0

I want to find out if economic sanctions effect the educations system of sanctioned countries. There are some studies that have worked already on providing evidence that sanctions negatively effect the health condition of countries, so I wanted to take a look if the same can be observed with education. My teacher/supervisor told me to use the OLS model for that. I have watched tons of videos on OLS and I understand the basics of it, I just don't know how to implement it on my data. So far I figured out that my independent variable should be countries' GDP connected to the economic effect of sanctions somehow; then my dependent variable would measure education outcomes - which could be for instance school enrollment, drop-out rates, or total years of schooling. Control variable I have theorized also - for instance, was the sanctioned country also effected by war/violent conflict, what's the political situation of the sanctioned country, ... and so on - all possible things that could also have an effect on education.

I am stuck on the practical implementation. I know how to merge and prepare my datasets: for instance, for a scatter plot, I would have my independent variable on the y axis and my dependent variable on the x axis.

My teacher said that I should not bother with time series (I first thought comparing before and after sanction entered the education variables) he explained me this is anyways included in my country observations of OLS - that is something I don't understand.

My data include all sanctions that have been and are being carried out against countries, which have different years, and I have all education outcomes for each sanctioned country observed throughout the years. How would the OLS look like? I understand the basic principle of it - when putting money paid for fuel and total km driven with that fuel e.g. but my imagination is not able to implement my study for this model and that s where i am asking for help - I would just need someone to explain what I put where, then I can go to work it out myself in R.

I have 0 experience with statistics so please be nice :)

Marco Plebani
  • 779
  • 4
  • 17
JacquelineANP
  • 1
  • I suspect that back-and-forth online communication to work this out will be difficult for both parties and that you would be best served by a real-time conversation in which both people can view the data, code, and output in question. – rolando2 Aug 01 '21 at 16:15
  • You need to give us more information about your data. Please list all the variables, and then draw a DAG to show us how they are all related in your mind. After drawing the DAG we will be able to help you further. Please edit the question, and add the information about the variables, and the DAG. See [here](https://stats.stackexchange.com/questions/445578/how-do-dags-help-to-reduce-bias-in-causal-inference/445606#445606) for details about how to draw the DAG. – Robert Long Aug 01 '21 at 18:53

0 Answers0