Supplementary Exercise 6.87 of IPS7e ------------------------------------ Mean nicotine content in cigarettes. The advertised value is 1.4 mg. A study was carried out to determine whether the mean content was higher. The study wants to test the null hypothesis H0: mu = 1.4mg against the alternative hypothesis Ha: mu > 1.4mg where mu is the cigarette population mean nicotine content. (a) A z-test statistic was computed: z_obs=1.75. This value is significant at the 5% level, because with a one-sided Ha we compute the P-value as P = P(Z>1.75) = 0.040. (For example, Table B of PSLS gives P(Z<1.75)=0.9599 and P(Z<-1.75)=0.0401.) Therefore, the P-value is indeed less than 0.05. Alternatively, we could note that the 5% cut-point (sometimes called the critical value) for the test with this one-sided alternative is the 95% percentile of N(0,1) because the 95% percentile has 95% to the left and 5% to the right. We can use the normal distribution table (with probability 0.95) or the table of critical values/percentiles (with upper tail probability 0.05) to determine the 95% percentile as 1.645. Because z_obs > 1.645, the test is significant at the 5% significance level. (b) We already computed the P-value, and it was greater than 0.01. Therefore, the test is not significant at the 1% level. Alternatively, the 1% cut-point for the test is the 99% percentile of N(0,1) = 2.326, and z_obs < 2.326. --- Minitab commands (to compute P-value and cut-offs) and output: MTB > CDF 1.75; SUBC> Normal 0.0 1.0. Cumulative Distribution Function Normal with mean = 0 and standard deviation = 1 x P( X <= x ) 1.75 0.959941 MTB > CDF -1.75; SUBC> Normal 0.0 1.0. Cumulative Distribution Function Normal with mean = 0 and standard deviation = 1 x P( X <= x ) -1.75 0.0400592 MTB > InvCDF .95; SUBC> Normal 0.0 1.0. Inverse Cumulative Distribution Function Normal with mean = 0 and standard deviation = 1 P( X <= x ) x 0.95 1.64485 MTB > InvCDF .99; SUBC> Normal 0.0 1.0. Inverse Cumulative Distribution Function Normal with mean = 0 and standard deviation = 1 P( X <= x ) x 0.99 2.32635