Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: Blinder-Oaxaca decomposition - Interpretation of output

 From Bettina Lamla To "statalist@hsphsun2.harvard.edu" Subject st: Blinder-Oaxaca decomposition - Interpretation of output Date Tue, 4 Dec 2012 10:44:38 +0000

```Dear all,

I perform a decomposition analysis using the Oaxaca command from Ben Jann. I have Stata 12.1.

In every specification I find the group difference to be statistically significant, however neither endowments, coefficients not the interaction between these two seem to explain the difference (not significant). I struggle with the interpretation. Can anyone help?

Thanks!

Here is a part of the output:

Blinder-Oaxaca decomposition
Number of obs   =   1076
Model           =         linear
Group 1: east = 0                                 N of obs 1      =        690
Group 2: east = 1                                 N of obs 2      =        386

------------------------------------------------------------------------------
flit3 |           Coef.       Std. Err.      z          P>|z|     [95% Conf. Interval]
-------------  +----------------------------------------------------------------
overall        |
group_1 |   .5391304   .0192225    28.05   0.000     .5014551    .5768057
group_2 |   .4378238   .0258589    16.93   0.000     .3871413    .4885063
difference |   .1013066   .0322209     3.14    0.002     .0381548    .1644584
endowments | .019734   .0314114     0.63    0.530    -.0418313    .0812993
coefficients |   .0345781   .0352739     0.98    0.327    -.0345575    .1037137
interaction |   .0469945   .0349906     1.34   0.179    -.0215857    .1155748

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```