Table 2.

Wilcoxon rank sum tests of the hold-out samples.

	TIV women classified as female versus classified as male	TIV men classified as male versus classified as female
AM hold-out sample
AM model	T = 12,722, z = − 2.3885, p = 0.0169, η2 = 0.0354	T = 12,829, z = 3.3879, p < 0.001, η2 = 0.0713
AM+cr model	T = 7514, z = 3.2204, p = 0.0013, η2 = 0.0644	T = 8858, z = − 2.6727, p = 0.0075, η2 = 0.0444
ATM model	T = 11,004, z = − 0.4390, p = 0.6606, η2 = 0.0012	T = 11,507, z = 0.0236, p = 0.9812, η2 < 0.001
AM+cr model	T = 11,236, z = 0.2778, p = 0.7812, η2 < 0.001	T = 11,284, z = 0.5097, p = 0.6103, η2 = 0.0016
ATM hold-out sample
AM model	T = 9908, z = − 4.7156, p < 0.001, η2 = 0.1381	T = 11,325, z = 6.2257, p < 0.001, η2 = 0.2407
AM+cr model	T = 8425, z = 0.8513, p = 0.3946, η2 = 0.0045	T = 10,341, z = − 2.3190, p = 0.0204, η2 = 0.0334
ATM model	T = 12,284, z = 1.3806, p = 0.1674, η2 = 0.0118	T = 12,239, z = 1.0910, p = 0.2753, η2 = 0.0074
AM+cr model	T = 12,403, z = 1.6918, p = 0.0907, η2 = 0.0178	T = 12,130, z = 0.8780, p = 0.3800, η2 = 0.0048

Comparison of individuals classified as female versus male (Wilcoxon rank sum tests) for the AM and ATM sample.