Different p-value at different node numbers of centrality comparison

[GangYLu]

Hi, everyone!

I used NCT package to compare two networks estimated by two dataset (the same group measured at timepoint 1 and timepoint 2, respectively). When I tried to compare the bridge strength of all nodes, the p-value of GAD6 is 0.031 (p < 0.05). But when I compared the bridge strength only for GAD6 or PHQ2 or GAD1, the difference and the p-value are the same (0.065, p = 0.821). Why is there a difference? What is the principle under this type of comparison?

I would appreciate any help you can give, thank you!

[Yjin49]

Sorry to disturb your question. I am very curious about the Settings of communities expectedinfluence in your code, which have puzzled me for a long time. My data do not have the bridgeExpectedInfluence results in any case, but only the EI results. Because my research also focuses on CESD and GAD