Dear Meragiaw,

Could you please clarify the problem?

Different scholars may classify the same dendrogram in different number of communities. From this point of view, I am asking just if there is the way/certain standard of statistical methods by which everyone can have the same number of groups in the same hierarchical cluster of dendrogram.

In other words, if you have four community types. How can you test their differences from each other?

https://cran.r-project.org/web/packages/ecodist/ecodist.pdf

Thank you Kflay for your suggestion. I see but even the level of dissimilarity is very subjective and you can level it whatever you want.