st: Reverse Causality calculation methods

[Alexandra Grabbe <alexandra.grabbe@googlemail.com>]

Hello everyone,

I am currently working on my Master's Thesis, analysing the impact of
sporting activities on job satisfaction using panel data over five
years.
Including additional control variables in the model, my regression
identifies Sport as a statistically significant factor influencing
satisfaction at work.
As a further step I would now like to exclude a possible "reverse
causality", such that job satisfaction affects sports instead of vice
versa.
Does someone have an idea how I can approach this problem?
(As a side note: the data consists of appr. 36,000 observations)

I have thought of analysing certain individuals of the data set by
looking at the change of their sporting behaviour and their job
satisfaction. But I am not sure of how to do this technically with
Stata. Another option would be to identify instrumental variables
describing sports, but this would be quite difficult and only
approaches the problem to a certain point. Thus, new ideas are
welcome. Please help!

Alexandra
*
*   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/