You should get 0.268 V, which you've said you do get.
You have a potential divider with 4.824 V at one end and -2.904 V at the other. So that's 7.728 V across the potential divider.
Your divider outputs (7.728 x 3900) / (5600 +3900) or 3.1725 V of the 7.728 V across it.
As the bottom of the divider is at -2.904 V, that's 3.172-2.904 = 0.268 V with respect to ground.
It has been commented that you're shorting out ground but I don't understand that idea - you're not shorting anything out.
As the potential divider output voltage is positive, the reverse-biased diode has no real effect.