When you apply a force to an aircraft it'll act initially about the CG. If the force vector doesn't go through the CG, then you have a moment about the CG, this will cause rotation of the aircraft about some point near to the CG - aerodynamic effects will generally stop it being spot on the CG.
So, at a first approximation, an aircraft with low thrustline and high CG will pitch up with power, and an aircraft with high thrustline and low CG will pitch down with power.
However, this isn't absolutely universal, because the horizontal stabiliser will usually to some extent be in the propwash. Hence increased power not only increases thrust, but increases local airspeed over some or all of the horizontal stabiliser (or for that matter over the centre section of the wings). So sometimes you'll see unexpected effects for that reason.
But, as a simplistic (i.e. JAR exam) explanation, I'd say CG will probably suffice as the answer.
G
N.B. Look at the engine mounting angle too. A low engine can still have a high thrustline and vice-versa - it's where the thrust vector passes relative to the vertical CG that counts, not the actual engine position.