You make excellent points that are generally true. Even a 0.9+ range for global scores may manifest as around a one standard deviation change in some cases such as 150 to 135 or 130 to 117. I would imagine that the shifts in different indices may also cancel out each other in certain cases as well. You are absolutely right that there is no way to directly measure IQ independently of environment. I have often advocated that in my DS' case, his real life functioning lies below what IQ results suggest. I also mentioned that direct observations and other measures like achievement tests were used in reviewing the IQ scores as I would truly question IQ results that contradicted real life observations.
