Hi everybody,

I am currently working on the last phase of my thesis and I struggle with the last part of building my model. I somehow still get insignificant results, but I think that is due to an error in my model specification. Is there someone who has tips to change something in my model to improve it?

I have a cross-sectional database where individuals (level-1) are nested in countries (level-2). I want to test the effect of a level-2 variable on a binary level-1 variable. I used the following command in Stata.

Code:
melogit SEA InstitutionalCol i.age9c i.gender i.EducationSecondary i.EducationTertiary EcoDev, vce (cluster country) or
I use SEA as dependent variable (a dummy, level-1), InstitutionalCol is a 7-point Likert scale (level-2), age9c denotes age categories (level-1), EducationSecondary is a dummy if someone's educational background is on the secondary level (level-1), EducationTertiary does exactly the same but for tertiary educational background (level-1) and EcoDev denotes the economic development level in a country measured in GDP per capita, so in dollars (level-2).

Every tip is welcome. Hopefully, it helps me with creating a better model. Thank you already for your time, much appreciated!