Statistical Methods for Comparison of Forms - An Invariant Approach to Statistical Analysis of Shapes - page 185

Table 4.5. Confidence Interval for Scale Difference with α 0.100
Distribution of Bootstrapped Scale Differences
-1.961 |**
0.146 |**
0.160 |**
0.174 |**
0.187 |***
0.201 |*****
0.214 |**********
0.228 |*************
0.242 |***************
0.255 |D********
0.269 |**************
0.283 |*********
0.296 |*****
0.310 |*****
0.323 |**
0.337 |**
0.351 |*
0.364 |
0.378 |*
0.391 |
0.405 |*
Upper bound: 0.32669 Lower bound: 0.15548

This confidence interval ( Table 4.5 ) does not contains zero and
therefore suggests that the two populations do not differ significantly
in scale. Remember, however, that there is no single value that repre-
sents “size” and that this result may change depending upon the
chosen scaling factor. With no evidence for a difference in size, confi-
dence intervals for the estimated shape difference matrix can be
examined.

Confidence Intervals for Shape Differences Based on 100
Resamples with α 0.10

Lower Limit Estimate Higher Limit

LND 1 LND 2 0.003 0.047 0.076

LND 1 LND 3 -0.159 -0.130 -0.112

LND 2 LND 3 0.055 0.082 0.107

LND 1 LND 4 -0.165 -0.142 -0.123

LND 2 LND 4 0.042 0.070 0.092

LND 3 LND 4 -0.027 -0.010 0.004