Statistical Methods for Comparison of Forms - An Invariant Approach to Statistical Analysis of Shapes - page 190

LND 7 LND 5 1.01102
LND 11 LND 3 1.01178
LND 9 LND 5 1.01432
LND 11 LND 4 1.01459
LND 6 LND 4 1.01522
LND 4 LND 3 1.01674
LND 6 LND 3 1.01710
LND 10 LND 3 1.01778
LND 10 LND 4 1.01859
LND 9 LND 7 1.02059
LND 7 LND 4 1.02395
LND 7 LND 3 1.02451
LND 9 LND 3 1.02529
LND 9 LND 4 1.02543
LND 5 LND 3 1.02920
LND 8 LND 5 1.03070
LND 11 LND 10 1.03194
LND 5 LND 4 1.03239
LND 8 LND 4 1.03454
LND 8 LND 3 1.03612
LND 9 LND 8 1.03828
LND 10 LND 7 1.04098
LND 11 LND 7 1.04132
LND 6 LND 2 1.04358
LND 11 LND 9 1.05073
LND 7 LND 2 1.05127
LND 10 LND 6 1.05230
LND 11 LND 6 1.05392
LND 8 LND 7 1.05422
LND 7 LND 6 1.05524
LND 11 LND 8 1.06053
LND 8 LND 2 1.06729
LND 8 LND 6 1.06781
LND 10 LND 2 1.06811
LND 9 LND 2 1.06964
LND 10 LND 8 1.07670
LND 11 LND 2 1.08413
LND 5 LND 2 1.10241
LND 9 LND 6 1.10527
LND 3 LND 2 1.14628
LND 4 LND 2 1.15038
T statistic [=max/min]: 1.16709
Notice that the value of the observed T statistic, T obs , has been reduced
from 1.587 (when all landmarks were included in the analysis) to
1.167. This indicates that landmark 1 was important to the overall
form difference. The test of the null hypothesis using the reduced land-
mark set indicates the same finding.
p -value = 0.06