Coaster is sort of on the right track, I think. AC power is a completely different animal than DC. For DC circuits, then P = VI, where P is the power dissipated, V is the voltage drop, and I is the current running through the element. However, in AC circuits, then the input voltage is a sinusoid, usually represented by a cosine function. If you have purely resistive loads, then the current stays in phase with the voltage function, and the P that you're looking for, the average power, is equal to VrmsIrms. (Vrms = Vmax/sqrt(2)). Inductive or capacitive loads can cause the voltage and current to get out of phase. The effect of this is a decrease in average power, since the current and voltage no longer line up at their maxima. There is another kind of power, called reactive power. Basically, this represents all of the losses in the circuit. All of this is usually represented using phasor or polar notation, so if you do it that way, it works out nicely and you can figure out your power factor and such from that information. In short, it goes like this: plot average power on the real axis of the complex plane, and reactive power on the imaginary axis. The power that they are quoting, in VA, is the length of the phasor (or vector) from the origin to the point you plotted. The distance on the real axis is the average power, in watts, and the distance on the imaginary axis is the reactive power, in VAR's. I believe the power factor is the arctan of the reactive power / average power. I could be wrong, but I'm pretty sure that's the gist of it.
If there's any EE's out there who want to correct me, go ahead. I'm sure I missed something somewhere.
