Dear all

I have a hierarchical mass opinion survey data, 83,961 individuals nested in 21 countries. My two main variables of interests are 1) meritocracy in the public sector (var name= justice_public) and 2) meritocracy in the private sector (var. name=justice_private). The data is collected from citizen survey in 21 countries.

I'd like to show that there exists statistically significant differences in meritocracy in the public sector and meritocracy in the private sector. I believe that the following paired t-test is one thing I should try.

However, I would like to hypothesize that there exists statistically significant differences in the degree of meritocracy between public and private sectors in any sample countries regardless of country level differences measured by GDP/capita and levels of corruption in the public sector. In fact, these 21 countries differ a lot in terms of these two country level measures. However, I am not sure if simply showing each country's t-test allows me to confirm this hypothesis. I wonder if there is anything I can do beyond this using STATA. I do not include results of t-test for each country here, but all results show statistically significant differences in two variables.

Regression analysis is not an option for me as the data structure does not allow me to do so. If the above meritocracy variable is asking meritocracy in respondent's own working organization, I can do regression analysis using a working sector as an independent variable and meritocracy as a dependent variable, and then I can claim that being a public sector worker shows a positive impact on meritocracy, controlling for some country-level factors. However, I am not able to do this due to the data structure.

Any suggestions would be greatly appreciated. Thank you very much in advance.

Code:
. ttest justice_private == justice_public
ttest justice_private == justice_public if country_name=="Austria"
ttest justice_private == justice_public if country_name=="Belgium"
ttest justice_private == justice_public if country_name=="Bulgaria"
ttest justice_private == justice_public if country_name=="Croatia"
ttest justice_private == justice_public if country_name=="Czech Republic"
ttest justice_private == justice_public if country_name=="Denmark"
ttest justice_private == justice_public if country_name=="Finland"
ttest justice_private == justice_public if country_name=="France"
ttest justice_private == justice_public if country_name=="Germany"
ttest justice_private == justice_public if country_name=="Greece"
ttest justice_private == justice_public if country_name=="Hungary"
ttest justice_private == justice_public if country_name=="Ireland"
ttest justice_private == justice_public if country_name=="Italy"
ttest justice_private == justice_public if country_name=="Netherlands"
ttest justice_private == justice_public if country_name=="Poland"
ttest justice_private == justice_public if country_name=="Portugal"
ttest justice_private == justice_public if country_name=="Romania"
ttest justice_private == justice_public if country_name=="Slovak Republic"
ttest justice_private == justice_public if country_name=="Spain"
ttest justice_private == justice_public if country_name=="Sweden"
ttest justice_private == justice_public if country_name=="United Kingdom"

ttest justice_private == justice_public if country_name=="Austria"