Pew runs a solid operation. Of course they suffer from all the problems that phone surveys suffer from, but they work hard and they do good work.
Try reconciling their results with this:
I'm not familiar with In-Stat, altho I will note they had a substantially smaller sample size.
I may re-edit this if I find out more about In-Stat.
ETA: I do not understand this. I'm having two separate and probably unrelated definitional issues.
(1) What the hell is "end-user research"? It appears to be some form of market research, but I don't understand how they are picking their sample. It is not population wide, which Pew's is. It seems that InStat's selection of a universe is oversampling for tablets. This is possibly driven by the kind of insight they have into the tablet supply chain vs. the kind of insight they (don't) have into the ereader supply chain. However,
(2) I'm not sure how people are categorizing the Nook Color. If the Nook Color is selling a lot (I'm having trouble answering this question), I could imagine a scenario in which Pew's number might be skewed because some of the respondents bought Nook Colors and think it is one or the other or possibly both. The question does not appear to help answer which it is.
I think what is happening is that InStat is looking at the massive volume of cheap tablets being dumped on a largely uninterested market, leading to this prediction: "As a result, In-Stat (www.in-stat.com) is forecasting that tablet shipments will outpace e-reader shipments by the end of this year." If I were a business, I would NOT find this Helpful. I certainly would not conclude, "“Of the two, the tablet market is the stronger and more sustainable opportunity,” says Stephanie Ethier, Senior Analyst."
Especially not in the face of the Pew numbers, which show slower growth and fewer buyers of tablets vs. ereaders.