STATS 2B03 Lecture Notes - John Tukey, Confidence Interval, Birth Weight
Document Summary
. 8828 from anova table, q from q table. 0. 32<1. 341 smoking would have the effect of decreasing the birth weight. Equivalently, we can compute tukey confidence intervals for i j and reject. H0: 1= j if and only if 0 is not in the confidence interval in our example, a tukey 95% c. i for 1- 2 is. = -1. 17 to 1. 87 since 0 is in this c. i. , we do not reject h0: 1= 2. 8. 2. 3 checking the anova assumptions the validity of the above analysis requires the following assumptions: the populations all have the same variance ( 12= 22= = k2) the samples are independent. E11 = x11 - xavg. 1 = 7. 5 7. 59 = -. 09. E21 = x21 - xavg. 1 = 6. 2 7. 59 = -1. 39. E13 = x13 - xavg. 3 = 7. 5 7. 59 = -. 09. Assumption 1 is equivalent to the residuals following a normal distribution.