- Joined
- Jan 23, 2016
- Messages
- 1,148
- Reaction score
- 1,880
Emphatically yes, it needs to be tested. Because the great majority of posts supporting the notion that these products do something are not logically sound.lol with all these successful comments and experience, do we still really need to test lol.
If you add a bottle to your tank, and are able to successfully establish a tank, this proves absolutely nothing. I can add a glass of DI water from my four year old's purple sippy cup to a new tank, then successfully establish a tank. By this same logic, purple sippy cups help a tank cycle faster.
Two essential factors for determining if a factor actually does something:
1. controls -- isolated tanks established at the same time, identical to the experimental tanks in all respects except for the factor being tested
2. replication -- multiple tanks in each group (multiple controls and multiple identical experimental tanks)
The starting assumption is always that the factor being tested does nothing. Only if the replicate experimental tanks show a consistent difference from the controls is there a logical basis for concluding the factor did anything.
Without those factors there is no logical basis for attributing the success of a tank to the bottle of special sauce the user added. Properly controlled and replicated experiments, or magical thinking.
I look forward to seeing results and encourage you to consider replication in your design.