Questions 3-5. Suppose a regression model is used to investigate respiratory health in children. The outcome variable is forced expiratory volume (FEV), a numerical measure of how much air a person can exhale during a forced breath. The exposure covariates are a binary indicator for allergies (0=no allergies, 1=any allergies) and age (in years). The output table for the regression is given below. Note that the p-value for the age parameter has intentionally been obscured. Variable Coefficient Std Error 95% Lower 95% Upper p-value allergies -0.21 0.08 -0.37 -0.05 0.0001 age 0.23 0.01 0.21 0.25 XXX 3. How do we interpret the effect of allergies on FEV based on this model? a. Each 1 unit increase in FEV decreases predicted number of allergies by 0.21. b. Controlling for age, each additional allergy a child has reduces predicted FEV by 0.21. c. Predicted FEV in children with allergies is 0.21 lower than children without allergies, adjusting for age. d. Controlling for age, the predicted FEV is higher in children with allergies. e. The effect of allergies on FEV is not significant after controlling for age. 4. State in words the null and alternative hypotheses corresponding to the obscured p-value. (Open Response, two sentences) 5. Based on the regression output table, which of the following must be true? a. The obscured p-value is less 0.05. b. The obscured p-value is greater than 0.05. c. The relationship between FEV and allergies is confounded by age. d. The relationship between FEV and age is confounded by allergies. e. None of the above