A closer examination of three small-sample approximations to the multiple-imputation degrees of freedom,erratum

In the print and electronic versions of Wagstaff and Harel (2011, Stata Journal 11: 403–419), the equation at the bottom of page 405 appeared as

$v_{LPZ}=\frac{{[\overline{W}+\{(m+1)/m\}B]}^2}{\{{\overline{W}}^2/N-1\}+{\{(m+1)m\}}^2\{B^2/(m-1)\}}$

but should have appeared as

$v_{LPZ}=\frac{{[\overline{W}+\{(m+1)/m\}B]}^2}{\{{\overline{W}}^2/(N-1)\}+{\{(m+1)/m\}}^2\{B^2/(m-1)\}}$

That is, the formula should read ${\{(m+\text{1})/m\}}^{\text{2}}$ and not ${\{(m+\text{1})m\}}^{\text{2}}$ in the denominator. The erroneous expression was not used in the simulation, so the results stand.