You understand it right. What you discovered is that manufacturers don't always report sensitivity accurately, or they use different methodologies.
The ohm rating of a speaker is listed as a single number, but in reality sometimes fluctuates dramatically in actual usage. A 8 ohm speaker can dip below 4 ohms at certain frequencies. Every speaker varies differently and the actual ohm of the speaker can significantly impact the sensitivity measurement.
For example, the standard methodology to measure sensitivity is a 1watt(or 2.83v) signal measured at 1 meter. Using this method, a speaker that dips to 4 ohms at the frequencies measured would have a 3 dB higher sensitivity than a speaker that was 8 ohms at that frequency.
Also, sensitivities are reported by some manufacturers as in-room and others anechoic. The in-room sensitivity will be higher due to room gain, and that could be 3+ dB. Many don't tell you which method they use.
So using the industry "standard" to measure sensitivity can lead to very different results.
Sent from my Galaxy Nexus using Tapatalk 2
JTR Noesis 212HT x 3 (LCR) powered by Lab Gruppen 10000Q amp
CHT SHO-10 x 4 (sides and rear) powered by Denon 4311
JTR S2 x 2
CHT 18.1 x2
Oppo BDP103D bluray player/Sonos/PS3