This is a two part question. I need it done asap. Thank you.
This week we will start getting into the more technical material. Among all the concepts discussed in the Ch 1 Appendix, the “Difference in Means” test is the most important, as you will be carrying out a diff-in-means test in your term papers.
Read pages 21-23 and also pp. 43-46 (starting with the section “Pairing off”, which describes a diff-in-means test.) Then answer the questions below, and upload your answers here by the deadline.
1. Consider a test of the null hypothesis that there is no difference in Average Family Income among individuals with any insurance plan, and individuals with the catastrophic plan. Examine the statistics reported in Table 1.3. What is the absolute value of the test statistic for this test?
Hint: Look at the formula for test-statistic reported in the middle of the page on page 45 of Mastering Metrics (this is described as the “t-statistic for a difference in means”, i.e. we are not asking about the numbered equation that also appears on that page. As it is written there, “When the null hypothesis is one of equal means…”, the formula for the test statistic is: t(0) = (diff in means – 0) / (standard error). From Table 1.3, we know that (diff in means)=-654. Plug this in the numerator of the right-hand side, and find the corresponding standard error from the table, and plug this in the denominator of the right-hand side. Then divide the top by the bottom to find the test statistic.
2. Considering your answer to Question 1 above, do you reject the null hypothesis of equal family income? Is the observed difference in family income between subjects in the control and treatment group statistically significant?