Question: The Relation Between Health Insurance Coverage and Clinical Outcomes Among Women with Breast Cancer

In their paper of the title above Ayanian et al. examined the potential relationship between health insurance status and clinical outcomes of women with breast cancer. A question the researchers had was where women who do not have health insurance coverage are less likely to be screened for breast cancer, and whether their disease is more advanced at the time of diagnosis. These data look at the insurance status (insured vs. uninsured) of the patient and the grade of cancer (1 = least advanced, …, 5 = most advanced) at the time of initial diagnosis. Datafile: Insure.JMP

a) Could a two-sample t-test be used to analyze these data? Why or why not? (2 pts.)

b) Test the null hypothesis that the stage of cancer is same for both insurance classes against the alternative that reflects that uninsured patients generally have more advanced cancer grade at the time of diagnosis. Summarize of your findings. (4 pts.)