Standardized mean difference > 1.0

[Reinout Heijboer]

Dear all,

For my most recent study I have done a propensity score matching 1:1 ratio in nearest-neighbor without replacement using the psmatch2 command in STATA 13.1.

When checking the standardized mean difference (SMD) before and after matching using the pstest command one of my variables has a SMD of 140.1 before matching (and 7.3 after).
Is it possible to have a SMD greater than 1.0, or is this an error?

Thank you in advance.

Best,

Reinout Heijboer

[William Lisowski]

I've installed psmatch2 (user-written software from SSC) and have the following version.

. which psmatch2
*! version 4.0.11 22oct2014 E. Leuven, B. Sianesi

The output of help psmatch2 and help pstest do not suggest that a "standardized mean difference" is reported. Instead, the one appearance of the word "standardized" in the two help files occurs in the following, from help pstest describing its reporting of an "absolute standardized difference of the means":

Rubins' B (the absolute standardized difference of the means of the linear index of the propensity score in the treated and (matched) non-treated group) and Rubin's R (the ratio of treated to (matched) non-treated variances of the propensity score index). Rubin (2001) recommends that B be less than 25 and that R be between 0.5 and 2 for the samples to be considered sufficiently balanced. An asterisk is displayed next to B and R values that fall outside those limits.

I had hoped to use psmatch2 and pstest to do an analysis and look closely at the results, but I could not determine the source of the data to which their sample code applies.