quote (carn): "That is the reason why independant confirmation for any result is fundamental. You explain give your yardstick to someone else and he closely looks for errors, can create his own ones and compare them and measure the stones as well. This gives some chances to recognize a wrongly marking..."
i already responded to this quote in my last, lengthy post, but i think it is important, and can be approached from a second angle. last time, i talked about the fact that, since the person double-checking your statistical process, is also trained in and presumably enamored of, to a degree, statistical methodology, he approaches the data with the same in-built bias, the same basic presumption of 'scientificness' or 'objectivity,' that he will tend to interpret things in its favor, where interpretation is possible, and will also simply not
introduce critical gestures with which he is unfamiliar: that is, he can't analyze the study on clinical grounds, as he is not familiar with clinical standards. or, like hans, refuses to even think about them.
but the other way to look at this is from the pov of clinical practice, for 'independent confirmation' is available here as well, because the case study - to take that as a shorthand for the body of technique of clinical practice generally - standardizes
data gathering, analysis, and description of treatment interventions. just to build a bridge to the literature, for example, a case study in homeopathy may describe the repertorization of the case, and reference the grading of symptoms found in boenninghausen. all of the data, from provings to mm, to case studies are drawn upon in presenting the case.
of course there is repetition of method, but by independent practitioners, which is also the way in which innovation is introduced. don't forget, boenni innovated on hahni, and kent innovated on both, and nash developed the characterization and understanding of remedies to a new height, and in the latter part of the 20th century, LM prescribing was (re)introduced....
although there were many splinters - as bigred likes to point out - there is in the mainstream practice a continuous re-implementation, re-calibrating, re-checking, and re-proving of tried technique, and a gradual enrichment of methodology, materia medica, diagnosis ....
it is even the same in history or journalism: independent confirmation, of course, being fundamental to verification of the story. also, closeness to the source, a primary document being preferrable, for the most part, to a secondary account, or a contemporary account preferable to an historical reconstruction, AOTBE.
and it works, of course. so the answer to your general question, how do we know an interpretation is true, is found in analytical standards like this. plus the consideration whether the subsequent intervention works. along the way, there are yes/no questions that pop up, which are usually pretty vital to the process; but the road less travelled by statistics, is the road of life in its fullness - you can not randomize biography, nor therapy, past a certain point, and you can never accomodate, within the randomization, the enormous variety of individual details of treatment, which might as well be infinite for all the chance we have of explaining what has 'just transpired.'
the explanatory power of statistics, in this regard is its achilles heal. it is this vulnerability, this incapacity to analyze objective events to their fullness, that is behind einstein's pan of the process: statistics takes us as close to reality as gross measurement can get, but stops there, with a fine but impenetrable veil draped over the blessed object, the 'it.'
and lest you think it is only me and einstein
that subscribe to such a notion, let me quote feynman, who to his credit grasped the problem and discussed it openly, instead of playing the know-it-all, though he was as well qualified as most to adopt that mantel, had he chosen:
"Another thing that people have emphasized since quantum mechanics was developed is the idea that we should not speak about those things which we cannot measure.... Unless a thing can be defined by measurement, it has no place in a theory. And since an accurate value of the momentum of a localized particle cannot be defined by measurement it therefore has no place in the theory. [But] The idea that this is what was the matter with classical theory is a false position. It is a careless analysis of the situation. Just because we cannot measure position and momentum preciesly does not a priori mean that we cannot talk abou them. It only means that we need not talk about them.... A concept or an idea which cannot be measured or cannot be referred directly to experiment may or may not be useful.... The question is whether the ideas of the exact position of a particle and the exact momentum of a particle are valid or not."
it is appropriate and even admirable to have a goal of measuring homeopathic process with statistics. but the inability to do so is not necessarily a judgement on facts collected by other means.