At least we are above the industry standard. But how did Porche come in first?<ul><li><a href="http://www.autoblog.com/2008/06/04/j-dd-power-releases-2008-initial-quality-study/">http://www.autoblog.com/2008/06/04/j-dd-power-releases-2008-initial-quality-study/</a</li></ul>

This was the "initial quality survey." I doesn't really have anything to do with long term reliability. About 3 months after you buy a new car, you get a letter from JD Powers with a long survey in it. If you choose to fill it out, you're part of the "initial quality survey." There are lots of flaws to this survey method, such as many owners not bothering to fill it out (are they reprsentative of all owners?) because it also asks too many personal questions, or people fudging answers based on their expectations of their car. If a car doesn't break unti month 4, then it might be perfect in the owner's eyes. Chevy owners might be easier on interior/paint quality than Audi or Porsche owners. People might want to "defend" their purchase of an American car, showing the world that it's every bit as good as one of them foreign cars. The list could go on.

Consumer reports data on actual reliability is a lot better, although even it can suffer from some of the same effects as above. (At least it isn't limited to 90 days. It goes unti the car is about 6 years old). CR puts the A3 at a little better than average for reliabilty.

aren't enthusiasts or car people they are after the image side of things. People that post on AudiWorld are the minority ;-)German cars are idealized by many and people pay more money for them. Then when they buy one they feel it should not break, so for some people one problem can be a big problem. Also some people don't like the expensive manitenance of these cars and usually are hit by surprise when they take the car in unless they are enthusiasts know. This also affects people's perceptions...

Good points about "initial quality" vs. long-term reliability, but I disagree about the number of respondents.

Many times the problem with surveys is not how they are conducted, but how the results are interpreted by others. As you note, JD didn't ask about long-term reliability, for example, but it was interpreted that way.

There are plenty of established statistical methods to "fix" the problem of non-respondents when designing a sample and I'm sure JD has specific validity numbers (usually expressed as plus or minus percentages or points) to help others evaluate the findings.

Also, good surveys include questions designed to catch "fudging" by comparing responses from seemingly unrelated questions.

That said, I agree that CR is somewhat better, though its questions IMHO are more flawed than JD's, given what each is trying to measure.

-dan

And, on top of all those flaws, JD Power services have to be bought. If you don't cough up the dough for their services, you can't claim an award.

Not to mention there are lots of categories for which manfacturers can buy, 19 just for different car types, sedan, SUV, etc, plus: Jaguar claimed a a #1 rating for car buying experience, Hyundai a #1 for best "non-premium" brand, Porsche for initial quality experience, and on it goes.

JD Power is simply a marketing scam.

comparison data

<ul><li><a href="http://forums.audiworld.com/a3/msgs/107045.phtml">;-)</a></li></ul>

Would a Chevy Aveo owner feel the same need to "defend" his purchase as a Mercedes owner? Could owners of some types of cars even be more or less likey to take the time to fill out and mail a survey? (Even literacy concerns come to mind here, like ability to actually understand a question, as well as having the time to fill out the survey or even put up with it, as it takes time to fill out a 4 to 6 page survey, like I received SIX of so far on my A3 over a 2 week period from different survey outfits. I filled out the first one and tossed the rest, after removing the crisp new dollar bill that was in half of them). Could owners of some brands or models be expected to have more automotive knowledge, and thereby know what is an actual fault and what isn't? Some buyers have no clue whether their car has ABS, stability control, traction control, or what, or even what the difference is. Could a Prius owner excuse other faults and be happy overall just because of good mpg? (You can see that kind of effect even on Tire Rack's tire surveys. Tires either seem to get good marks all across the board in different categories, or get bad marks, or medium marks. Rarely do you see a tire that gets realistic marks, like it sticks very well but is very noisy and is completely worthless in snow - people are biased in their answers, in my opinion, plus buyers of a performance tire might rate on a different scale than people that think all tires are the same and buy by cost alone.) What I'm trying to say in a long way is that it might be the same for car surveys. Different brands might face different challenges on the same survey.

It's pretty much impossible to get a completely valid survey on a consumer product, in my opinion. Given that, they are still worth looking at and I think they at least have some value.

Your Tire Rack example is excellent and makes the point. If you ask, "Which lasts longer, Michelin or Conti?" you will get garbage info.

If you ask, "How many miles did you drive on your Contis before replacing them," then ask Michelin owners the same, and also ask both groups questions like "Which do you agree with: a) It's smart to run tires until the cords show to get your money's worth; b) Once I sense hydroplaning at &gt; 40 mph it's time for new ones ..." you get the idea (grin).

-dan