The confidence degree sustained by a statistical inquiry which shows a sufficient divergence from purely random effects.
The matter is to determine what is a "sufficient divergence". On the borderline, some significant effect may seem to emerge, but may also remain somewhat (or largely) hypothetical.
On the other hand, statistical significance must be interpreted, i.e. given a meaning. At roulette, a 18/20 appearence of red would be statistically significant. But what does it mean? Has somebody "nursed" the wheel? which way? Why?, etc… These are systems problems.
From a systemic viewpoint, as stated by W.D. GROSSMANN and K.E.F WATT: "… several phenomena of great recent importance are strongly dependent on the behavior of a very small proportion of all individuals, at the tails of frequency distribution curves. AIDS became a pandemic in part because of the atypical behavior of a tiny proportion of all individuals, not the average population-wide behavior" (1992, p.20). The authors give various other examples of the shortcomings in the use of statistical average values (and of grossly misleading financial accounting (see "Patrimonial accounting").
This problem reflects an insufficient understanding due to excessively narrow frames of reference used to give significance to mere data and information.
- 1) General information
- 2) Methodology or model
- 3) Epistemology, ontology and semantics
- 4) Human sciences
- 5) Discipline oriented
To cite this page, please use the following information:
Bertalanffy Center for the Study of Systems Science (2020). Title of the entry. In Charles François (Ed.), International Encyclopedia of Systems and Cybernetics (2). Retrieved from www.systemspedia.org/[full/url]
We thank the following partners for making the open access of this volume possible: