The 2008 GSS asked about the number of hours you watch TV per day. An analysis that evaluates this by race shows the results (note the codes: 1 = Black and 2 = White):

T-Test of difference = 0 (vs not =):
T-Value = 5.96 P-Value = 0.000
a. Do you believe that TV watching has a normal distribution for each race? Why or why not? What effect does this have on inference comparing population means?
b. Explain how to interpret the reported confidence interval. Can you conclude that one population mean is higher? If so, which one? Explain.
c. On what assumptions is this inference based?

Transcribed Image Text:

Sample N en 1 2 Differen ce = mu (1) -mu (2) Estimate for difference: 1.620 95% CI for difference: (1.085, 2.155) StDev SE Mean 188 4.38 3.58 0.26 1014 2.76 2.39 0.075