Mean Bias on Propensity Score Matching

[Imam Agus Faisal]

Hi, so I am using - psmatch2 - to analyze my data, after I got the results, I use - pstest - and I got this result:

-----------------------------------------------------------------------------------
 Sample    | Ps R2   LR chi2   p>chi2   MeanBias   MedBias      B      R     %Var
-----------+-----------------------------------------------------------------------
 Unmatched | 0.093   3539.25    0.000     20.2      12.1      80.0*   0.51    100
 Matched   | 0.002    225.10    0.000      2.9       2.6       9.6    1.01    100
-----------------------------------------------------------------------------------

I know that I supposed to look at the Mean Bias, I don't know if the numbers are good enough, is there any literature or research that recommends a good Mean Bias or tolerable Mean Bias?