Hi everybody!
I am doing a fixed effects panel regression on the country level where I look at the effect of private sector investment on female labor force participation. Among others, I want ot include female education as a control variable, however, I am not sure whether to include primary, secondary or tertiary education as research on this is mixed. So I ran several regressions where I included all three levels at one or seperately and received different results. I don't know, however, what really to do with these results (see below). I also decided to lag education as there possibily are dynamic relationships between enrollment and labor force participation.
In column 1-3, I included either of the three education levels at once. Every time they are significant, however the main independent variable (investment) is only significant in column 3. In columns 4-6, I included several education levels at once. I really don't know what the different results and significance levels mean. In column 3, when I only include tertiary education I get the only significant results for investment, however, that might be biased because when including secondary education investment is not significant anymore (?). Based on these results, is there a way to tell which education variable I should include???
Maybe someone could help me out here?
Thanks in advance!
Best regards
Clara
I am doing a fixed effects panel regression on the country level where I look at the effect of private sector investment on female labor force participation. Among others, I want ot include female education as a control variable, however, I am not sure whether to include primary, secondary or tertiary education as research on this is mixed. So I ran several regressions where I included all three levels at one or seperately and received different results. I don't know, however, what really to do with these results (see below). I also decided to lag education as there possibily are dynamic relationships between enrollment and labor force participation.
In column 1-3, I included either of the three education levels at once. Every time they are significant, however the main independent variable (investment) is only significant in column 3. In columns 4-6, I included several education levels at once. I really don't know what the different results and significance levels mean. In column 3, when I only include tertiary education I get the only significant results for investment, however, that might be biased because when including secondary education investment is not significant anymore (?). Based on these results, is there a way to tell which education variable I should include???
Maybe someone could help me out here?
Thanks in advance!
Best regards
Clara
Comment