|
NC IA | P value does equate to the “success rate” of the product. It just tells you the trial data is X% accurate.
So if becks does 6 replicated trials, gets +5bu response in 4 trials and -2bu in 2 trials it will have a higher P value than if they conduct 12 trials and have +5bu response in 8 trials and -2bu in 4 trials. The win rate is the same, but the P value is lower because the sample size is higher.
Basically, when more replicated trials are conducted, the data becomes more statistically valid because the sample size is higher.
Hope that makes sense. | |
|
How much do you believe the Becks PFR trials?
