Regarding datasets

While reading "Hints for using files" section in the "Important Notes for p6", i came across Task D where its mentioned that we have to use "previousapplications.csv" as our secondary data set. Tried to understand what it meant so i came to the conclusion that i might have to perform the univariate, segmented univariate and bivariate task on both the datasets for only Task D. And no other tasks are to be performed on "previousapplications.csv". I just want to make sure that i'm not headed in wrong direction to avoid deduction in marks.