The following are output from SPSS general linear model univariate procedure. Model 4 had all three predictors (AGE, QUET, and SMK) and two interactions (SMK x QUET and SMK x AGE), while Model 5 had the three predictors only. a. By including SMK and the two interaction terms (Model 4), how much does the R2 number increase from that of Model 3? b. Suppose we have two nonsmokers with the same QUET number, but one is 1-year older than the other. According to Model 4, how much will their SBP differ on average? c. Suppose we have two smokers of the same AGE, but one is bigger than the other by 1 QUET. According to Model 4, how much will their SBP differ on average? d. Using the output from Models 4 and 5, conduct an F test to determine whether the two interactions together are necessary or not using a significance level of 0.05?

Model 3

Tests of Between-Subjects Effects Dependent Variable: SBP Type Ill Sum Source of Squares dif Mean Square F Sig. Corrected Model 4915.630* 5 983.126 16.924 <_001 intercept smk .031 .982 quet .122 age .002 .119 .376 error total corrected a. r squared=".765" estimates dependent variable: sbp confidence interval parameter b std. sig. lower bound upper .005 .537 .248 .426 .344 .545 this is set to zero because it redundant.tests of between-subjects effects type ill sum source squares df mean square f model .001 .066 .549 redundant diagnostics condition variance proportions dimension eigenvalue index .00 .011 .94 .95 .84>