diff command and dissimilar standard deviations??

[Eva Lan]

Hello everyone

I am using the diff-command for my DiD-Estimation, to including all results in one table.
T1 = a --> b
T2 = a --> c
T3 = b --> c
I receive the same Difference-values, BUT the standard deviations are different...

	(1) Diff. a	(2) Diff. b	(3) Diff. c	delta Diff.
a --> b	2.859*** (0.642)	-0.809 (0.900)	-	-3.668*** (1.105)
a --> c	2.859*** (0.687)	-	2.237** (1.019)	-0.623 (1.229)
b --> c	-	-0.809 (0.876)	2.237** (0.926)	3.046 (1.274)

Aren't the calculations within the same column the same ones?
Is it okay, that the standard deviations are different within my calculations? Or do I have to use another method for my reg. interpretation?

Thank you very much in advance!


The Stata-Outputs:

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Also, do not post pictures - they're hard to work with.

I don't understand why you have the same parameter values for the different dv's. You'll do better if you post your code - I don't know how to understand a ->b etc. I would normally think you've got different dv's but you can't really have different dv's and get identical parameters. Your posted results do have different sample sizes, which should change the standard errors (and the parameters when reported with this many figures.)