Rick, Thasp hasn't checked back in and didn't give full details of how he measured and calculated, but yes, what you're suggesting is correct. The applicable form of Ohm's Law is that power equals the square root of (Vsquared/resistance), so the standard measure is 2.83V input into 8ohms, which is equivalent to 1 watt. If 2.83V was instead input into 4ohms, the power would be 2 watts rather than 1 watt. To actually get a 4ohm number for 1 watt the input would be 2.0V rather than 2.83V(or the result would be adjusted down by 3dB to compensate for the lower impedance).
Another point is that the basic sensitivity measure is supposed to be taken under anechoic conditions, and as we discussed with Gena quite a while ago, measured in room at normal listening distances the reverberent sound field can be as strong or stronger than the direct field(so the 6dB loss per doubling of distance isn't the complete picture and less power is required than that measure would imply).
