Once again regarding the statistical significance, what do we want to proof?
Quality of one pill, one bottle, a whole batch, the credibility of a supplier?
However it might seem an obvious one, this is a very important question. Testing the credibility of a supplier by testing one or two bottles might not proof anything. Testing these bottles on a statistical significant way will not help in achieving our goal of testing the credibility of a supplier, on that level our perfect bottle test will still be statistical insignificant. You just cannot compare apples with oranges.
Especially if we want to warn ignorant future customers of this supplier, we will have to be very precise in defining our aim first and after that the testing method that is required to reach that aim. Not the other way round.
The assumption we have now is that in testing one or two bottles, we will find something negative. Maybe we will. But what will that add to the case we are trying to make? As far as I understand, we will only have the possibility to follow this up with legal action initiated by the original buyer of this bottle. As an individual case. To have this test to escalate to the level we want it to be, I think we should be prepared to invest a lot of time in legal activities.
Don’t understand me wrong here, I’m all in favour of doing something. But what are our capabilities in this situation? Is my view to pessimistic?
Edited by brainbox, 27 May 2006 - 11:02 AM.