Negative betweenness centrality values
Hi all!
I'm fairly new to the the use of network analysis and I've used Gephi before but just recently started using JASP in order to avoid prepping data.
I might be clueless (probably I am) but I've run into a problem I'm not sure how to interpret, and as far as I've seen, it's something that's happened even in the tutorial for NA available here.
Basically, as far as I know, betweenness centrality should take up a value between 0 and 1, but both in the tutorial example and in my data, its numbers take up negative values (in my data up to -1.14, picture) and I'm not sure how to interpet it.
Usually if some variable has low betweenness I know it's not central in any way and that node isn't used often in terms of traversing the network via shorter paths. So a value of 0 would mean that specific node is never used as a point of travel.
How to interpret the fact that a specific node has a betweenness value of -1.14? Not only it's never used in a shortest paths walk, but it...demotivates info/actors/stuff from using it?
The only thing that jumps to mind is the fact that I'm using a Cov correlation method with normalized centrality measures but I'm still lost in interpreting the results, and the same goes for closeness, strength and expected influence measures.
Any help would be much appreciated, be it via comments or links to papers that I should read.
Thanks in advance!
Comments
Hi! The centrality indices are shown as z-scores by default. You can select "raw" with "centrality measures" to get the raw values!
Best, Sacha
Ended up doing exactly that, thanks!