If we ignore variations in mass density as well as variations in the radial distance from the center of gravity to the surface, (mountains, equatorial bulge etc.), and external forces such as the gravity of the Moon and Sun, then:
From anywhere on or above the Earth�s surface, gravity can be considered a point source at its center, equivalent to the entire mass of the Earth and the sphere of the Earth�s surface is the radial distance of maximum gravitational acceleration . . . However . . .
Beneath the surface of a sphere, gravity plays a different kind of ball game.
As Peter Pedant pointed out we do not consider the mass in overlying spheres (above the radius from which we are determining gravitational acceleration) because the mass of the spheres above this radius �attract� from all directions equally and effectively cancel each other out.
A spheres volume increases with the cube of the radius. Therefore, comparing two solid spheres of equal density, a sphere of twice the size would have eight times the mass. Remember that gravity varies inversely to the radius squared, consequently, doubling the radius doubles the gravitational acceleration. The result is that below the surface gravitational acceleration should diminish linearly as we approach the Earth�s center.
But the Earth has another curve ball to throw our way.
Beneath the Earth�s surface mass density increases as we approach the core. The Earth�s core has a mass density five to six times that of surface material. The net effect of this is that gravitational acceleration actually increases below the Earth�s surface reaching 10.7 m/s� at the core-mantle boundary before declining to zero at its center. Go figure!