Question

In the intervention group, 210 of 1267 girls had a pregnancy event (birth or induced abortion)

In the comparison group, 168 of 1567 girls had a pregnancy event

1. Using these numbers, prepare a contingency table like figure:

	Intervention	Control	Total
Pregnancy Event
No Pregnancy Event
Total

2. Calculate the relative risk for pregnancy between the intervention and comparison group.

(# participants with event in exposed group/ total participants in exposed group) Probability of event in exposed group RR = Probability of event in unexposed group (# participants with event in unexposed group / total participants in unexposed group)

3.  Based on your calculation of the relative risk, what effect did participation in the infant simulator intervention have on pregnancy events?

4. The researchers calculated a χ2 statistic for the association between exposure to the intervention and experiencing a pregnancy event, with the following result:

                                                χ2 = 20.8

                                                p=0.000044

What are the hypothesis and the null hypothesis for the χ2 test in this scenario?

5. Write a sentence that interprets the result of the χ2 test.

6. Based on these results, can we reject the null hypothesis?

Answer 1

1. The contingency table is prepared as follows-

	Intervention	Control	Total
Pregnancy event	210	168	378
No Pregnancy event	1057	1399	2456
Total	1267	1567	2834

2. Relative risk of pregnancy = (Probability of pregnancy in the intervention group) / (Probability of pregnancy in the control group) = (210/1267) / (168/1567) = 1.5457

3. Based on the calculation of relative risk, it can be concluded that the risk of the outcome that is pregnancy events increased by 50% due to participation in the infant simulator intervention.

4. The chi-square test in this scenario is used test whether there is any association between 'exposure to the intervention and 'experiencing a pregnancy event' . The Null Hypothesis is given as, H0 : there is no association between the two given events vs the alternative hypothesis is H1 : there exists association between the two variables.

5. If the calculated chi-square value is greater than the tabulated chi-square value then we reject the null hypothesis or else we fail to reject it. Here, we consider the p-value which is 0.000044. Hence, the test is highly significant at 1% level of significance since the p-value is very very less than the alpha value of 0.01.

6. Based on these results, we reject the null hypothesis at 1% level of significance and conclude on the basis of the given information that association exists between 'exposure to the intervention' and experiencing a pregnancy event'.