Just to be clear on something... If we assume that all manufacturers were 100% honest with their sensetivity ratings, and that each driver in question is in a optimal box: Does that mean that a 12" speaker rated at 100db avg sensetivity 1w/1m is louder than a 15" speaker rated at 97db avg sensetivity 1w/1m ? For more info I am refering specifically to a 12" Eminence Delta vs a 15" Carvin speaker.
Pretty much. But you should also try to find out *at which frequency or frequencies* efficiency was measured. What is most important is whether the speaker is more efficient at the frequencies that matter to you. For instance, a subwoofer might be horribly inefficient at 1 kHz but much more efficient at 50 Hz, which is cool because that's where you want it to be efficient.
JBL is the only manufacturer I am aware of that publishes the actual measurement specs. They warm the driver up for X hours, then sweep the signal through a broad range of frequencies. Voice coil resistance increases with temperature and brings about a phenominon of power compression. The warm-up takes this into consideration. Link to JBL 2245 Datasheet on My Site The link above is an HTML page for the JBL 2245 subwoofer. Read the sensitivity part for how they take the measurement. The 2245 is awesome as a subwoofer in every respect, except for the huge (10+) cubic foot cabinet it requires.