Cross-national comparison best methodology and research questions advice

#1
I am working on my master thesis based on European Social Survey Round 9 from 2018 (link). In a nutshell, I want to use justice principles (the ESS Round 9 is meant to measure four different distributive justice principles, link) to investigate the differences in justice principles between post-soviet states and Western Europe. I was also thinking about exploring how justice principles influence attitudes to inequality (according to theoretical research that principles are essential for forming attitudes).

My main hypothesis is that there should be a significant difference in justice principles between post-soviet states and other European countries.
Hence, my main research question is to see how (1) justice principles differ between post-soviet states and Western Europe and maybe (2) how those differences translate in different attitudes to inequality in post-soviet states and Western Europe (theoretical assumption that values influence attitudes). Basically the survey contains of four justice principles, each measured by three items. Attitudes to inequality are measured by four items, as far as I remember.

Hence my question; what type of statistical analysis you think would be best for the research? I had some in mind, but I will share them yet to see an unbiased answers. Furthermore, if you have any additional ideas about the direction of the research and hypothesis I would be more than happy to hear them. I have conducted thorough literature review, however I still have troubles with settling on methodology and formulating sharp research questions. Arguably the hardest part of the research. Thanks!

I was thinking about using multilevel modelling, basically having individual level, country level and post-soviet state/non post-soviet state level. I don't know though if it makes sense. I'd be happy to hear your thoughts!
 

AngleWyrm

Active Member
#2
I hypothesize attitudes toward justice correlate with annual wages.
Does the survey taker earn more than the median for 1 person? More than ten people? Less than 1/5th?
 
#3
I hypothesize attitudes toward justice correlate with annual wages.
Does the survey taker earn more than the median for 1 person? More than ten people? Less than 1/5th?
I do not know that, but that is something I can definitely control for. However my main focus is on actual differences between post-soviet states and Western Europe to understand how those regions differ in support for principles of justice. Hence also my question: what type of statistical analysis you think would be best for the research? Should I use multilevel modelling with regression? I would be thankful for any advice since it's my first that complex project in statistics and I feel a little bit lost.