STAT 509 2016 Fall HW6-2 Solution
Instructor: Shiwen Shen
                                 Collection Day: October 5
1. In a random sample of 85 automobile engine crankshaft bearings, 7 have a surface finish rough-
   ness that exceeds the specifications. Does this data present sufficient evidence that the pro-
   portion of crankshaft bearings, say p, exhibiting excess surface roughness is greater than 0.06?
   We will address this using a hypothesis test.
   (a) State the null and alternative hypotheses.
                                                 H0 : p = 0.06
                                                 Ha : p > 0.06
   (b) Calculate the appropriate test statistic.
                                                 7
       We know the sample proportion is p̂ = 85    = 0.082, so the test statistic is
                                        p̂ − p0           0.082 − 0.06
                                  z0 = q                = q            = 0.85
                                           p0 (1−p0 )        0.06×0.94
                                                n                85
   (c) What is the p-value of the test?
       Given Ha : p > 0.06, we have
             p-value = P (Z > z0 ) = P (Z > 0.854) = 1 − P (Z < 0.85) = 1 − 0.8023 = 0.1977
   (d) What is your conclusion based on the p-value if we use α = 0.1?
       Because p-value= 0.1977 > 0.1 = α, we fail to reject the null hypothesis H0 . Therefore,
       with 90% confidence level, we fail to conclude that the proportion of crankshaft bearings
       exhibiting excess surface roughness is greater than 0.06.
   (e) Calculate a 90% confidence interval for the population proportion. Interpret the confidence
       interval using the context of the question.
                                                  r
                                                    p̂(1 − p̂)
                                 C.I. = p̂ ± zα/2
                                                      rn
                                                         0.082(1 − 0.082)
                                      = 0.082 ± 1.64
                                                                85
                                      = 0.082 ± 0.058
                                       = (0.024, 0.14)
    (f) Compare the results in (d) and (e), are they similar?
        0.06 is included in the 90% confidence interval (0.024, 0.14). It is consistent to the conclu-
        sion that we fail to reject H0 : p = 0.06.