Both the scion and the rootstock have an effect on the overall characteristics of the tree. Also, a rootstock that dwarfs one specific variety may not necessarily dwarf another variety.
If you are using non clonal rootstocks, the effect can be completely unpredictable, especially if using Monoembryonic seedlings because of the randomness of the parent pollen.
All this aside, I usually focus on the scion because it controls much of the shape of the top of the tree including branching structure.
For example, if I used Kent as a rootstock and grafted Lemon Zest onto it, ignoring the fact that the scion could significantly influence vigor, the resulting tree will likely be tall and upright because of the longer internode distance associated with this variety.
If I took that same Kent rootstock and grafted Carrie onto it, the resulting tree will likely have a more compact, dense canopy because of the attributes associated with the Carrie variety.
The rootstock will help influence how fast each of the trees described above will reach a specific height/size.
I had a thread on mango rootstocks but it’s so old that most of the links are broken.
Here’s a link that hopefully works
https://www.mango.org/wp-content/uploads/2017/10/Mango_rootstock_Final_Report_Eng.pdfSimon