan experiment I am currently doing with 3 mice at home

Some points are starting to be discussed there and there will likely be more before during and after the mprize at home experiment: because we are going to tune everything. I thought we could discuss it here along the way.

Edited by AgeVivo, 15 August 2012 - 11:39 PM.

[AgeVivo]

To start, for now,
- One point I see is how to make a "perfect" blind study: I am planning to do this and I think I see, but comments before and after mine are always welcome.

It's really important for the experiment to be double blinded if we want it to be taken seriously. This means that neither you nor anyone involved connected with the test subjects knows who got the placebo and who got the active compound.

I am planning to consider groups of pets under similar conditions, and make a little computer program that randomly decides who among those pets will receive treatments or placebos. To have someone independent put letters on the bottles and tell me which bottle to send to which participant, and he will make the information officially available for the end of the experiment (and that person would be observed by another person, to avoid issues)

- Another is what measures to look at and how to reach a valid mprize registration: I had looked at it for myself and concluded that it was too difficult to start with. But we can think ahead of time.

http://www.mprize.or...j_mprize_record
(...)
http://www.mprize.or...n=mj_mprize_how
http://en.wikipedia....tion#The_Mprize

Edited by AgeVivo, 15 August 2012 - 11:46 PM.

[AgeVivo]

I found the way to do the randomizing:

the randomizing wil be videotaped! By someone, under the control of someone else. (who? Mind and I won't know). And the video will be put on the web after the very end of the experiment.
In the video a treatment or placebo will be associated to each pet owner, with a rolling dice. To do that I first need to make and "algorithm", and make it public, based on characterics of group of pets/pet owners that is currenlty being built.
The rolling dice will allow to "run the algorithm".

I have to think of how to organize it in fine details: I see a complex solution (sending bottles at least twice), I'll take time to think about a paper-based system another day (other things to do right now).
If some of you like "algorithms", of course please be free to propose such details. Collaborating is fun.