Re: Springtime Fry growout event may happen
It might be interesting to start with a survey of what parameters folks think should be tested.
Re: Springtime Fry growout event may happen
Quote:
Originally Posted by
RogueDiscus
It might be interesting to start with a survey of what parameters folks think should be tested.
Thats a good thought Steve. Something we can definitely do.
Re: Springtime Fry growout event may happen
Quote:
Originally Posted by
Mando
I like this idea, however, you'll need a lot more than 8 participants. Like you mentioned, you have 3 Iv's with two levels each. You'll need a group of participants for each level to get credible results. And, you'll need another group to compare against. Maybe the control group could follow what we all follow today to raise our discus - hobby standard. Considering the amount of participants needed, time, effort, and the amounts of fry that would have to be donated, I'd go with Al's idea to slowly collect the data we essentially want from your experimental method, but in a slower scale. What I understood from Al was that he wants to test one or the other. Either Holding WC's constant and changing the foods or vise versa. This method requires about 8 participants.
I'd love to participate on this spring challenge. I wish I could donate fry, but my pair are not good parents.
For three variables at two levels the minimum number of trials is 8, each participant will have 10 discus so giving you multiple trials for each condition. More is better obviously. As a scientist optimizing how drugs are made, we use these principles often. They have repeatedly been shown to be more rapid, robust and predictive to establish both optimal conditions and acceptable 'design space' than the more traditional and logical, 'one variable at a time' type of experimentation. You can get more value from fewer trials by applying 'design of experiment' principles, assuming it is designed well.