network analysis - Why do normalized centrality values the exceed 0-1 range?

MBD

Hi JASP users,

I ran a network analysis on 4 nodes (continuous variables). I chose EBICglasso estimator. Under the Analysis option, I chose 'Cor', and the 'Normalized' option under centrality measures. The centrality plot however showed negative values such as -0.912 and also positive values such as 1.301 for other nodes.

I think normalized values are supposed to be between 0 and 1 (?)

I'd appreciate it if anyone could explain this to me and how to interpret negative values of centrality measures if the negative values that I got are not a result of an error.

Thank you,

MB

EJ

I've asked the team!

EJ

vandenman

Hi MB,

The values are normalized by subtracting the mean and dividing by the standard deviation (i.e., they are z-scores). So there is no need for them to lie between 0 and 1.

Hope that helps,

Don