Statistical Methods for Comparison of Forms - An Invariant Approach to Statistical Analysis of Shapes - page 178

Table 4.3. Distribution of Bootstrapped T Statistics:
T
1.028 |*****
1.057 |*****************************************
1.087 |***************************************************************
1.116 |*************************************************
1.146 |***********************
1.175 |************
1.205 |***
1.234 |****
1.264 |
1.293 |
1.322 |
1.352 |
1.381 |
1.411 |
1.440 |
1.470 |
1.499 |
1.528 |
1.558 |
1.587 | T obs
Testing the null hypothesis of similarity in form using EDMA-II
EDMA-II was developed for testing the difference in shapes of two
populations where the variance-covariance matrices are not equal. As
described in Chapter 1 , the Ts65Dn mouse is a model for Down syn-
drome. Previous quantitative studies of the craniofacial phenotype in
children with Down syndrome have documented increased variability
in this group as compared to unaffected children (Kisling 1966;
Thelander and Pryor, 1966; Frostad, Cleall et al., 1971; Cronk and
Reed, 1981). Because Ts65Dn mice are at dosage imbalance for many
of the same genes duplicated in Down syndrome (Richtsmeier, Baxter
et al., 2000), it may be necessary to adopt a working hypothesis that
the Ts65Dn mouse will also show increased variability as compared to
euploid littermates. If evidence for the working hypothesis is found,
application of EDMA-I to these samples may be inappropriate and
EDMA-II testing should be implemented.