Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.


Luis A. Afonso
Posts:
4,699
From:
LIsbon (Portugal)
Registered:
2/16/05


The insignificance of some antiNHST people
Posted:
May 15, 2013 10:38 AM


The insignificance of some antiNHST people
When do exists someone, somewhere, sometime, that is concerned about the Null Hypotheses is true or untrue in strictly sense as a fundamental topic of NHST, I am prone to a stringent and irrepressible desire to send him to revise theory without delay. No matter it seems odd/heretic the objective/aim of Hypothesis Tests is rather different. In fact we are moving in other world, that of systemic/unavoidable uncertainty at which the Aristotelian duality true/untrue is banned as completely out of context. The Logic, not Deductive but Inferential, is other: concerning the result of an observation/experiment it did or didn´t brought/provide sufficient evidence that our previous knowledge was correct? Instead the uninspiring: all humans are mortal, Socrates is human, therefore Socrates is mortal, we proceed from general premises to a necessary and specific conclusion: all birds have ails, this animal does them, then it is likely a bird, I risk. Because it could be a mammal (bate) or an insect (bug) an error is inherent. Throughout statistical tests we cannot prove, by all means, that the null hypothesis is true or not: try to calculate how much splits you have to make in order to prove a coin is fair, so the probability to heads occurrence is exactly ½.
Luis A. Afonso



