As some of you know from previous threads, I don't put a lot of faith in CR surveys because they are not a random sampling by ANY measure - their data is biased and skewed.
IMHO, only a particular type/subset of consumer subscribes to CR, and therefore even has the chance to respond (Bias #1). Furthermore, those responding may have an agenda - an axe to grind, overwhelming praise, a brand fanatic, an employee of the manufacturer, etc (Bias #2).
SO HOW CAN IT BE AN ACCURATE SURVEY AT ALL???????? They tout their customer survey base as a "feature", but it really exposes a "bug" in their system.
"Our 2005 reliability survey, the largest of its kind, reached a milestone this year—we've gathered responses on more than 1 million vehicles from Consumer Reports and ConsumerReports.org subscribers, the most we've ever received. These results underpin the most comprehensive reliability data you will find anywhere. Here, we give you a first look at our new Predicted Reliability Ratings for new cars, based on this survey, and the models that we expect to be the most and least reliable."