Family-based assciation study Vs. Population-based case-control study

Question

[MAPK]

I have gone through a few literature to understand family-based association study and population based case-control study. I know family based association study is where we have in-family cases and controls samples and population based case-control study is where we have unrelated samples from the same ethnic group. However, I am still finding it difficult to understand specifically under which circumstances we use one over the other (is it just when the samples are related vs non-related?). Could someone please explain me the major differences of these to study in a few points like you would to a fifth grader. Thanks.

Answer 1

[bio.informatics.016]

One of the most common confounding factors that limit the statistical power of case control study is population stratification. This factor can lead to a false positive or false negative associations. This problem can be avoided by using family based rather than case control study.

[firestar]

In family-based association testing, is it still necessary to correct for family-level stratification?

[MAPK]

If you are looking at association study, you need to remove all the ethnicity specific common variants.

[firestar]

They all belong to the same strain/population. This is not human data. In fact they all have the same great-great great grandparents or something.