Notices
996 Forum 1999-2005
Sponsored by:

996 Reliability Survey - Admin Approved!

Thread Tools
 
Search this Thread
 
Old 05-06-2010, 01:03 PM
  #46  
LVDell
Nordschleife Master
 
LVDell's Avatar
 
Join Date: Oct 2003
Location: Tobacco Road, NC
Posts: 5,225
Likes: 0
Received 27 Likes on 23 Posts
Default

Uh oh!

As Ricky said to Lucy.........
You got some splainin' to do!
Attached Images  
Old 05-06-2010, 01:17 PM
  #47  
Bob Rouleau

Still plays with cars.
Lifetime Rennlist
Member
Thread Starter
 
Bob Rouleau's Avatar
 
Join Date: Jul 2001
Location: Montreal
Posts: 15,078
Received 255 Likes on 119 Posts
Default

Originally Posted by BruceP
Wait, I thought that was the point. I thought this was a heroic effort to set the record straight about the reliability of 996s.
Bruce, in fact setting the record straight is MY hope. Truedelta will simply report. For years I have watched Dell (and others) dish out pretty harsh treatment to anyone who suggests that the M96 engine is inherently flawed. I hope that with enough data we might be able to provide evidence that Dell, Wellardmac et al were right all along. My own experience with an M96 was similar to theirs - no issues at all.

That's why I am puzzled by the objections to the survey. I am not qualified to debate stats with pros like Dell or Michael Karesh. On the the other hand anything reasonable which is positive will be a lot better than the conventional wisdom of the "ticking bomb" don't you think?

Best,
Old 05-06-2010, 01:24 PM
  #48  
BruceP
Drifting
 
BruceP's Avatar
 
Join Date: Aug 2006
Posts: 2,508
Likes: 0
Received 24 Likes on 18 Posts
Default

Originally Posted by Bob Rouleau
Bruce, in fact setting the record straight is MY hope. Truedelta will simply report. For years I have watched Dell (and others) dish out pretty harsh treatment to anyone who suggests that the M96 engine is inherently flawed. I hope that with enough data we might be able to provide evidence that Dell, Wellardmac et al were right all along. My own experience with an M96 was similar to theirs - no issues at all.

That's why I am puzzled by the objections to the survey. I am not qualified to debate stats with pros like Dell or Michael Karesh. On the the other hand anything reasonable which is positive will be a lot better than the conventional wisdom of the "ticking bomb" don't you think?
I do. I just think that people are worried about the potential impact of bad information, having endured all this controversy. It really takes the fun out of owning the car sometimes.
Old 05-06-2010, 01:33 PM
  #49  
BruceP
Drifting
 
BruceP's Avatar
 
Join Date: Aug 2006
Posts: 2,508
Likes: 0
Received 24 Likes on 18 Posts
Default

Sidebar:

Bob, you know what I would really love to have? A survey like this among all Porsche owners rather than among the universe of car owners. It's only among Porsche owners that the 996 is the Rodney Dangerfield of cars, and it's only among Porsche owners that it has any kind of specific reputation at all. It would be fascinating to look at a database of maintenance and repair diaries of everybody on Rennlist, and an awesome resource. That I would sign up for.
Old 05-06-2010, 01:35 PM
  #50  
mkaresh
Racer
 
mkaresh's Avatar
 
Join Date: Feb 2010
Posts: 310
Likes: 0
Received 0 Likes on 0 Posts
Default

Originally Posted by BruceP
Wait, I thought that was the point. I thought this was a heroic effort to set the record straight about the reliability of 996s.
I never said I was trying to affect resale values, nor have I ever made any claims that my stats could affect resale values.

Earlier you got your knickers in a twist because you felt I had distorted what you said. Now you're entirely putting words in my mouth. Hypocrite much?
Old 05-06-2010, 01:37 PM
  #51  
Sneaky Pete
Rennlist Member
 
Sneaky Pete's Avatar
 
Join Date: Nov 2008
Location: Mooresville, IN (Life Long Cheesehead)
Posts: 5,815
Likes: 0
Received 54 Likes on 35 Posts
Default

This thread is making me ill. Why isn't there a smiley face for puking?
Old 05-06-2010, 01:40 PM
  #52  
BruceP
Drifting
 
BruceP's Avatar
 
Join Date: Aug 2006
Posts: 2,508
Likes: 0
Received 24 Likes on 18 Posts
Default

Originally Posted by mkaresh
I never said I was trying to affect resale values, nor have I ever made any claims that my stats could affect resale values.

Earlier you got your knickers in a twist because you felt I had distorted what you said. Now you're entirely putting words in my mouth. Hypocrite much?
Ah, the personal attack. That always works with me. Gosh, you're right. I surrender. You never had any altruistic motives at all. It's just a commercially motivated survey.

Grownup much?
Old 05-06-2010, 01:51 PM
  #53  
mkaresh
Racer
 
mkaresh's Avatar
 
Join Date: Feb 2010
Posts: 310
Likes: 0
Received 0 Likes on 0 Posts
Default

BruceP,

Pointing out hypocrisy is hardly a personal or juvenile attack. For an example of such an attack see the earlier thread, where one of you ended up deleting such an attack.

I haven't the faintest idea what works with you. Quite likely nothing.

You do seem to have trouble thinking beyond a false dichotomy regarding my alleged motives. Following your logic, anyone who is paid for what they do is only in it for the money. Clearly false.
Old 05-06-2010, 02:07 PM
  #54  
BruceP
Drifting
 
BruceP's Avatar
 
Join Date: Aug 2006
Posts: 2,508
Likes: 0
Received 24 Likes on 18 Posts
Default

Where I come from, "hypocrite" is an insult.

And you know what "works" with me? Integrity. That works.

You are a commercial enterprise who is taking advantage of the hospitality of an online community to generate content for your site. The quid pro quo for this is that this same community might benefit from the resulting information. There is nothing wrong with this. It happens all the time. But if you're going to do this, in the real live grownup marketing world, you do it respectfully and honestly and with humility and gratitude. You listen. You be enthusiastic. You make people want to help you. You accept their advice, since you're also accepting their time and personal information.

What you do not do is march into the community, a perfect stranger, pitch up your site as a service to mankind and then become argumentative, defensive and dismissive with people who question you. It's the equivalent of ringing a doorbell and interrupting a party to sell aluminum siding, and then yelling at people for not having the good taste to buy it.

If you want to make a go of this, you need people on your side. You will absolutely fail if the character you're showing in here ends up attached to your brand.

People usually pay me for that advice, but I'm giving it to you for nothing in the spirit of community. Isn't that nice?
Old 05-06-2010, 02:17 PM
  #55  
Palmbeacher
Banned
 
Palmbeacher's Avatar
 
Join Date: Oct 2009
Posts: 731
Likes: 0
Received 0 Likes on 0 Posts
Default

Originally Posted by LVDell
Actually you have confused reliability (and I have no doubt his measure is reliable) with validity. In this case, it's validity, more specifically construct validity and criterion (predictive) validty, that is of concern.
Forgive me for not being bilingually-fluent in statistician-speak vocabulary. I'm not a statistician for whom everything is an abstract science of figures and mathematics, I'm the guy who has to decide whether conclusions drawn from those statistics are the result of flawed methodology and hence, erroneous, so innocent people don't get sick or hurt. The proposed methodology in this case is seriously flawed, and I would not put faith in any conclusions drawn from it. If you can't tell whether data is corrupt, due diligence demands one assume it is.

Originally Posted by mkaresh
Analysis of survey data ALWAYS extrapolates from the sample to the population. It's the entire point. I'm not sure how you think the actual number of cars in the population might play into this--it's irrelevant.
If the sampling was random, which this is not. You're surveying people who read internet forums. It would be like doing a study on cardiovascular health in America sampling only people interviewed eating at McDonalds... and then extrapolating your stats to the general population.


I run a large number of checks for errors in the data
You don't even make an attempt to verify that respondants own or did own a Porsche, let alone verify that they actually had the repair issues they indicate. What other errors in the data would you deem important to check for, if not those ?
Old 05-06-2010, 02:32 PM
  #56  
mkaresh
Racer
 
mkaresh's Avatar
 
Join Date: Feb 2010
Posts: 310
Likes: 0
Received 0 Likes on 0 Posts
Default

To the extent hypocrite is an insult, it's one with a clear, specific basis. Unlike, to pick a random example, "moronic."

The general gist of suggestions has been that the survey is fatally flawed so I should abandon it and the thousands of people who've been participating for the past four years. My apologies for not accepting this "advice" with enthusiasm, and for becoming defensive when attacked.

When people do offer constructive advice, I do react as you suggest. But there has been very little such advice here.

I have "yelled" at no one simply for not signing up, here or anywhere else. So this is also a blatant mischaracterization.

Before I ever posted in this forum I asked for, and received, permission from the owner. This particular thread was posted by the admin at his own initiative. So I honestly do not feel like a salesman interrupting a dinner party.

Throughout I have responded honestly, openly, and with an emphasis on the facts, so I see no problem with my character or integrity as represented here. True, I do not speak like a PR professional. I am not a PR professional.

Oddly, when people get the PR spin, they claim to want someone who speaks honestly and plainly. But when someone speaks honestly and plainly, they ask for sugar-coated spin.
Old 05-06-2010, 02:45 PM
  #57  
mkaresh
Racer
 
mkaresh's Avatar
 
Join Date: Feb 2010
Posts: 310
Likes: 0
Received 0 Likes on 0 Posts
Default

Originally Posted by Palmbeacher
The proposed methodology in this case is seriously flawed, and I would not put faith in any conclusions drawn from it. If you can't tell whether data is corrupt, due diligence demands one assume it is.

If the sampling was random, which this is not. You're surveying people who read internet forums. It would be like doing a study on cardiovascular health in America sampling only people interviewed eating at McDonalds... and then extrapolating your stats to the general population.

You don't even make an attempt to verify that respondants own or did own a Porsche, let alone verify that they actually had the repair issues they indicate. What other errors in the data would you deem important to check for, if not those ?
There is not a reliability survey I'm aware of that does anything to verify that repairs were actually performed. We all assume that if someone reports a repair then the repair actually occurred, because there's really very little reason to assume otherwise and verifying every repair would be expensive. Even if cost were not a factor, and it certainly is, the negative impact of the reduced sample size would very likely outweigh the positive impact of verification.

There is similarly very little reason to assume that someone is taking the time over a period of many months to report on a car that does not actually exist. It would be a fair amount of effort with no real point. One or two such responses would have no substantial impact on the results. And more than one such response would be difficult to provide without being picked up by the checks I can and do run.

It's hardly the same as conducting a survey at McDonalds, as that population is engaging in a behavior with a clear connection to the dependent variable.

My focus is on the overall repair frequency. The great majority of car repairs are not dependent on how the car was driven or cared for. It is not possible to maintain the electrical system, to give a key example. And, even with those repairs that could have been influenced by how the car was driven and cared for, most cars are cared for well enough that they should not have required these repairs.

Major mechanical repairs before 120,000 miles are rare these days, in nearly all car models. In those cases where they are relatively common, the reason is not primary because of how the car was driven or maintained, but because of how the car was engineered.
Old 05-06-2010, 03:05 PM
  #58  
chsu74
Rennlist Member
 
chsu74's Avatar
 
Join Date: Oct 2006
Location: CT
Posts: 9,615
Received 311 Likes on 259 Posts
Default

If anything, this is entertainment. I am getting no work done and will get more popcorn. Someone pass the salt...
Old 05-06-2010, 03:08 PM
  #59  
mkaresh
Racer
 
mkaresh's Avatar
 
Join Date: Feb 2010
Posts: 310
Likes: 0
Received 0 Likes on 0 Posts
Default

Originally Posted by chsu74
If anything, this is entertainment. I am getting no work done and will get more popcorn. Someone pass the salt...
Your taste in entertainment clearly differs from mine
Old 05-06-2010, 03:20 PM
  #60  
BruceP
Drifting
 
BruceP's Avatar
 
Join Date: Aug 2006
Posts: 2,508
Likes: 0
Received 24 Likes on 18 Posts
Default

Sorry, no, Mike. It's a pejorative, and thus a personal attack. You may observe that you think someone has presented contradictory reasoning or evidence. But once you impute motive to this, it's just an insult.

I can't believe you think any of this is going to help your cause.

Last edited by BruceP; 11-29-2012 at 08:27 PM.


Quick Reply: 996 Reliability Survey - Admin Approved!



All times are GMT -3. The time now is 07:45 PM.