A health researcher studying child birth is interested in the possible effect that smoking by the mother during pregnancy has on the baby's birth weight. To investigate, the researcher conducts an observational study by reviewing 400 records of women in their twenties who have recently given birth for the first time. For each record, he notes whether or not the woman smoked during pregnancy and the baby's birth weight, in addition to the woman's education level, and age. From the data, the researcher creates two groups: women in their twenties who smoked during pregnancy and women in their twenties who did not smoke during pregnancy. Then he compares the average birth weight between the two groups. (a)Why might the researcher have chosen to perform an observational study (by gathering information from past records) and not a randomized experiment (by assigning pregnant women to either the smoking or nonsmoking group at random)? Choose the best answer from the choices below. In a randomized experiment, it could be that some pregnant women who do not smoke would be assigned to the smoking group, meaning that they'd be forced to smoke. It would not be ethical to force people to smoke. For a randomized experiment to be performed, the researcher must ask women in the population to volunteer to take part. This would mean that there would not be any chance to include women who smoked during their pregnancy in the sample. A randomized experiment should never be performed when it is possible to perform an observational study. When groups are randomly assigned, the researcher cannot control the types of participants in each group, making the results of a randomized experiment unreliable. (b)The variable age of the woman is not a possible confounder in this study. Choose the best reason why. A variable is a confounder if its effect on the outcome cannot be distinguished from the effect on the outcome from different treatments. (In our context, the outcome is birth weight, and the treatments are smoking and not smoking.) In this study, both treatment groups (smoking and nonsmoking) were similarly comprised of only women in their twenties. Thus, the researcher designed the study so that the woman's age would not be a confounder. A variable is a confounder if, based on its value, it prohibits some members of the sample from participating in the study. For this study, the sample consisted of only women in their twenties. So, each woman in the sample could participate since they could be placed into one of the two groups. (c)The variable education level of the mother is a possible confounder in this study. Choose the best reason why. A variable is a confounder if, based on its value, it prohibits some members of the sample from participating in the study. For this study, there may be some women in the sample who did not graduate from high school making it not possible for them to participate in the study. A variable is a confounder if its effect on the outcome cannot be distinguished from the effect on the outcome from different treatments. (In our context, the outcome is birth weight, and the treatments are smoking and not smoking.) There is already a noticeable difference between the two groups of women in terms of smoking. It's possible the women in the smoking group differ from the women in the nonsmoking group in other characteristics, such as education level, making it difficult to determine which variable is affecting birth weight. (d)Suppose the researcher is interested in reducing the effect that differences in education level might have on birth weight between the smoking and nonsmoking groups. What is a reasonable approach for the researcher to take? Choose the best answer from the choices below. The researcher could increase the sample size. Increasing the sample size helps to reduce the effect that confounding variables have on the outcome of the study. Because there are so many schools the women could have attended, it would be necessary to have a sample size much larger than 1000 pregnant women in order to have enough variety for a reasonable study. The researcher could aim to select a woman who smoked and then find a nonsmoking woman who has a higher education level than the woman who smoked. She can continue to create pairs of mothers in this way, such that each pair contains one smoker and one nonsmoker and such that the nonsmoker in the pair always has a higher education level. Then she could compare the birth weights between the smoking group (having a lower education level) and the nonsmoking group (having a higher education level). The researcher could divide the women in the smoking group into two groups: those who completed high school and those who did not. She could divide the women in the nonsmoking group into the same two educational groups. Then she could compare the birth weights between the smoking and nonsmoking groups, by education level.