13C InClass

Recall the maternal smoking study presented in Preview Assignment 13.C.

Researchers wanted to study the difference in birth weight of babies born to mothers  who smoked during pregnancy (smoke_now = yes) and mothers who did not smoke  during pregnancy.

Question 1

1) Discuss the differences that you saw in the visualization between the birth weights of babies born to women who smoked during pregnancy compared to mothers who did  not smoke during pregnancy.

Question 2

2) How can we make an inference about the difference when the population refers to  all pregnant women?

In the previous in-class activity, you reviewed and applied the steps to perform a one sample t-test (a hypothesis test about a population mean). We can apply the same  steps to analyze and test a hypothesis about the difference in means for two  independent populations. A hypothesis test for comparing two population means is often  referred to as a two-sample t-test.

In the preview assignment, you verified that the assumptions for a two-sample t-test  were met for the maternal smoking study. We know that the two samples were  independent, representative of the populations of interest, and large.

Question 3

3) Group 1 consists of mothers who smoked during pregnancy, and Group 2 consists  of mothers who did not smoke during pregnancy.

a) What is the parameter of interest?

b) Write the null and alternative hypotheses, with proper notations for each.

The test statistic to compare two population means is calculated using the following  formula:

[latex]t=\frac{estimate\;of\;parameter-null\;hypothesis\;value}{standard\;error}=\frac{(\bar{x}_{1}-\bar{x}_{2})-(\mu_{1}-\mu_{2})}{\sqrt{\frac{s^{2}_{1}}{n_{1}}+\frac{s^{2}_{2}}{n_{2}}}}[/latex]

Question 4

4) Use the DCMP Compare Two Population Means tool at https://dcmathpathways.shinyapps.io/2sample_mean/ to complete the test. Select  “Summary Statistics” and enter the values from the following table:

Group 1:

smoke_now = Yes

Mothers who

smoked during

pregnancy

Group 2:

smoke_now = No

Mothers who did not  smoke during

pregnancy

Sample Mean [latex]\bar{x}_{1}[/latex] = 114 [latex]\bar{x}_{2}[/latex] = 123
Sample Standard  Deviation [latex]s_{1}[/latex] = 18.2 [latex]s_{2}[/latex] = 17.3
Sample Size [latex]n_{1}[/latex] = 480 [latex]n_{2}[/latex] = 733

a) What is the value of the test statistic?

b) What is the P-value?

c) Given a 5% significance level, what do you conclude? Write your conclusion in context and support your answer.

d) Is this conclusion consistent with what you concluded using only the  visualizations in Question 1?

Question 5

5) We often learn more from constructing confidence intervals than from the hypothesis  test because it shows a range of plausible values for the difference between the  population means.

a) Use the same DCMP tool to calculate and interpret the 95% confidence  interval to estimate the difference in the population means. Round your  answer to the nearest hundredth.

b) Interpret the result. Include the relevance of 0 in your description.

Question 6

6) An original article published in 1972 in the Journal of Epidemiology concluded that  there was “a positive association between maternal cigarette smoking and reduced  infant birth weight.”

a) Why did the authors choose the word “association” rather than “causation?”

b) Read the following excerpt from the same article. Discuss how this statement  might impact your conclusion.

text reading "Bias in selection. Selection of mothers by smoking habit may result in a study population consisting of two groups that are dissimilar in other respects. The basic problem is that some well-known risk factors affecting mortality are also independently related to the frequency of smoking."