[Maarten Buis]

I had previously estimated the following regression which is similar to what you recommend, the only difference being that I also interacted sa4 and year with unemp_av6. Is that incorrect?

That is not incorrect, it is just different. I would be hesitant to add that many interactions, but it is not wrong per se.

Also, wouldn't I divide the p-value by two instead of multiplying it?

That is correct, sorry.