from a hole in the ground when it
comes to mechanical engineering quality. So as long as the *think* they got
a great deal on their latest japmobile they are in heaven.
In the future, try not to drag out JD Powers as proof of anything important.
For reasons I won't mention here (unless he wants me to), I think Matt
knows better. Otherwise, your response is the same as mine would be.
The JD Power survey of *initial* customer satisfaction asks people who
*bought them* if they're satisfied with their cars. Well, *duh*!
Your average GM customer seems to me to be a helluva lot more tolerant
of crappy quality than some of the typical
Audi/BMW/Benz/(fill-in-high-buck-car-here) whiners we see here. Not
that that's a good or bad thing, but it *does* skew the results. A
Chevy buyer will normally be satisfied if the dealership promptly
fixes things while a significant percentage of A-B-B-? buyers expect
their cars to be perfect and tolerate no less before complaining about
it - especially to JD Power.
In addition, damn few of them have ever actually bothered to *compare*
the different makes or they'd instantly realize (as I did when I sat
down in a 600-mile brand new Cavalier rental last year) that there are
*other* things in a car that constitute 'quality' than simply getting
one dependably from one place to another. *Comfort*, for one.
*Performance* for another. A couple of good reasons I'll buy BMWs old
enough to have their own driver's licenses. ;^)
(More rust than that)
JD offers multiple surveys, only one of them is the initial quallity survey.
They also have a dependability survey and GM does rather well there as well.
I do not work for JD (nor have I ever worked with them), but as someone that
works in the data analysis industry, they are respected and their survey
methods are scientific and proper. I have noticed that whenever someone is
unhappy about the results of their "favorite" in a survey Mark Twain's
"lies" quote (or a variation thereoff) comes out.
Before someone gets too excited, I do not own a GM vehicle, the closest I
ever got to one was when I had a GM engine in an AMC Jeep. Like it or not,
GM has consistently got good results in JD and other surveys - I would argue
that this is proof enough that their build quality is competitive in the
industry and not the joke people try to make.
IIRC, when Japanese cars got the good results in the 70's and 80's - the
same attitude was relevant against their quality because of previous
results. I guess that there is always a lag between quality and percieved
Unless you can provide a scientific study that proves this, it is nothing
more than a theory. I would argue that the percieved value for purchase
price might skew the results somewhat, but scientific surveys that use
proven statistical tools always provide a margin of error and a degree of
confidence in the results. From what I have seen about the JD dependability
surveys, they are using the same standards used in the industry for non
It is a survey conducted using a proper SRS. What do you need a control
group for if you have no motive to prove a theorem (or hypothesis)?
Once you take a large enough sample you can use scientific methods to create
a confidence interval. Hypothesis testing is just one use for the
statistical methods used in surveys like these.
Anyway, how are you going to have a control group? Sell people placebo cars
that just look like a BMW but are actually a Hyundai under the skin?
Since the same formulas and survey methods are used for the creation of the
confidence intervals as what would have been used in a hypothesis testing, I
fail to see how one is more scientific than the other.
You are pointing out the differences between statistics and science.
Statistics do not *scientifically* prove or disprove anything. That is my
point. Not that what they say is or is not valid, but that it is not
Statistics use science (math) in order to predict population parameters
based on a sample with confidence levels and margins of errors (that's the
science part of it). If a survey is done using true SRS with large enough
sample sizes, we can say that we are presenting statistical data analyzed
and collected scientifically from it. That's all I mean by that.
They do not provide a e=mc^2 kind of proof, but than the world we live in is
too complicated to reduce into simple linear equations - and statistics
(when properly interpreted, but this is an entirely different issue) can
provide us with very accurate prediction of the real population parameters.
All I saying is that from my understanding, JD follows the required
procedures in order to present information that is not biased, and thus (as
the science tells us) is a good indication of the true situation.
Before we get into sematics of language - my original post was that slamming
JDPower as a mere marketing organization that provides useless information
is far from the truth as I see it.
FWIW - We know from Quatum Mechanics that even e=mc^2 breaks down in certain
situations - so there is no true scientific "silver bullet" out there.
I never said it was *totally* useless. Just *marginally* useless.
Marketing analysts simply want to predict what people *will* want and
dammit, *I* am just not that predictable! You can pretty much tell
where one of a flock of *sheep* will end up, but that doesn't apply to
real Car Guys (TM).
There is no way that J.D. Power or any marketing organization can
provide any meaningful input to an enthusiast that will tell me, for
example, whether I'll be happier or feel more at home in a 2002 Jaguar
X-Type or in a 1988 BMW 535is or in a 2003 Chevy Cavalier. Anybody
want to hazard some guesses as to the "C.R. Krieger Satisfaction
Rating" of those three? Those who know me well from my years here
will have no problems ...
(Been there; done that)
Statistical data about non-critical applications (like this) is useful to
help you make a decision. If you have a couple of cars that appeal the same
to you from a driving dynamic pov, for example - it is nice to know which
one is more likely to be reliable and provide more usage enjoyment. But here
we are getting into the "interpration" issue I mentioned before.
That's not their goal and anyone that approaches this data this way has no
idea what it means.
Having said that - there is no way that anyone can predict what you will
like more other than you - including enthusiast media, your buddies at the
local car club, people on this board or even your significant other - but we
all have practical limits to the amount of time and money we can devote to
our decisions and have to use other sources of information in order to help
us make a decision. If you take it in this grain - it is a very useful
source of information.
I find this hard to believe since the near-new rental I had seemed like it was
falling apart. Beyond that, the Cavalier is one of the worst cars in it's class
on a consistant basis, and I know someone who's 96 Impala SS nearly came apart
at the joints. Add to the cheap interiors, lifeless drive qualities and usually
sub-par perfomance and you've got a formula for disaster. The V8 powered Impala
couldn't break 7.5 seconds to 60 MPH, long gone are the days of 13 second SSs'
from starught off the factory floor. Besides, the Malibu hardly qualifies as a
sports sedan. I guess next you're gonna try to tell me that the Bonneville SSEi
and Impala/Monte carlo SS are class-leading sports sedans, right?
Motorsforum.com is a website by car enthusiasts for car enthusiasts. It is not affiliated with any of the car or spare part manufacturers or car dealers discussed here.
All logos and trade names are the property of their respective owners.