Your friends who say yolks don't get salmonella infections are dangerously wrong.
It seems that Salmonella infections tend to start on either in the albumen (the egg white) or the layer between the yolk and the albumen, called the Vitelline Membrane:
In egg contents the most important sites of contamination are either the outside of the vitelline membrane or the albumen surrounding it. In fresh eggs, only few salmonellas are present and as albumen is an iron-restricted environment, growth will only occur once storage-related changes to vitelline membrane permeability, which allow salmonellas to invade yolk contents, have taken place. When this happens high populations are achieved in both yolk contents and albumen.
Do you count that layer as the yolk? If not, strictly, there are more infections in the albumen than the yolk. If so, you might get a practical measurement more like this:
[...] incidence of yolk contamination (approximately 2.5% overall) was significantly greater than the incidence of albumen contamination (approximately 0.5% overall).
But that doesn't take a more important issue into account.
[...] egg yolk and albumen differ substantially in their ability to support bacterial growth. [...]
Thus, it may be that more eggs are infected in the albumen + vitelline membrane than the yolk proper, but there may be more risk to health in the cases where the eggs are left unrefrigerated and the yolk infections take place. In the case of eggs (under)cooked in their shells, it is more likely the yolk will not reach the necessary temperature to kill the Salmonella cells.