M:Metrics and IDC both recently conducted iPhone surveys, and the results were somewhat different. The difference was in the methodology, and the methodology has everything to do with the final results, according to ITWire on Wednesday.

Right after M:Metrics published its findings on potential iPhone customers, IDC released its own results. While M:Metrics said that 14 percent ofthose surveyed had a strong interest in buying the iPhone, IDC claimed that only 10 percent were interested.

That’s not a huge difference as far as surveys go, but what was interesting was the difference in the experimental design.

"The most obvious difference is in the sample size," Stephen Withers wrote. "M:Metrics had 11,060 respondents, IDC just 456. The sample space was also different, with M:Metrics apparently sampling from mobile phone subscribers, while IDC looked at online mobile phone shoppers."

The upshot is that mobile phone shoppers wouldn’t be shopping for a new phone just before the iPhone is released; they probably had already decided against the iPhone.

It requires some training to design these surveys. Some informal surveys get off the hook by claiming that it wasn’t a "scientific survey." In other cases, like these two, where there is more rigor, the methodology still needs to be scrutinized and understood. Inattention to subtle factors, sample size, and the phrasing of questions can seriously skew the results.

It’s very late in the game. The buying public will vote with their dollars, and we’ll know the real world results soon enough.