“Is Individual Job Performance Distributed According to a Power Law? A Review of Methods for Comparing Heavy-Tailed Distributions”, Seth M. Spain, Marcus Credé, P. Harms, Bradley Brummel2017-09 (economics, order statistics)⁠:

It has recently been proposed that individual job performance follows a power law distribution (O’Boyle & Aguinis2012).

We present an argument and evidence for why the conclusion does not follow from the premises. We discuss the nature of generating mechanisms of statistical distributions, and compare the normal, log-normal, and Pareto distributions.

We review statistically principled methods of testing power-law distributions, and point out how it is necessary to compare them to a plausible alternative distribution (Clauset et al 2009). We reiterate the importance of testing the assumptions of statistical models, and review the methods that are available to organizational researchers when the norm of normality is violated.

[Keywords: distribution theory, statistical assumptions, power laws]

See Also:

• Scale-free networks are rare

• Modeling bursts and heavy tails in human dynamics

• Emergence of good conduct, scaling and Zipf laws in human behavioral sequences in an online world

• Power Law Trends in Speedrunning and Machine Learning