The Computation of Pearson's r from Ranked Data
Journal of Applied Psychology · 1921 · Volume 6 · Issue 4
Abstract
A method of obtaining Pearson's r from ranks is to convert the series of ranks into an array of scores on a convenient linear scale, the linear scores then to be used in the computation of r in the ordinary way. In translating the ranks into linear units, it is assumed that the ability in question is distributed according to the Gaussian law. With Sheppard's table one can determine the per cent of cases falling between the various successive points on the scale. From Psych Bulletin 20:09:00842. (PsycINFO Database Record (c) 2016 APA, all rights reserved)
Enable JavaScript for citation tools and additional paper interactions.