For the raw and expected cell counts it is clear that

This preview shows page 11 - 20 out of 33 pages.

that swing states and political office do not have association. For the raw and expected cell counts, it is clear that raw value is nearly the same as the expected value, which also means that these two variables do not have association with each other.
Image of page 11

Subscribe to view the full document.

Image of page 12
Problem 1g. Construct a contingency table with the political office and contribution group variables, showing the chi square contribution as well as the raw and expected cell counts. Describe the association test results which are appropriate for the table. proc freq data = politics ; tables can_off * contributiongroup / chisq expected nopercent norow nocol ; run ; I also used Chi-Square for this question to see whether political office and contribution group have association. In the table below, it is clear that the probability is less than 0.05, which means that there is strong relationship between these two variables: political office and contribution group. Seeing the raw and expected value, they have much difference, which also means that these two variables have strong association. However, the null hypothesis is that there is not association between political office and contribution group, which is not proved by the test result. Therefore, we should reject the null hypothesis and the final conclusion is that there is strong association between political office and contribution group.
Image of page 13

Subscribe to view the full document.

Image of page 14