Yeah I'm not really sure I hold much weight to these satisfaction surveys - just too many variables that could potentially skew results on way or another which may not be actually genuinely reflective outside of the survey group.
Too many variables, no doubt. And then there is the margin of error. Not to mention the type of use and knowledge of the average user. It would be interesting to see what the details are in terms of satisfaction vs dissatisfaction items...
Another point is: I may enjoy my bicycle but it does not make it a car or a motorcycle; or, I may dislike my Porsche because it was simpler for me to drive my Camry on daily commutes with its automatic transmission AND it had much better mileage... Of course such real comparisons never really happen as the car owner will typically never own the other models so it's all hypothetical and theoretical.
Too many variables is right.
The other interesting statistic here would be sales numbers. Although accuracy is an issue here as well when "shipped" is often substituted for "sold" and full disclosure is often lacking...
Over time another good statistic is "switcher" numbers: iPhone to Samsung, Samsung to iPhone, iOS to android and vice versa. Considering most people keep their phones at least 2 years to get the best deal, there have not been too many purchasing cycles yet.
Interesting stuff. The war goes on.
Enjoy your great phones in the mean time!