1.Using the EXERCISE data set, create a new variable PUL GAIN which is the difference in the post-exercise ( PULSE 2) and pre-exercise pulse ( PULSE 1) rates. Create four groups; male smokers, female smokers, male nonsmokers, and female nonsmokers.

(a) Write down a contrast that compares

i. smokers and nonsmokers

ii. female and male

iii. female smokers and female nonsmokers

iv. male smokers and male nonsmokers

(b) Use the Bonferroni method to test all the comparisons in (a). Consider = .05.

(c) Use SAS to verify your results from (b).

Analysis Variable : PUL GAIN N newfactor I Mean Variance MALESMOKER 58.4 110.78 FEMALESMOKER 10 35.4 913.82 MALENONSMOKER 10 10 56.60 174.27 FEMALENONSMOKER 10 10 63.80 72.40 Model Information Test for continuous variables Mean t-test Degrees of Freedom Method Pooled Tails for continuous tests I wo-tailed Strata weights Non P-value adjustment Bonferromi Contrast Coefficients newfactor Contrast MALESMOKER FEMALESMOKER MALENONSMOKER FEMALENONSMOKER smokers and Centered -1 -1 nonsmoker female and male Centered female smokers Centered and fe male smokers Centered 0 -1 and mal Continuous Variable Tabulations Standard Variable newfactor Num Qbe Mear Deviation PUL GAIN MALESMOKER 58.4444 10.5251 PUL GAIN FEMALESMOKER 10 85.4000 30.2293 PUL GAIN MALENONSMOKER 10 56.6000 13.2010 PUL GAIN FEMALENONSMOKER 10 63.8000 8.5088 p-Values Variable Contrast Raw Bonferroni PUL GAIN smokers and nonsmoker 0.0498 0.199 PUL GAIN female and male 0.005 0.0219 PUL_GAIN female smokers and fo 0.0110 0.0441 PUL GAIN male smokers and mal 0.8247 1.000