One-way ANOVA: No, Short, Long Source DE SS Factor 2 252.00 Error 15 98.00 Total 17 350.00 S = 2.556 R-Sq MS 126.00 6.53 = 72.00% F P 19.29 0.000 R-Sq (adj) = 68.27% Since the intelligence of birds differ for at least two types of migrants, we have to conduct the post hac tests to identify which pairs of migrants differ significantly. Fisher's multiple comparisons procedure is one of such methods. In this method we have to obtain confidence intervals for all pairwise differences between level means using Fisher's LSD procedure using the following formula, 1 1 {[(x - X}) #a2(x+1)8 yn, nj where n be the total number of observations and k be the total number of treatments. If the interval so obtained contains the value '0' in it, then there is no significant difference between the means of ith & jth treatments, otherwise they are treated as differ significantly. Level N Mean StDev No 6 15.000 3.098 Short 69.000 2.608 Long 6 6.000 1.789 Pooled StDev 2.556 No subtracted from: Individual 95% CIs For Mean Based on Pooled StDev Short subtracted from: () Fisher 95 Individual Confidence Intervals All Pairwise Comparisons Simultaneous confidence level = 88.31% -------) 7.0 Lower Center Upper Long -6.145 -3.000 0.145 Lower Center Upper --- Short -9.145 -6.000 -2.855 Long -12.145 -9.000 -5.855 (---) 10.5 -10.0 -10.0 (----------) -5.0 () -5.0 14.0 (----------) 0.0 0.0 17.5 5.0 5.0 From the above MINITAB output it is clear that the mean intelligence of non- migrating birds differs significantly that of short distance migrating and long-distance migrating birds. There is no significant difference between the mean intelligences of short distance migrating and long-distance migrating birds. -