Citation

Peters CC. Am. J. Sociol. 1933; 39(2): 231-236.

Copyright

(Copyright © 1933, University of Chicago Press)

DOI

10.1086/216374

PMID

unavailable

Abstract

Ross's criticism of the Faris investigation of insanity rates, in insisting upon a critical ratio of exactly three of more, makes appeal to an element of magic which is not warranted by the actual nature of reliability measures. Ross, further, does not take cognizance of the increased reliability due to the fact that the differences are prevailingly in the same direction. In computing standard errors it is important to take account of the element of correlation.