A regression analysis of data on n = 53 males considered the following variables:
Y = SBPSL (estimated slope based on the straight-line regression of an individual's blood pressure over time)
X1 = SBP1 (initial blood pressure)
X2 = RW (relative weight)
X3 = X1X2 = SR (product of SBP1 and RW)
The accompanying computer printout was obtained by using a standard stepwise regression program (SPSS). Using this output, complete the following exercises.
a. Fill in the following ANOVA table for the fit of the model Y = β0 + β1X1+ β2X2 + β3X3 + E.

b. Test H0: ρYX2|X1 = 0.
c. Test H0: "The addition of X3 to the model, given that X1 and X2 are already in the model, is not significant."
d. Test H0: ρY(X2, X3)|X1 = 0.
e. Based on the tests in parts (b) through (d), what is the most appropriate regression model? Use α = .05.
f. Based on the information provided, can you assess whether X1 is a confounder of the X2-Y relationship? Explain.
QUESTION CONTINUE TO NEXT PAGE

Transcribed Image Text:

Source d.f. | SS | MS egressionXX Residual Total 52 DEPENDENT VARIABLE SBPSL VARIABLE(S) ENTERED ON STEP NUMBER 1.. SBP 1 MULTIPLE R R SQUARE STANDARD ERROR 0.45834 0.21007 04428 SUM OF SQUARES 14.79083 55.61661 MEAN SQUARE ANALYSIS OF VARIANCE REGRESSION RESIDUAL DF 13.56308 VARIABLES IN THE EQUATIONVARIABLES NOT IN THE EQUATION BETA -0.45834 STD ERROR B 0.01265 VARIABLE RW SR BETA IN PARTIAL 0.23166 0.23074 TOLERANCE F 0.99553 3,627 .99933 3.611 VARIABLE 0.26007 0.25953 13.563 0.04660 5.10797 (CONSTANT) VARIABLE(S) ENTERED ON STEP NUMBER 2. .RW MULTIPLE R R SQUARE STANDARD ERROR SUM OF SQUARES MEAN SQUARE 0.51332 0.26350 1.01838 ANALYSIS OF VARIANCE REGRESSION RESIDUA DF 2. 50. 9.27620 1.03710 8.94435 18.55240 51.85504 VARIABLES IN THE EQUATION........._..-------VARIABLES NOT IN THE EQUATION BETA 047382 0.23166 BETA İN PARTIAL TOLERANCE 0.04646 STD ERROR B VARIABLE VARIABLE SBP 1 15.174 SR 0.00450 0.00690 0.001 -0.04817 0.02252 5.38484 0.01237 3.627 CONSTANT) VARIABLE(S) ENTERED ON STEP NUMBER 3. SR MULTIPLE R R SQUARE STANDARD ERROR SUM OF SQUARES MEAN SQUARE 0.51334 0.26352 1.02871 ANALYSIS OF VARIANCE REGRESSION RESIDUAL DF 6.18448 1.05824 5.84409 18.55345 5185399 VARIABLES IN THE EQUATIONVARIABLES NOT IN THE EQUATION VARIABLE BETA STD ERROR B BETA IN PARTIAL TOLERANCE F VARIABLE SBP 1 RW SR CONSTANT) 0.01391 0.14372 0.00122 0.04798 0.01801 0.18527 0.001 5.36183