Why Gaussian statistics are mostly wrong for strategic organization

Abstract

Consider the coasts of England and Norway. They appear jagged no matter what kind of measure is used: miles, kilometers, meters, centimeters, millimeters. This is called ‘scalability’: no matter what the scale of measurement, the phenomenon appears the same. Scalability results from what Benoit Mandelbrot (1983) called ‘fractal geometry’. A cauliflower is an obvious example. Cut off a branch; cut a smaller branch from the first branch; then an even smaller one; and then even another, etc. Now set them all on a table, in line. Each fractal subcomponent is smaller than the former; each has the same shape and structure. They exhibit a ‘power law effect’ because they shrink by a fixed ratio. Power laws underlie fractal geometry. They are ‘indicative of correlated, cooperative phenomena between groups of interacting agents at the microscopic level’ (Cook et al., 2004). They often take the form of rank/size expressions such as F ~ N-β, where F is frequency, N is rank (the variable) and β, the exponent, is constant. In most exponential equations the exponent is a variable. Power laws call for ‘scale-free theories’ because the same theory applies to each of the different levels, that is, the explanation of the generative process is the same across all levels of analysis. The cauliflower is an obvious example of this as well. STRATEGIC ORGANIZATION Vol 3(2): 219–228 DOI: 10.1177/1476127005052700 Copyright ©2005 Sage Publications (London,Thousand Oaks, CA and New Delhi) http://soq.sagepub.com

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 102 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.