[Jakob Andersen]

1. Regarding "Did the interventions affect sickness days?"
1.1) Good.

1.2) So when it says the below it means that my model predicts that at wave 2 (post) the number of sickness days should be slightly higher?
Control#0: 1
Control#1: 1,02

1.2.1) The word "predict" confuses me. Does the coefficients express that there are actual difference between the two time periods?
All my coefficients is strongly significant (0,000), which is quite strange since the

xtpoisson y i.treatment##i.time, fe

command only give slightly significant results for only one of my treatment groups.

1.3) I thought you just said that is this command there is no space for controls?

1.4) I have 4.000 unique observations. When I run these commands it's down to 2000?


3. Regarding "whether perceptions of leadership style influence sick days (regardless of interventions)"
Super!

Although I'm pretty sure my employee_id and time DO identify my observations.


4. Reporting
I'm not sure how to report this in a paper. Would my table look like something below? (I wouldn't put p in parenthesis in the actual paper).

[Clyde Schechter]

You have posted a single table without context, and I don't know which command(s) it is the output from. So I can't interpret it. I think it would be best if you posted the complete output that Stata gave you (including the commands) by copying from the Results window or your log file and pasting into a code block here. (See FAQ #12 for how to use code blocks.) I would be better able to advise you if I saw the whole thing.

[Clyde Schechter]

Additional clarification: please show the actual Stata output directly from the regressions and margins commands, not passed through -esttab- or -estout- or -outreg- or anything like that.

[Jakob Andersen]

I made it work!

Thanks again