A problem of interest to health administrators is to determine the effects of smoking (cigs) during pregnancy on infant health. One measure of infant health is birth weight (bwght in ounces); a birth weight that is too low can put an infant at risk of contracting various illnesses. To look at this question you write down the following simple linear regression equation: bwght=β0+β1cigs+error

(i) What is the most likely sign for β1? Explain.

(ii) Interpret your model estimates below for β0 and β1 in full sentences, appropriate for a report. Also, what if a pregnant woman smoked 15 cigarettes per day; what would be the predicted reduction in the birthweight (grams) of her child? The p-values are statistically significant and the equation for this output is:

bwght = 119.81 - 0.479cigs + error 

(iii) The R-squared value is 0.022. Interpret.