Ok so I know nothing about econometrics or research methods but I have this project for my Humanomics class and I'm totally lost on the data analytics part.

We are trying to write a research proposal on a project that uses twin studies data (TEDS) to evaluate the impact of parents highest education level on their child's likelihood to have depression while controlling for heritability of depression, as well as controlling for other factors such as gender and race. My professor suggested that we use a diff in diff estimate to control for heritability but noted that race would not be controlled for between a set of twins and would just be a variable across families. The data is like 95% white people so we kind of need to control for race but I have no idea how to control for intra-twin differences and family differences simultaneously. I also don't really understand differences in differences estimates to be honest so any tips or help is much appreciated.

Also, we have to do all of this without knowing what the data would look like bc we cant access it, all we know is that the race of the twins, gender of the twins, parental education level, and type of twin (fraternal or identical) was recorded. We have to write a proposal of how we would analyze the data.
And I only started using stata less than a month ago and our prof just expected us to teach it to ourselves so I have no idea if this is a dumb question or not.

My original plan was to use multivariate logistic regression for this but the prof told me to use dif in dif estimates and I'm lost. SOS