n the Heart and Estrogen Replacement Study, a research question of interest was to quantify the effect of physical activity (1 to 5 increasing scale) and drinking status (1=any drinking, 0 = no drinking) on serum glucose levels at year one of the study. These data were analyzed using two different models. Results from the analysis are provided below labeled SAS Output 1 and SAS Output 2. To answer some of the questions that follow, you need to fill in some critical pieces of information that have been deleted.

For questions 45-52, refer to the following study and the SAS output on the next page: In the Heart and Estrogen Replacement Study, a research question of interest was to quantify the effect of physical activity (1 to 5 increasing scale) and drinking status (1=any drinking, 0 = no drinking) on serum glucose levels at year one of the study. These data were analyzed using two different models. Results from the analysis are provided below labeled SAS Output 1 and SAS Output 2. To answer some of the questions that follow, you need to fill in some critical pieces of information that have been deleted. SAS OUTPUT 1 Dependent Variable: GLUCOSE1 year 1 fasting glucose (mg/dl) Standard Parameter Estimate Error t Value Pr > It) Source DF Sum of Squares Mean Square F Value Pr > F Intercept 133.5214559 B 3.32553319 40.15 <.0001 model physact b error corrected total drinkany r-square coeff var root mse glucose1 mean least squares means for effect source df type i ss square f value pr> F Pr > It] for HO: LSMean(i)=LSMean(j) PHYSACT GLUCOSE1 LSMEAN LSMEAN Number Dependent Variable: GLUCOSE1 PHYSACT 4 99414.06422 24853.51605 13.17 <.0001 i drinkany w n source df type ill ss mean square f value pr> F w 114. 184838 0.7301 0.1223 0.0014 <.0001 in physact drinkany sas output dependent variable: glucose1 year fasting glucose standard parameter estimate error t value pr> It) Source DF Sum of Squares Mean Square F Value Pr > F Intercept 136.1311475 B 3.93349578 34.61 <.0001 model physact b error corrected total r-square coeff var root mse glucose1 mean drinkany source df type i ss square f value pr> F DRINKANY 0 0.0000000 B PHYSACT 4 99414.06422 24853.51605 13.17 <.0001 physact b drinkany source df type ill ss mean square f value pr> F PHYSACT*DRINKANY 3 1 11.0550220 B 7.76171494 1.42 0.1545 PHYSACT 4 72177.32993 18044.33248 9.56 <.000 physact drinkany b what type of analysis should you perform to test the given hypothesis a. ancova b. twoway anova c. linear regression d. logistic e. one-way in sas output number insert for error degrees freedom which model is more appropriate these data: or statistic and p-value use make this decision because interaction signicant pvalue .0001 not value mean square sum squares dividing by df we obtain omnibus null ho do any physical activity groups differ significantly glucose yes group differs from all significantly. choices c d are true. consider estimated can be written as: y="133.52" p5 p4 p3 p2 d1 independent variables equation above defined as follows: if subject category otherwise a regular drinker level at year who shown figure below fasting comparative current alcohol consumption there same direction interaction. reverse no lines fit data parallel. differences between irregular drinkers each approximately equal difference depends strongly on>