Bias Against Domestic Cars

On Nov 26, 9:52*am, Vic Smith > wrote:

Hi Vic!

> Not getting into statistical analysis methods, I'll note the CR has
> NEVER given sampling sizes on a model basis. *Only total surveys.
> I am perfectly willing to be corrected on that.

You won't be.

> And I won't make any claims about those numbers beyond the fact that
> they could reveal a built in bias that is common in self-selected
> surveys. *Won't even say there is a useful element to the bias in the
> end. *Because I happen to think CR got it pretty right, and the bias
> is a side issue in the end. *

> --Vic


While they provide nothing for statistical information on specific
models, they probably do have enough data to see basic trends. You
can find some interesting things if you dig into the charts and
numbers they bury in the print. For example, over the 7 years that is
a typical length of new car ownership, the average Ford has roughly 5
issues a CR reader would report in a survey, about the same as Nissan
or Hyundai. The average Honda/Toyota owner would have roughly 3
problems reported by a CR subscriber. (I'll even give a margin of
error of 1 problem, as the chart was small and integrating in my head
is subject to mistakes these days). Granted, these could be minor or
major problems, covered under warrantly or very expensive. There's no
way to know from CR. Also, you still have the issue of self-sampling
and and a non-scientific survey method. See the Chicago Tribune's
"Dewey Defeats Truman" case study on why not to use data obtained in
this manner...

On Thu, 26 Nov 2009 11:24:51 -0800, SMS >
wrote:

>Ashton Crusher wrote:
>
>> That, in large part, goes to the heart of the CU bias. Most of their
>> respondents were already of the mind set that domestic was crap and
>> imports were perfect. And they repeat that when they fill in their
>> survey forms.
>
>You don't understand how the survey works. They aren't asking for a
>subscriber's opinion of one product versus another, they're asking a
>series of very specific questions that ask what, if any, problems an
>owner has had with their vehicle.
>

It's ALL opinion since no one has to send in any documentation to
support their claims. If someone wants to ignore the problems they
can just answer "no problems". You don't seem to understand human
nature. For this kind of survey to have any real validity beyond
being a popularity contest it would have to be based on DOCUMENTED
work orders for repairs of all these vehicles, say from the records of
a leasing company or taxi fleet. We've had mostly GM in our
government fleet for over 20 years and we just don't have much in the
way of problems with them yet to listen to the critics we'd be lucky
if half of them were running at any given point in time. The last new
GM I had (95 caprice) required perhaps $1000 in actual non-wear item
repairs (both in and out of warranty) in the 12 years and 140K miles
of service. The entirety of it's repairs was to fix two electric
windows and one sensor on the transmission. Yet some fool on the CU
survey could, and probably would, put down 3 major repairs while
swearing they'd never buy another domestic.

>There seems to be a perception that the CU survey ask questions like
>'which is more reliable, a Toyota or a Chevy.'
>

Not at all, I've seen their questionnaire.


> I see people all the time who buy an "import" for no
>> reason other then they don't want a domestic. And that includes
>> people who have run their previous domestics for way past 100K with no
>> more problems then the imports have. In other cases it's in spite of
>> having owned imports that fell apart around them. When someone has a
>> problem with their Toyota they just shrug it off, when someone has a
>> problem with their domestic they can't stop badmouthing it long enough
>> to breath.
>
>Actually it's the opposite. The Toyota owner with a problem is upset
>because they have the mindset of 'hey this is a Toyota, I'm not supposed
>to have these problems,' while the Chevy owner shrugs it off with "well
>it's a GM product, I expected this.'
>

Sorry but human nature is such that the typical Toyota owner, who
bought it because it was supposed to be so great, is not going to
acknowledge that he paid a higher price and has just as many problems
as he's had with every other car. So he's going to forget about half
the repairs and not even list them. And to cement the rightness of
his decision he'll say he'd buy one again. OTOH, the Chevy owner
shrugs nothing off. He feels like he did GM a favor buying the car in
the first place and by god if there is a defect he's going to make
them pay dearly by giving it lots of poor ratings - that'll teach em.

>In any case the number of owners that would lie when asked questions
>regarding specific sub-systems of their vehicle, is small. No one has
>ever shown any bias in the Consumer Reports surveys, either by the
>respondents, or in the questions that are asked. It's all sour grapes
>when someone complains. The complaint is usually along the line of 'well
>_I_ never had xyz problem, so that means the survey is bogus.' These
>people don't understand statistics or the necessity of a large sample
>size in order to have valid data.


You don't understand that people rarely tell the truth for a wide
variety of reasons. And the CU survey depends 100% on the respondents
telling the truth about what is, to a large number of people, a very
emotionally charged large $$ purchase.

"SMS" > wrote in message
> They aren't asking for a subscriber's opinion of one product versus
> another, they're asking a series of very specific questions that ask what,
> if any, problems an owner has had with their vehicle.
>
> There seems to be a perception that the CU survey ask questions like
> 'which is more reliable, a Toyota or a Chevy.'


There are two parts to the auto issue. One part is the survey, the other
part is the editorial comment by their testers.

IMO, the bias comes in the editorial part. Like the Suzuki rollover issue
that was proven to be problematic with CR.

The survey portion has some merit, but we don't know a lot of how the
questions were answered. Do owners of some makes tend to forget about minor
problems more or less than owners of other makes? One respondent may think
nothing of the time the radio did not work for a week and had to be replaced
and forget to mark it while another may be PO'd that one day driving under
power lines he had static on his favorite station 150 miles away and he
tells everyone about the crappy radio. .

wrote:

> I'll bite. I'm looking at the April, 2009 auto issue. Let's take the
> Ford Fusion on page 89. I see lots of red circles. I'm not seeing
> them state any numbers aside from the model year and the "6" in V6.
> No sample size for this model is given, though you claim it is always
> stated. No margin of error for the reliability projection. Nothing
> of statistical value whatsoever. You must still be getting the
> special edition? Please share.

They state the sample size for the whole survey, and they also state
that they leave out models for which they get too low a number or
responses for the data to be statistically valid. Fortunately the survey
is so large that only really niche vehicles are left out for lack of data.

> Well, I engaged in the amusement provided by helping to debunk your
> anti-Saturn rhetoric and finding all your contradictions, anyway.

LOL, sure you did. It's interesting that all the things I reported on so
early turned out to be completely true, and resulted in recalls or
special service campaigns by Saturn. You debunked nothing. But if it
makes you feel good to think that you did, go right ahead believing it.

Saturn's demise was a direct result of their basing an entire company on
marketing hype that could not fool enough people to sustain the
business, and could not overcome the fact that the vehicles were
unreliable. When CR and J.D. Power pointed out these facts, you saw the
same kind of sour grapes you see now with the claims of bias.

On Fri, 27 Nov 2009 02:26:09 -0800, SMS >
wrote:
>
>> Well, I engaged in the amusement provided by helping to debunk your
>> anti-Saturn rhetoric and finding all your contradictions, anyway.
>
>LOL, sure you did. It's interesting that all the things I reported on so
>early turned out to be completely true, and resulted in recalls or
>special service campaigns by Saturn. You debunked nothing. But if it
>makes you feel good to think that you did, go right ahead believing it.
>

You were pretty much waging a trolling campaign against Saturn.
Way over the top. Fake trolling names and all.
Pretty much like the domestic guys will go after Toyota now for the
gas pedal problem that's killed some people, and how they went after
the Toyota head gasket and sludging problems.
No real difference. Just one biased person complaining about others'
bias.
Happens all the time. Mob mentality.
Don't feel bad about. But fight it. No sense being a punching bag.
And try to look at the big picture.

--Vic

On Thu, 26 Nov 2009 16:26:58 -0800 (PST), wrote:

>On Nov 26, 9:52Â*am, Vic Smith > wrote:
>
>Hi Vic!
>
>> Not getting into statistical analysis methods, I'll note the CR has
>> NEVER given sampling sizes on a model basis. Â*Only total surveys.
>> I am perfectly willing to be corrected on that.
>
>You won't be.
>
Hey Caviller! Long time no hear. Those were fun days when Steve was
playing the fool. Persistent cuss, wasn't he?
He did have some good points about Saturn timing chain
lubrication/failures, just somewhat over the top.
He was touting the Toyota he owned of course.
Pretty proud of how he flew from S.F to LA to cut a deal on it and
take advantage of dealer price differentials. He did a good job on
that, and is to be commended. He really "loved" that Toyota.
And here he is - a CR subscriber I guess - claiming people aren't
biased.
hehe.

--Vic

On Nov 27, 4:26*am, SMS > wrote:
> wrote:
> > I'll bite. *I'm looking at the April, 2009 auto issue. *Let's take the
> > Ford Fusion on page 89. *I see lots of red circles. *I'm not seeing
> > them state any numbers aside from the model year and the "6" in V6.
> > No sample size for this model is given, though you claim it is always
> > stated. *No margin of error for the reliability projection. *Nothing
> > of statistical value whatsoever. *You must still be getting the
> > special edition? *Please share.
>
> They state the sample size for the whole survey, and they also state
> that they leave out models for which they get too low a number or
> responses for the data to be statistically valid. Fortunately the survey
> is so large that only really niche vehicles are left out for lack of data..


Lol. "Hi! We surveyed a million of our own readers, so OF COURSE that
means our results are accurate!" If that's enough to satisfy your
intellecutual curiousity, so be it.


> > Well, I engaged in the amusement provided by helping to debunk your
> > anti-Saturn rhetoric and finding all your contradictions, anyway.
>
> LOL, sure you did. It's interesting that all the things I reported on so
> early turned out to be completely true, and resulted in recalls or
> special service campaigns by Saturn. You debunked nothing. But if it
> makes you feel good to think that you did, go right ahead believing it.


Thanks to Google, thousands of your anti-Saturn and anti-domestic
rants still exist in the archives. A monkey can hit the bullseye with
a dart once in a while, but it's all the horrible misses that provide
a good chuckle.


> Saturn's demise was a direct result of their basing an entire company on
> marketing hype that could not fool enough people to sustain the
> business, and could not overcome the fact that the vehicles were
> unreliable. When CR and J.D. Power pointed out these facts, you saw the
> same kind of sour grapes you see now with the claims of bias.


Haven't owned a Saturn in nearly a decade, so I'm really not
interested except to note your bias on the topic of this thread.
Since CR showed the S-series to be generally reliable, contrary to
your claims above, that doesn't even provide a basis for sour grapes.
It only provides irony that you put so much trust into their
reliability data, but ignore any results that you don't like.

As for bias, it would be nice to live in a world where you can believe
everything you see from the media at face value and know that every
study and statistic is accurate. Sadly, in the world where I live,
being naive usually means being a sucker and being duped. Here, we
would just have to blindly assume CR has no bias, because they hide
away all their methodology and statistical information for some
mysterious reason. I admit, I do like your world better.

On Nov 26, 9:02*pm, "Ed Pawlowski" > wrote:
> "SMS" > wrote in message

>
> There are two parts to the auto issue. One part is the survey, the other
> part is the editorial comment by their testers.
>
> IMO, the bias comes in the editorial part. *Like the Suzuki rollover issue
> that was proven to be problematic with CR.

Plus, the unintentional bias comes from the part that because they
survey only their own readers, 100% of the resonses are influenced by
the editorial content.

Put aside a survey of questionable scientific validity that wouldn't
pass an undergraduate marketing exam. Put aside the results that
wouldn't hold up in any peer reviewed statistics journal. Put aside
any personal biases on the part of the management or editors. This
one issue makes everything else suspect. The results could be
reasonably accurate. Or maybe not. You'll never know. Even if they
are, I don't find the difference of 2 reported issues (of unknown
severity and cost) over 7 years to be all that significant between a
typical Ford and a typical Honda. The reality is that their own
results are exaggerating small differences by showing them as
percentages above/below an average number instead of showing them as
absolutes.

On Fri, 27 Nov 2009 02:26:09 -0800, SMS fired up the etcha-a-sketch and
scratched out:

>> Well, I engaged in the amusement provided by helping to debunk your
>> anti-Saturn rhetoric and finding all your contradictions, anyway.
>
> LOL, sure you did. It's interesting that all the things I reported on so
> early turned out to be completely true, and resulted in recalls or
> special service campaigns by Saturn. You debunked nothing. But if it
> makes you feel good to think that you did, go right ahead believing it.
>
> Saturn's demise was a direct result of their basing an entire company on
> marketing hype that could not fool enough people to sustain the
> business, and could not overcome the fact that the vehicles were
> unreliable. When CR and J.D. Power pointed out these facts, you saw the
> same kind of sour grapes you see now with the claims of bias.

Saturn vehicles are unreliable?

They're small, but I never knew them to be unreliable.



--
perfectreign
www.perfectreign.com || www.ecmplace.com
a turn signal is a statement, not a request

wrote:

> Plus, the unintentional bias comes from the part that because they
> survey only their own readers,

So you believe that a CR reader that bought a Ford or Chevy is more
likely to admit to problems than a CR reader that bought a Toyota or
Honda? Where's your evidence of that, LOL?

The survey isn't "what's your opinion of the reliability of xyz brand,"
it's detailed questions on problems you've had with the vehicle you own.