In my analysis I determine the intergenerational transmission of social status. In the literature it is common to use OLS regressions for this purpose. In addition to these regressions, I also calculate the Spearman rank correlation coefficient for two generations, because it does not assume that the correlation is linear.

But now I would like to include the grandparental generation as well. Can the Spearman correlation coefficient be extended in such a way that I can also include this third variable in the analysis?

  • 1
    $\begingroup$ Spearman along with many other metrics of dependence are pairwise only. If you have 3 generations, this would require 3 separate Spearman correlations. Two regression models, one with two generations and a second with three, would permit broader comparisons of dependence. You don't mention this but have you considered an age-period-cohort model framework? Yang and Land's book Age-Period-Cohort Analysis is a good starting point. $\endgroup$ – user234562 May 7 '20 at 16:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.