Hi everyone,

I new to this forum so i hope I'm not violating any forum rules. I've the following question.

I'm currently doing rounding up my thesis and conducting multi-level modelling and trying to look what the effect is of board gender diversity (BGD) on corporate social performance (CSP) being moderated by cultural values (hofstede's cultural values).

As I'm new to multi-level modelling, I interpreted my research question as follows.

The dependent variable (corporate social performance) is a firm variable which is observed over several years (panel data, level 1)

Firms are nested in industries (level 2)

Industries are nested in countries (level 3)

However, I do believe that the industry hierachy is not of interested for my research question. Is it important to also include the industry hierachy in the stata formula? My professor did mention that I cannot ignore the industry hierarchy as it explains variation in the dependent variable.

I generated the following stata command. Would this suffice or should i also let the slopes for the industry hierachy randomly vary?

mixed CSP c.BGD##c.DPD c.BGD##c.DID c.BGD##c.DMAS ROE ROA TBQ LEV BINDEPD BSIZE BSTRUC CEODUA logTA Year || GEOL:BGD, covariance(independent) || ICBIN: