Answer Happy

3) Using data from a lifespan study a researcher has found the surprising result that cigarette smokers have a much lower mortality risk as compared with pipe smokers. Before submitting the result for publication, she decided to take a closer look. She made a histogram of the ages of the cigarette smokers (solid) and the pipe smokers (dotted line). Histogram of ages Normal 0.09 status 0 1 0.08 0.07 Mean StDev N 41.00 5.143 1029 73.96 4.741 971 0.06 0.05 Density 0.04 0.03 0.02 0.01 0.00 32 40 48 64 72 80 56 ages Noticing the differences in the age distributions, the researcher says “Aha, I will need to control for age in my regression model. My new model will predict mortality using smoking status AND age as predictors.” Based on our class discussions of causal inference, what could go wrong in her analysis?