Home | Contact Us | FAQ | Search & Site Map | Link to Us
Sign In | Join | Other 45 Sites in Network
HomeAnnouncements
Discussion Groups
By Brand
BMWChevroletDodgeFordGMHondaLexusMercedes-BenzNissanPeugeotToyotaVolkswagenOther Brands
By Topic
4x4 CarsRVsDrivingMaintenance & RepairCar AudioCollectible Cars
Country Specific
Australian ForumsUK Forums
ArticlesAuto InsuranceBuyingCars & TechnologyMaintenanceMiscellaneousSafety
DMV Resources
Related Topics
MotorcyclesBoatsMore Topics ...

Car Forum / Driving, Maintenance, Tuning / General Car Topics / May 2004

Tip: Looking for answers? Try searching our database.

Re: Consumer Reports about cars

Thread view: 
Enable EMail Alerts  Start New Thread
Thread rating: 
Steven M. Scharf - 28 Apr 2004 17:30 GMT
"Caviller" <caviller@my-deja.com> wrote in message

>It's also amusing to note that Mr. Scharf himself has had his share of
>criticisms for CR when he doesn't agree with their editorial content. It's
>always nice to be able to bash a publication when you disagree
>with them.  It's amusing when someone turns around and praises the
>same publication as being unbiased and extremely valuable when they
>agree with it.

<snip>

You are confusing the product evaluations done by Consumer Reports with
their surveys of subscribers. The latter are extremely valuable because of
the very low margin of error. They will leave out a model when they fail to
get insufficient data for a low margin of error, but with vehicles this
isn't usually an issue (with cellular they looked at 12 regions, and if,
within a region, there were insufficient responses then they omitted the
data).

The demographics isn't really an issue because they aren't asking a
bunch of college professors what their favorite vehicle is, they're asking
each subscriber to evaluate their own vehicle. Perhaps someone might claim
that more well-off and better educated individuals are ore (or less) likely
to complain about specific defects, but again, since it's all relative, it
isn't an issue.

Their product evaluations definitely have a dose of editorial content. For
vehicles, they tend to value safety and reliability over the biggest engine
(though they don't hesitate to call a vehicle underpowered). They are
typically unimpressed with gimmicks and gadgets. This is a reflection of
their subscriber base, who typically aren't buying Corvettes or Maseratis,
and are more interested in Camry versus Accord versus Passat, or Acura
versus Lexus versus Infiniti.

For some products they lack an understanding of the cost to benefit ratio of
some product features, which often has them recommending products that are a
bit too downmarket.
Caviller - 29 Apr 2004 02:18 GMT
> "Caviller" <caviller@my-deja.com> wrote in message

> You are confusing the product evaluations done by Consumer Reports with
> their surveys of subscribers. The latter are extremely valuable because of
[quoted text clipped - 3 lines]
> within a region, there were insufficient responses then they omitted the
> data).

Please reread my post.  I clearly was not commenting about their
evaluations.  In regard to the subscriber surveys, I asked a few
simple questions about your basis for these claims.  Again, what is
their margin of error, exactly?  What is the threshold for
insufficient data, exactly?  How about their standard deviation?
After they normalize their results into ratings circles, is a half-red
circle significantly better than a half-black one?  Do you know any
detail at all about their methodology or statistics?  If not, how can
you possibly claim they have a very low margin of error?

Just out of curiousity, if the surveys do indeed have a very low
margin of error as you claim, how do you explain the relatively good
showing for the small Saturn sedans?  These are the ones you've bashed
for widespread reliability issues on usenet for many years...

> The demographics isn't really an issue because they aren't asking a
> bunch of college professors what their favorite vehicle is, they're asking
> each subscriber to evaluate their own vehicle. Perhaps someone might claim
> that more well-off and better educated individuals are ore (or less) likely
> to complain about specific defects, but again, since it's all relative, it
> isn't an issue.

Anytime you don't randomly sample, it can lead to errors.  Whether the
errors are significant or not in this case is unclear.  For example,
readers can be biased by the results from previous surveys, or any
editorial bias.  That can affect how they respond on their surveys.
As one textbook example goes, the Chicago Tribune once made a mistake
of this type.  Supposedly, a strike had caused a manpower shortage.
As one version of the legend goes, the remaining staff did a cheap and
easy last minute survey of their readers for a presidential poll.  The
demographic of their readers happened not to be very random
politically, as their [in]famous "Dewey Defeats Truman" headline would
indicate.  This is why professional surveys are almost always random,
and not self-selected from a subscriber base. (
http://www.drjohnmiller.com/perform.html ).
misterfact - 02 May 2004 15:55 GMT
> > "Caviller" <caviller@my-deja.com> wrote in message
>  
[quoted text clipped - 41 lines]
> and not self-selected from a subscriber base. (
> http://www.drjohnmiller.com/perform.html ).

 Attention Mr. Scharf- you're strangely SILENT !
misterfact - 06 May 2004 20:57 GMT
> > "Caviller" <caviller@my-deja.com> wrote in message
>  
[quoted text clipped - 7 lines]
>
>Just out of curiousity, if the surveys do indeed have a very low
margin of error as you claim, how do you explain the relatively good
showing for the small Saturn sedans?  These are the ones you've bashed
for widespread reliability issues on usenet for many years...

 With all the complaints and un-resolved problems with Saturns-
sounds to me that Consumer Reports' product evaluations leave a lot to
be desired-both in their sampling process and specifics.
Steven M. Scharf - 06 May 2004 10:22 GMT
<MikeHunt2@mailcity.com> wrote in message
news:40990590.B101F20A@mailcity.com...
> I though I explained that in my post.  For a vehicle to be
> considered made in the USA it must be assembled of at least
> 75% American made parts.

This is true. But it's cute how you used "American." The parts don't have to
be from the U.S.; Canada or Mexico counts as "American." And labor is not
included in the calculations. A vehicle can actually have 0% U.S. content
and be considered "Made in the U.S.A." The U.S. makers are adept at working
the domestic content law to their benefit. Building new factories in Mexico
is one of the biggest problems.
 
Sign In
Join
My Latest Posts
My Monitored Threads
My Blog
My Photo Gallery
My Profile
My Homepage

Start New Thread
Enable EMail Alerts
Rate this Thread



©2008 Advenet LLC   Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.