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:
4,725
From:
LIsbon (Portugal)
Registered:
2/16/05


Are there insignificant people ...
Posted:
Jun 19, 2013 2:55 PM


Are there insignificant people ... First of all a caution note: The matter is strictly the Null Hypotheses Significant Tests, never its usefulness or not in the Douglas H. Johnson´s investigation branch to which, surely, I would not dare to write a line even; the matter his article ?The insignificance of Statistical Significance Testing?, 1999, Journal of Wildlife Management 63(3) : 763772. _______________
Conclusion ? The most common and flagrant misuse of statistics, in my view, is the test of hypotheses, especially the vast majority of them known beforehand to be false ? . (end of citation).
My comment Doubtless the author intend to demand that an assertion posed by the null hypotheses should be answered by a yesno conclusion. In fact everybody agree that it´s a structural error only possible to be done by strongly apart of the inductive (not deductive, of course) logic commanding random events. When we perform a NHST we do not aspire to state if H0 is true or not: yet Ronald Fisher said (a century ago ) that we are able only to measure, by its probability, how likely an assertion is: not less, not more. To those that ignore this basic principle would be expected sufficient wise to stay silent about and not to fight against what is well established forever. . Luis A. Afonso



