True, they don't. But they also might not be very accurate if the sample is self-selected (like the Lifestyles one in Cancun at 5.88") or any other kind of selection bias (anytime the sample population knows they are going to be measured simply for that purpose only, which tends to exclude those who are smaller). Most of the studies that minimize this bias seem to give averages closer to 5.25, rather than 5.75...
Lifestyles was definitely a little higher due to selection bias, but the main reason people keep quoting that one is because it's one of the few that shows the full distribution of their data publicly.
The 5"-5.25" is almost always for non-bonepressed. I think there was one study in the low 5's for a pharmacologically-induced erection, but in that case the sample bias went the other way - those measured suffered from ED, and that generally takes the length down. That length can be regained if the ED is cured or the penis is actively exercised.
For longer samples, about 6.2" is as long as I've seen it bone-pressed, and that was in Wessells (self-selected.)
The British and Brazilian studies both bone-press and both get around 14.5cm, (n=609 and n=160 respectively.) Italy was n=3300, and they're 12.5cm NBP, so again pretty much 14.5cm BP. The Wessells study was only n=80 and self-selected, so to give 80 self-selected guys the same weight as 4069 random subjects is just silly.
