back to index

Correlation/prediction results for subject measure 98 (SSAGA_ChildhoodConduct)

819 subjects had a valid SSAGA_ChildhoodConduct measure.

Multivariate prediction (GLM-based, automatic feature selection, leave-one-family-out prediction)
Original data space: r=0.10 CoD=-0.11     Deconfounded space: r=-0.05 CoD=-0.16
Scatterplot shows predicted-SSAGA_ChildhoodConduct vs measured-SSAGA_ChildhoodConduct (in original and deconfounded data space).

Univariate regression (regressing each netmat element independently against SSAGA_ChildhoodConduct, correcting for multiple comparisons across elements, using PALM permutation testing, taking into account family structure).
Number of significantly correlated edges at p<0.05 (two-tailed, FWE corrected) = 0 (minimum corrected p = 0.4110)
Number of significantly correlated edges at p<0.05 (two-tailed, uncorrected) = 277 (248 expected by chance)
Image shows edges (node-pairs) whose connection most strongly correlates with SSAGA_ChildhoodConduct (in decreasing order), with t-statistic listed at the top of each node-pair.