Validation of analysis models depends on what is known of the business domains, and such knowledge is by nature partial and specific, if not hypothetical. Nonetheless, even if general proofs are out of reach, the truth of models can still be disproved by counter examples found among the instances under consideration.
I would like some feedback on that approach.
Validation implies that somehow we are able to "measure" concepts in a generic way and establish causation. Having read 1000+ papers in management, business and economics I come to the conclusion that we are not able to do so for two main reasons. The first is the role of language in science. The second is endogeneity that - despite what statistics and econometrics would like us to belief - cannot be solved meaningfully.
So, your puzzle perhaps starts with the main reason why you want to validate in the first place. Exceptions to the rule from a statistics point of view simply do not matter given that a statistics view works with means and standard devations. Hunting for each and every exception to the rule implies that by definition you reject the opportunity of uncovering laws such as that of gravity in science.
Perhaps the best is to start with the point of departure that exceptions are much more informative and generic cases because exceptions might point to black holes in sciences that inform that mainstream thinking is "wrong".
So you agree: whereas validation is out of reach counter examples can still be used to disprove models.
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity