A university administrator conducted a multiple regression analysis to predict a student's COLL_GPA as a function of their SAT quantitative score, HS_GPA, the number of hours they study per day (HRS_STUDY), and the number of hours they use a cell/mobile telephone per day (HRS_CELL). Use the two partial ANOVA tables from the regression analysis below to determine how much additional variation (as a percentage) in COLL_GPA can be explained by HRS_STUDY and HRS_CELL, together, over-and-above SAT and HS_GPA: MODEL 1: COLL_GPA = f(SAT, HS_GPA) SST = 20.3 SSE = 13.1 MSR = 3.6 MODEL 2: COLL_GPA = f(SAT, HS_GPA, HRS_STUDY, HRS_CELL) SST = 20.3 SSE = 7.5 MSR = 3.2 2a. 7.59% b. 12.50% c. 36.95% d. 42.75%