Date:         Fri, 3 Apr 1998 16:09:40 -0800
Reply-To:     "Steven X. Schwenk" <sxs@CONCENTRIC.NET>
Sender:       Vanagon Mailing List <Vanagon@vanagon.com>
From:         "Steven X. Schwenk" <sxs@CONCENTRIC.NET>
Subject:      Re: Head (gasket) fix
Comments: To: Bob Berkemer <berky@WIN.BRIGHT.NET>,
          vanagon List <vanagon@vanagon.com>
Content-Type: text/plain; charset=us-ascii

Bob:

How do you do the variables for the ground theory, though?  It is only a
rare few who have checked or augmented their ground wires.

Besides the "coolant/water" and the "inadequate ground" camps, there are
also the "warm up before driving" camp., the "daily driver v.
occassional driver" camp, the "easy driver v. higher rpm" camp and the
"everything is random, nothing matters" camp.

I have gotten a few suggestions for the Statistical Survey Form, or
whatever we'll call it.  I need more feed back, though.  Please send
variables you think are improtant, or tell me which of those I proposed
are no good or need modification.  I thought of adding some for
repairs:  privatre rebuilt heads/ vw rebuilt heads / jb weld heads /
ceramic coated heads, etc., so we can get some measure on whether one
method of repair lasts longer than others.

Ian...I know nothing as far as whether the variables I have proposed are
appropriate or sufficient for meaningful statistical analysis.

steve



Bob Berkemer wrote:

> God I hated stats!      Lies, lies and damnable lies!
>
> Oh well..I regress....... It seems that there are two basic hypothesis
>
> for the failure of the head gaskets even though they seem to be
> linked.
> First is the bad water/antifreeze camp. Second is the bad grounding of
>
> anything/everything camp. Shouldn't our instrument allow for these two
>
> major categories for data input? Then would we need to run an analysis
>
> of covariance? Sorry but I'm not an empirical guy. More of a
> touchy/feely like "how do you feel about your heads failing"
> Remember,
> it's friday!
>
> O.K. Ian.... your on!
>
> Bob
> 84 westy, "der bus"