It is well documented that active maternal smoking during pregnancy is associated with​ lower-birth-weight babies. Researchers wanted to determine if there is a relationship between paternal smoking habits and birth weight. The researchers administered a questionnaire to each parent of newborn infants. One question asked whether the individual smoked regularly. Because the survey was administered within 15 days of​ birth, it was assumed that any regular smokers were also regular smokers during pregnancy. Birth weights for the babies​ (in grams) of nonsmoking mothers were obtained and divided into two​ groups, nonsmoking fathers and smoking fathers. The accompanying data are representative of the data collected by the researchers. The researchers concluded that the birth weight of babies whose father smoked was less than the birth weight of babies whose father did not smoke.

Nonsmokers Smokers
4194 3522 3454 3998 3455 3066
3062 3771 3783 3150 2986 2918
3544 3746 4019 4216 3502 3457
4054 3518 3884 3493 3255 3234
4248 3719 3668 2860 3282 2746
3128 3290 3423 3686 2851 3145
3471 4354 3544 3807 3548 4104
3994 2976 4067 3963 3892 2768
3732 3823 3302 3769 3509 3629
3436 3976 3263 4131 3129 4263

a. Is this an observational study or a designed experiment? Why?
b. What is the explanatory variable? What is the response variable?
c. Can you think of any lurking variables that may affect the results of the study?
d. In the article, the researchers stated that "birthweights were adjusted for possible confounders …." What does this mean?
e. Determine summary statistics (mean, median, standard deviation, quartiles) for each group.
f. Interpret the first quartile for both the nonsmoker and smoker group.
g. Draw a side-by-side boxplot of the data. Does the side-by-side boxplot confirm the conclusions of the study?