M. Nurminen, STATISTICAL SIGNIFICANCE - A MISCONSTRUED NOTION IN MEDICAL-RESEARCH, Scandinavian journal of work, environment & health, 23(3), 1997, pp. 232-235
The P-value is the significance probability of obtaining a value of th
e rest statistic that is as extreme, in relation to the null hypothesi
s, as that observed. Medical researchers may, in some situations, disa
gree on its appropriate use or on its interpretation as a summary meas
ure of consistency with the null hypothesis in a particular data set.
More informative statistical measures such as the likelihood ratio and
the Bayesian posterior probability have been suggested for drawing in
ferences from clinical trials and epidemiologic studies. Causal infere
nce is not statistical in nature; rather it strives to provide scienti
fic explanations or criticisms of proposed explanations that would des
cribe the observed data pattern. In this context, it is important to r
emember that a finding may not be medically important, or a causal hyp
othesis may even not be true even if a study shows a significant P-vaI
ue.