Attrition and OLS estimatos

[Blanca Jara]

How can I prove that the attrition bias affects or does not affect my OLS estimator?
Can I do it with a balance table between the group of which I have information and with the one of which I lost information? How would you interpret that balance table? (I have complete data of both groups in other variables)

[Carlo Lazzaro]

Blanca:
you may want to take a look at the community-contributed programme -mcartest- (just type -search mcartest- within Stata to spot and install it) and at missing values theory at large (see -mi- entries in Stata .pdf manual).
Among different textbooks on how to deal with missing data, my favourite one is: https://www.crcpress.com/Flexible-Im.../9781138588318