Biology Reference
In-Depth Information
Next, consider the confidence interval calculated for the distance
between landmarks 1 and 3. This interval does not include 1, but unlike
the previous estimate examined, this estimate is less than 1. We inter-
pret this as indicating that the distance between landmarks 1 and 3 is
significantly smaller in the mean normal hemi-mandible relative to the
mean Ts65Dn hemi-mandible.
Testing the null hypothesis of similarity
in form using EDMA-I
One way to organize EDMA output is to write the form difference
matrix as a vector and sort the elements from smallest to largest val-
ues. This format is shown below. It is convenient to write the output in
this format when the EDMA-I test is used to determine whether or not
we can reject the null hypothesis of similarity in shape. The Bootstrap
test, described previously in this chapter, is applied and the output is
given below.
Form Difference Matrix (Sorted from Minimum to Maximum)
LANDMARK 3 with LANDMARK 1 0.79328
LANDMARK 4 with LANDMARK 1 0.88377
LANDMARK 6 with LANDMARK 5 0.98774
LANDMARK 10 with LANDMARK 5 0.99459
LANDMARK 10 with LANDMARK 9 1.00024
LANDMARK 11 with LANDMARK 5 1.00576
LANDMARK 7 with LANDMARK 5 1.01100
LANDMARK 11 with LANDMARK 3 1.01177
LANDMARK 9 with LANDMARK 5 1.01416
LANDMARK 11 with LANDMARK 4 1.01458
LANDMARK 6 with LANDMARK 4 1.01521
LANDMARK 6 with LANDMARK 3 1.01765
LANDMARK 10 with LANDMARK 3 1.01787
LANDMARK 10 with LANDMARK 4 1.01864
LANDMARK 9 with LANDMARK 7 1.02009
LANDMARK 4 with LANDMARK 3 1.02135
LANDMARK 7 with LANDMARK 4 1.02390
LANDMARK 7 with LANDMARK 3 1.02461
LANDMARK 9 with LANDMARK 3 1.02535
LANDMARK 9 with LANDMARK 4 1.02547
LANDMARK 5 with LANDMARK 3 1.02934
LANDMARK 8 with LANDMARK 5 1.03074
LANDMARK 11 with LANDMARK 10 1.03110
LANDMARK 5 with LANDMARK 4
1.03244
LANDMARK 8 with LANDMARK 4
1.03452
Search WWH ::
Custom Search