Correction

This article refers to:

Misunderstood and Unattributed: Revisiting M. H. Doolittle's Measures of Association, With a Note on Bayes’ Theorem

Armistead, T. (2016), “Misunderstood and Unattributed: Revisiting M.H. Doolittle's Measures of Association, with a Note on Bayes' Theorem,” The American Statistician, 70, 63–73.

Due to errors during production, Equation (4.2(4.2) $$\begin{array}{ccc}\hfill i_1& =& {[a-(a+b)(a+c)/n]}^2/[(a+b)-(a+b)(a+c)/n]\hfill \\ & & [a+c)-(a+b)(a+c)/n]\hfill \\ & =& {(28-1.8)}^2/(100-1.8)(51-1.8)\hfill \\ & =& 0.142.\hfill \end{array}$$(4.2) ) and Equation (6(6) $$i_2=2[an-(a+b)(a+c)]/\{(a+b)[(n-(a+c)]+(a+c)[n-(a+b)]\},\mathrm{or},$$(6) ) are incorrect. The correct equations are (4.2) $$\begin{array}{ccc}\hfill i_1& =& {[a-(a+b)(a+c)/n]}^2/[(a+b)-(a+b)(a+c)/n]\hfill \\ & & [a+c)-(a+b)(a+c)/n]\hfill \\ & =& {(28-1.8)}^2/(100-1.8)(51-1.8)\hfill \\ & =& 0.142.\hfill \end{array}$$(4.2)

and (6) $$i_2=2[an-(a+b)(a+c)]/\{(a+b)[(n-(a+c)]+(a+c)[n-(a+b)]\},\mathrm{or},$$(6) (6.1) $$i_2=\left\{[a-(a+b)(a+c)/n)\right]+[d-(b+d)(c+d)/n]\}/\{[(a+b)-(a+b)(a+c)/n]+[(a+c)-(a+b)(a+c)/n]\}$$(6.1) (6.2) $$=\left[\right(28-1.8)+(2680-2653.8\left)\right]/\left[\right(100-1.8)+(51-1.8\left)\right]=(26.2+26.2)/(98.2+49.2)=0.355.$$(6.2)

The American Statistician sincerely regrets the error.