I work for a OEM and most of my time is spent analyzing and dissecting JD Power results. Two important things:
1. Statistically, JD Power studies are nonsense. There are routinely variations in the study that make zero sense and cannot be explained. However, we still use the studies as a benchmarking tool for one simple reason: Executives like that you can supposedly sum up the performance of the company in a single number, JDP PPH (problems per hundred). It’s really nonsense.
2. We have something called the “Porsche Effect”. Basically brand prestige leads to low JDP scores. Customers want to feel validated in their purchase, so they don’t complain to JDP. Don’t get me wrong, obviously Porsches are excellent cars, but are they 20%-30% better than BMW, Audi or Mercedes? Doubtful. 

http://jalopnik.com/how-the-hell-are-porsches-so-reliable-1709270257